| *** epim has joined #openstack-dev | 00:00 | |
| clarkb | dolphm: not yet | 00:00 |
|---|---|---|
| clarkb | dolphm: just started wrapping my head around the problem | 00:00 |
| *** nachi1 has left #openstack-dev | 00:00 | |
| clarkb | dolphm: it doesn't seem like this is a keystone bug though. It is a linux bug :) | 00:00 |
| *** sushils has quit IRC | 00:01 | |
| dolphm | clarkb: i was about to ask "what's actually misbehaving here?" | 00:01 |
| *** ftcjeff has quit IRC | 00:01 | |
| clarkb | dolphm: the linux default local port range overlaps with the range of ports that IANA does assignemnts over | 00:01 |
| *** nati_ueno has quit IRC | 00:01 | |
| *** epim has quit IRC | 00:01 | |
| dolphm | clarkb: is there anything we can do to prevent that in the gate? | 00:02 |
| *** yamahata_ has joined #openstack-dev | 00:02 | |
| dolphm | clarkb: (start keystone before whatever happened to claim that port?) | 00:02 |
| clarkb | dolphm: we can shift the range on the hosts we use in the gate | 00:02 |
| bknudson | you can set the port rang - /proc/sys/net/ipv4/ip_local_port_range | 00:02 |
| clarkb | dolphm: we can't garuntee that nothing else doesn't use that port without shifting the range as it is semi random | 00:02 |
| morganfainberg | clarkb, i assume this doesn't happen everytime... just enough to register as a "wtf is that" | 00:03 |
| *** CaptTofu has quit IRC | 00:03 | |
| dolphm | clarkb: bknudson: are ya'll contradicting each other? | 00:03 |
| dolphm | morganfainberg: ++ | 00:03 |
| clarkb | morganfainberg: correct | 00:03 |
| morganfainberg | dolphm, no they are in agreement...afaict | 00:03 |
| clarkb | we are in agreement | 00:03 |
| *** CaptTofu has joined #openstack-dev | 00:03 | |
| dolphm | bknudson: is that possible for devstack? | 00:04 |
| bknudson | # echo "49152 65535" > /proc/sys/net/ipv4/ip_local_port_range | 00:04 |
| bknudson | devstack does sudo all over the place. | 00:04 |
| dolphm | ^^ | 00:04 |
| bknudson | installing packages... | 00:04 |
| clarkb | bknudson: or 35358 61000 | 00:04 |
| morganfainberg | drop things into sysctl.d (or distro equivalent) | 00:04 |
| *** nermina has quit IRC | 00:04 | |
| dolphm | clarkb: shouldn't you start above the IANA range? | 00:04 |
| clarkb | dolphm: maybe? I don't have an opinion on linux vs IANA here | 00:05 |
| morganfainberg | dolphm, ++ but i think we use the highest port... we might be the only project that uses a port in linux's ephemeral range | 00:05 |
| bknudson | maybe "49152 61000" | 00:05 |
| morganfainberg | clarkb, but i'd argue the safest bet is what bknudson just said. | 00:05 |
| morganfainberg | as long as it doesn't overly limit available ports (should be fine.... i think) | 00:05 |
| bknudson | since wikipedia says "dynamic or private ports" so maybe 61000+ are private ports. | 00:06 |
| clarkb | there are several problems here 1. keystone's default port conflicts with default local port ranges on linux boxes, we should probably document that for deployers 2. linux and IANA don't agree 3. wtf linux :) | 00:06 |
| bknudson | torvalds: ping | 00:06 |
| *** teran has quit IRC | 00:06 | |
| *** jhesketh_ has quit IRC | 00:06 | |
| dolphm | morganfainberg: i believe we are the only project with any reason to appeal to IANA | 00:06 |
| *** anniec has quit IRC | 00:06 | |
| morganfainberg | dolphm, that is fairenough. | 00:06 |
| dolphm | clarkb: ++ | 00:07 |
| clarkb | I will propose a change to devstack that shifts the range to 49152 61000. I will also submit a bug against keystone and devstack to track the work (keystone really should document this I htink) | 00:07 |
| bknudson | the other projects should use an ephemeral port and tell keystone what they picked so we can put in in the service catalog | 00:07 |
| *** stevemar has quit IRC | 00:07 | |
| *** stan_ivanov has quit IRC | 00:07 | |
| dolphm | clarkb: will you send me a link to that review? | 00:07 |
| morganfainberg | clarkb, feel free to include me if you want on that review. | 00:08 |
| *** derekh has quit IRC | 00:09 | |
| clarkb | dolphm: sure, morganfainberg will do | 00:09 |
| *** anniec has joined #openstack-dev | 00:11 | |
| *** dolphm is now known as dolphm_afk | 00:12 | |
| *** amotoki is now known as __amotoki__ | 00:14 | |
| *** CaptTofu has quit IRC | 00:14 | |
| *** sthaha has joined #openstack-dev | 00:14 | |
| *** CaptTofu has joined #openstack-dev | 00:14 | |
| *** xqueralt has quit IRC | 00:15 | |
| clarkb | dolphm_afk: bknudson: morganfainberg: https://bugs.launchpad.net/keystone/+bug/1253482 is the bug | 00:15 |
| uvirtbot | Launchpad bug 1253482 in keystone "Keystone default port in linux local ephemeral port range. Devstack should shift range." [Undecided,New] | 00:15 |
| *** pete5 has quit IRC | 00:15 | |
| morganfainberg | clarkb, thanks! | 00:15 |
| *** rongze has joined #openstack-dev | 00:16 | |
| *** carl_baldwin has quit IRC | 00:16 | |
| *** MaxV has joined #openstack-dev | 00:16 | |
| *** fbo is now known as fbo_away | 00:16 | |
| *** epim has joined #openstack-dev | 00:16 | |
| *** Ryan_Lane has quit IRC | 00:17 | |
| clarkb | dtroyer: if you are around, do you have an preference for where that goes in devstack? lib/keystone maybe? | 00:18 |
| *** armax has joined #openstack-dev | 00:20 | |
| *** rongze has quit IRC | 00:20 | |
| *** armax has left #openstack-dev | 00:21 | |
| *** yamahata_ has quit IRC | 00:22 | |
| *** yamahata_ has joined #openstack-dev | 00:22 | |
| *** pixelb has quit IRC | 00:22 | |
| *** dosaboy has quit IRC | 00:23 | |
| clarkb | hmm it almost needs to do it upfront to have a shot at keeping other processes away from the keystone ports | 00:24 |
| *** shinylasers has quit IRC | 00:24 | |
| clarkb | doing it at start keystone is too late | 00:24 |
| bknudson | clarkb: can the source image be updated? | 00:24 |
| bknudson | then it wouldn't be in devstack | 00:24 |
| *** faramir1 has joined #openstack-dev | 00:24 | |
| clarkb | bknudson: it can, but this is in theory a problem for everyone else | 00:24 |
| clarkb | we try to keep the delta between our images and devstack as small as possible | 00:25 |
| clarkb | maybe I need to account for it in both places | 00:25 |
| bknudson | clarkb: that might be the best way... then users of devstack would only hit this at most once. | 00:25 |
| bknudson | once per vm. | 00:25 |
| clarkb | ya | 00:25 |
| *** shinylasers has joined #openstack-dev | 00:25 | |
| *** shinylasers has quit IRC | 00:26 | |
| *** jasdeepH has quit IRC | 00:26 | |
| bknudson | or could put in a wait for the ephemeral port to be released | 00:26 |
| *** tdruiva has quit IRC | 00:26 | |
| *** salv-orlando has quit IRC | 00:26 | |
| *** lbragstad has quit IRC | 00:26 | |
| *** tdruiva has joined #openstack-dev | 00:27 | |
| dolphm_afk | clarkb: marked as won't fix on the keystone side, based on the above | 00:29 |
| *** roampune has quit IRC | 00:29 | |
| clarkb | dolphm_afk: I think keystone needs to document it | 00:29 |
| morganfainberg | dolphm_afk, clarkb, wasn't the bug simply to document in keystone? | 00:29 |
| dolphm_afk | clarkb: ah, how so? | 00:29 |
| dolphm_afk | totally missed the last paragraph.. | 00:29 |
| notmyname | ttx: why are you (or your robot self) unprioritizing and untargeting my swift blueprints? | 00:29 |
| clarkb | dolphm_afk: this is something that may not be apparent to deployers, so they should be made aware of it | 00:30 |
| clarkb | dolphm_afk: basically if you are running keystone you probably want a different local port range on linux | 00:30 |
| morganfainberg | maybe a big fat "IANA says we use port # but linux defaults to # so.. fix might get a conflict in some cases on startup" | 00:30 |
| dolphm_afk | clarkb: morganfainberg: re-opened against keystone! apologies for missing that paragraph | 00:30 |
| clarkb | np | 00:30 |
| bknudson | btw, I think keystone is broken somehow... something in keystoneclient. | 00:31 |
| bknudson | I hope it wasn't me. | 00:31 |
| *** boris-42 has joined #openstack-dev | 00:32 | |
| morganfainberg | bknudson, ? | 00:32 |
| bknudson | morganfainberg: http://logs.openstack.org/53/57553/1/check/gate-keystone-python27/23d7f1a/console.html | 00:32 |
| *** atiwari has quit IRC | 00:32 | |
| *** tdruiva has quit IRC | 00:32 | |
| *** matsuhashi has joined #openstack-dev | 00:32 | |
| bknudson | lots of things getting Unauthorized: The request you have made requires authentication. (HTTP 401) | 00:32 |
| bknudson | keystone.tests.test_keystoneclient.KcMasterTestCase.test_tokens_after_user_update_passwd | 00:32 |
| morganfainberg | bknudson, that happens occationally, i meant to submit that to elastic_recheck and take a look at it | 00:32 |
| dolphm_afk | dstanek: ^^ | 00:33 |
| morganfainberg | it's related to setup failing and teardown ducking then | 00:33 |
| dolphm_afk | morganfainberg: ?! | 00:33 |
| morganfainberg | teardown* | 00:33 |
| bknudson | the last thing we want is to be adding to the recheck problems! | 00:33 |
| morganfainberg | dolphm_afk, i think last time i looked that was because a setUp() failed. | 00:33 |
| morganfainberg | dolphm_afk, and then tearDown doesn't run and screws all subsequent tests up. | 00:34 |
| morganfainberg | dolphm_afk, it has been transient, so i haven't had a chance ot chase it (seen it three times now) | 00:34 |
| bknudson | why would only 3 tests fail? | 00:34 |
| morganfainberg | order of tests? | 00:34 |
| morganfainberg | at least this _looks_ like that issue. | 00:34 |
| morganfainberg | i _think_ we even have a bug on it. | 00:34 |
| morganfainberg | if it's the same issue | 00:35 |
| morganfainberg | let me see | 00:35 |
| *** dolphm_afk is now known as dolphm_for_real_ | 00:35 | |
| *** dolphm_for_real_ is now known as dolphm_really_af | 00:35 | |
| *** dolphm_really_af is now known as dolphm_reallyafk | 00:36 | |
| dolphm_reallyafk | there we go | 00:36 |
| bknudson | must have text-to-speech | 00:36 |
| *** clayb has quit IRC | 00:36 | |
| *** hemna_ is now known as hemnafk | 00:37 | |
| morganfainberg | bknudson, i can't find the bug now :( | 00:37 |
| *** teran has joined #openstack-dev | 00:37 | |
| bknudson | too many bugs! | 00:37 |
| morganfainberg | bknudson, yes. | 00:37 |
| dstanek | dolphm_reallyafk: what's up? | 00:37 |
| morganfainberg | dstanek, ^ | 00:37 |
| *** vuil has quit IRC | 00:39 | |
| dstanek | morganfainberg: the 401s or the port issue? | 00:39 |
| morganfainberg | dstanek the 401 | 00:39 |
| morganfainberg | i think ... think that is related to setUp and tearDown, but i can't find the bug i was referencing earlier | 00:40 |
| *** buddha_dai has joined #openstack-dev | 00:40 | |
| * morganfainberg gets back to grizzly port of a bug that _needs_ to be fixed. | 00:40 | |
| morganfainberg | ugh, repro script assumes havana =/ | 00:41 |
| *** arata has joined #openstack-dev | 00:42 | |
| *** CaptTofu has quit IRC | 00:43 | |
| *** sdake_ has quit IRC | 00:44 | |
| *** CaptTofu has joined #openstack-dev | 00:44 | |
| *** MaxV has quit IRC | 00:44 | |
| *** MaxV has joined #openstack-dev | 00:45 | |
| *** buzztroll has joined #openstack-dev | 00:45 | |
| jamielennox | keystoneclient review for today: https://review.openstack.org/#/c/53038/ | 00:46 |
| jamielennox | morganfainberg: bknudson, dstanek, ayoung, gyee ^^ | 00:46 |
| morganfainberg | jamielennox, might need to wait for approval, gate has been unhappy. | 00:47 |
| morganfainberg | jamielennox, but it can be reviewed (just an FYI) | 00:47 |
| jamielennox | this is going to be my new approach i think, pick the most necessary one, one per day | 00:47 |
| bknudson | I got the 401 error locally | 00:47 |
| dstanek | bknudson: really? | 00:48 |
| bknudson | wish I had this so I could debug easier... https://review.openstack.org/#/c/54789/ | 00:48 |
| jamielennox | yea, i saw the thing with the gate - this is also about the first patch i've been able to get that passed through jenkins | 00:48 |
| jamielennox | testr and pdb? awesome i've wanted that for ages | 00:49 |
| *** MaxV has quit IRC | 00:49 | |
| bknudson | dstanek: just running tox -e py27 , got the same output as http://logs.openstack.org/53/57553/1/check/gate-keystone-python27/23d7f1a/console.html | 00:49 |
| bknudson | jamielennox: it's a hack but it's worked for me. | 00:50 |
| *** buzztroll has quit IRC | 00:50 | |
| bknudson | oh, but I need to fix it because I used the wrong filename. | 00:50 |
| jamielennox | bknudson: do you know if it will still work if testr goes parallel? | 00:50 |
| bknudson | jamielennox: I believe it will work but it won't be running in parallel. | 00:51 |
| jamielennox | i think that's a good enough compromise. i tend to resort to just using nose when i need pdb but it doesn't always seem to give the same ordering in keystone | 00:52 |
| dstanek | jamielennox: eventually testr will support pdb | 00:52 |
| jamielennox | dstanek: i've heard lots of people annoyed by the issue, but nothing about upcomming support | 00:53 |
| dstanek | bknudson: do you get the error everytime you run the tests? | 00:53 |
| bknudson | dstanek: I just kicked it off again... takes 8 minutes | 00:53 |
| bknudson | actually 571.704s | 00:54 |
| *** epim has quit IRC | 00:54 | |
| *** vuil has joined #openstack-dev | 00:54 | |
| *** CaptTofu has quit IRC | 00:55 | |
| *** CaptTofu has joined #openstack-dev | 00:55 | |
| *** devlaps has quit IRC | 00:56 | |
| *** nati_ueno has joined #openstack-dev | 00:56 | |
| dstanek | hey it breaks for me! | 00:57 |
| *** nati_ueno has quit IRC | 00:57 | |
| bknudson | dstanek: yep, it does happen every time. | 00:58 |
| bknudson | well, it happened twice. | 00:58 |
| dstanek | do you know when this started? yesterday i was running tests without issues | 00:58 |
| bknudson | dstanek: I just noticed it in the last few hours. | 01:00 |
| *** nati_ueno has joined #openstack-dev | 01:00 | |
| *** nati_ueno has quit IRC | 01:00 | |
| *** ericw has joined #openstack-dev | 01:01 | |
| *** vuil has quit IRC | 01:01 | |
| *** matsuhashi has quit IRC | 01:02 | |
| *** matsuhashi has joined #openstack-dev | 01:02 | |
| gyee | jamielennox, sure, I'll review | 01:02 |
| *** tdruiva has joined #openstack-dev | 01:04 | |
| *** nati_ueno has joined #openstack-dev | 01:05 | |
| *** mikeoutland has quit IRC | 01:05 | |
| *** nati_ueno has quit IRC | 01:06 | |
| *** epim has joined #openstack-dev | 01:06 | |
| *** nati_ueno has joined #openstack-dev | 01:06 | |
| *** jhesketh has joined #openstack-dev | 01:06 | |
| *** matsuhashi has quit IRC | 01:07 | |
| clarkb | dolphm_reallyafk: morganfainberg I added you guys to https://review.openstack.org/#/c/57577/ it solves the problem as suggested by jeblair due to the reasons I have noted in a comment there | 01:07 |
| *** tonix has quit IRC | 01:10 | |
| morganfainberg | cool | 01:10 |
| *** reed has quit IRC | 01:10 | |
| *** zul has quit IRC | 01:11 | |
| *** morazi has quit IRC | 01:11 | |
| *** gyee has quit IRC | 01:12 | |
| dstanek | morganfainberg, bknudson, dolphm_reallyafk: this is what broke http://git.openstack.org/cgit/openstack/python-keystoneclient/commit/?id=1e856a860b540fe7116534b0c8b0eef9798757b8 | 01:12 |
| *** thomasem has quit IRC | 01:13 | |
| bknudson | this is why we need keystoneclient in tempest! | 01:13 |
| *** xarses has quit IRC | 01:13 | |
| dstanek | morganfainberg, bknudson, dolphm_reallyafk: don't know why yet | 01:13 |
| morganfainberg | bknudson, ++++++++++++++++++++ | 01:13 |
| morganfainberg | bknudson, got a sec? need to run something by you | 01:14 |
| bknudson | dstanek: btw - how were you able to verify that's what broke it? | 01:14 |
| bknudson | morganfainberg: sure. | 01:14 |
| *** xingchao has joined #openstack-dev | 01:14 | |
| *** matsuhashi has joined #openstack-dev | 01:15 | |
| *** kenperkins has quit IRC | 01:18 | |
| dstanek | bknudson: use 8a0732e262cd9de508399eabeb705b681c2503c0 as the revision to checkout here: http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/test_keystoneclient.py#n860 | 01:18 |
| dstanek | bknudson: that's the revision right before the broken one | 01:19 |
| bknudson | post a revert. | 01:19 |
| *** boris-42 has quit IRC | 01:20 | |
| *** singhs has quit IRC | 01:21 | |
| *** DennyZhang has joined #openstack-dev | 01:24 | |
| *** adjohn has quit IRC | 01:25 | |
| *** asalkeld has quit IRC | 01:25 | |
| *** kenperkins has joined #openstack-dev | 01:25 | |
| *** mikeoutland has joined #openstack-dev | 01:26 | |
| *** mikeoutland has quit IRC | 01:28 | |
| dstanek | bknudson: k - the problem seems to be that the auth_token setter needs to be updated | 01:29 |
| *** adjohn has joined #openstack-dev | 01:30 | |
| bknudson | dstanek: can re-submit with the fix afterwards. | 01:30 |
| *** nosnos has joined #openstack-dev | 01:30 | |
| *** rods1 has joined #openstack-dev | 01:31 | |
| *** xchu has joined #openstack-dev | 01:31 | |
| *** amcrn has quit IRC | 01:31 | |
| *** yaguang has joined #openstack-dev | 01:31 | |
| *** dvarga has joined #openstack-dev | 01:32 | |
| *** rods has quit IRC | 01:32 | |
| jamielennox | i haven't been paying attention here, but how did that patch that hasn't been commited break the gate? | 01:32 |
| dstanek | jamielennox: it merged today | 01:32 |
| jamielennox | :O | 01:32 |
| jamielennox | wtf | 01:33 |
| *** haomaiwang has quit IRC | 01:33 | |
| jamielennox | steve and i were talking about that patch, specifically cause i said that if it went through like that it would break a lot of peoples expectations | 01:33 |
| *** xingchao has quit IRC | 01:34 | |
| *** teran has quit IRC | 01:34 | |
| bknudson | obviously we need better testing. | 01:34 |
| jamielennox | damnit, i thought i had a -1 on it but i didn't | 01:34 |
| bknudson | is there a quicker way to make it easier to test? a tox env that tests against local keystoneclient? | 01:34 |
| *** zhikunliu has joined #openstack-dev | 01:34 | |
| *** arata has left #openstack-dev | 01:35 | |
| bknudson | now that I've made a debug tox env I think it's the answer to everything. | 01:35 |
| dstanek | bknudson: you'd just need to make the checkout process more configurable - maybe use env variables if they are there | 01:36 |
| bknudson | dstanek: good idea. | 01:36 |
| jamielennox | grr, that should never have gone through, have we got a revert patch coming? | 01:36 |
| dstanek | jamielennox: coming up | 01:36 |
| *** jobewan has joined #openstack-dev | 01:37 | |
| *** sarob has joined #openstack-dev | 01:37 | |
| *** mriedem has joined #openstack-dev | 01:37 | |
| jamielennox | dstanek: thanks | 01:38 |
| *** zhikunliu has quit IRC | 01:38 | |
| jamielennox | having backed off a little bit i'd be interested to know how did that change manage to break the gate? The problem was more of an issue of changing behaviour than something i expected would cause failures like that | 01:42 |
| jamielennox | oh, not a gate failure a keystone test failure | 01:43 |
| jamielennox | but still | 01:43 |
| dstanek | jamielennox: or maybe not - having git review issues - trying to fix now | 01:43 |
| dstanek | init is shitting out with a 'fatal: Needed a single revision' :-( | 01:43 |
| jamielennox | never seen that, want me to do it? | 01:44 |
| dstanek | jamielennox: sure if you have the time | 01:45 |
| *** matsuhashi has quit IRC | 01:45 | |
| *** matsuhashi has joined #openstack-dev | 01:46 | |
| *** buzztroll has joined #openstack-dev | 01:46 | |
| *** rongze has joined #openstack-dev | 01:47 | |
| dstanek | jamielennox: nm it finally worked | 01:48 |
| dstanek | jamielennox: i just needed to try harder - https://review.openstack.org/#/c/57583 | 01:49 |
| jamielennox | dstanek: just in time, i was about to push go | 01:50 |
| *** buzztroll has quit IRC | 01:50 | |
| dstanek | jamielennox: it turns out my memory sucks - i was using 'git review init' instead of 'git review -s' | 01:50 |
| jamielennox | ? | 01:51 |
| *** rongze has quit IRC | 01:51 | |
| jamielennox | oh right, first commit to client | 01:51 |
| dstanek | the -s version is how you init a repository for use with git-review | 01:51 |
| dstanek | i just re-did my devstack | 01:51 |
| jamielennox | ok, +1ed | 01:53 |
| *** haomaiwang has joined #openstack-dev | 01:53 | |
| *** xingchao has joined #openstack-dev | 01:53 | |
| dstanek | morganfainberg, bknudson: here's the revert commit https://review.openstack.org/#/c/57583 | 01:54 |
| bknudson | dstanek: can you also submit the original change? | 01:54 |
| bknudson | re-submit | 01:55 |
| *** jobewan has quit IRC | 01:55 | |
| *** angdraug has quit IRC | 01:55 | |
| dstanek | bknudson: with my fix? | 01:56 |
| bknudson | I'd prefer without the fix and then include your fix, but just with the fix is fine. | 01:56 |
| *** sarob has quit IRC | 01:58 | |
| dstanek | bknudson: should i have it roll a new changeset-id? i think i'll need to so gerrit is happy | 01:59 |
| bknudson | dstanek: I think you'll need a new changeset. | 01:59 |
| bknudson | dstanek: but can you keep the author shardy? | 02:00 |
| *** rods1 has quit IRC | 02:00 | |
| *** nati_uen_ has joined #openstack-dev | 02:00 | |
| dstanek | bknudson: yes, i'm just going to cherrypick it and remove the changeset-id; pretty sure that will keep shardy as the author | 02:01 |
| dstanek | bknudson: https://review.openstack.org/57585 | 02:03 |
| bknudson | dstanek: you're the Owner. | 02:03 |
| bknudson | but the author is set... so that's probably ok. | 02:03 |
| dstanek | i think the owner will always be the owner of the ssh key that pushed the review | 02:03 |
| bknudson | jamielennox: https://review.openstack.org/57585 | 02:03 |
| *** nati_ueno has quit IRC | 02:04 | |
| *** rods has joined #openstack-dev | 02:04 | |
| *** DennyZhang has quit IRC | 02:05 | |
| bknudson | dstanek: took your advice: https://review.openstack.org/#/c/57586/ | 02:05 |
| *** renlt has joined #openstack-dev | 02:05 | |
| *** tanisdl has quit IRC | 02:06 | |
| dstanek | bknudson: nice, trying it out now | 02:06 |
| *** zhikunliu has joined #openstack-dev | 02:07 | |
| *** _mattf is now known as mattf | 02:07 | |
| *** bingbu has joined #openstack-dev | 02:09 | |
| *** sgran has quit IRC | 02:09 | |
| *** sgran has joined #openstack-dev | 02:10 | |
| *** michchap_ has joined #openstack-dev | 02:11 | |
| *** moted has quit IRC | 02:11 | |
| *** ericw has quit IRC | 02:12 | |
| *** Mandell has quit IRC | 02:12 | |
| *** msmedved has quit IRC | 02:13 | |
| *** buzztroll has joined #openstack-dev | 02:13 | |
| *** adjohn has quit IRC | 02:13 | |
| *** buzztroll has quit IRC | 02:14 | |
| *** buzztroll has joined #openstack-dev | 02:14 | |
| *** michchap has quit IRC | 02:14 | |
| morganfainberg | bknudson, added you to this https://review.openstack.org/#/c/57577 just so you can see it. | 02:15 |
| morganfainberg | port change thing in devstack. | 02:15 |
| *** nati_uen_ has quit IRC | 02:15 | |
| *** msmedved has joined #openstack-dev | 02:16 | |
| *** nati_ueno has joined #openstack-dev | 02:16 | |
| bknudson | morganfainberg: can't say I agree with changing keystone port. | 02:16 |
| morganfainberg | bknudson, thats my point of the comment. | 02:16 |
| bknudson | not what was discussed here. | 02:16 |
| *** erkules_ has joined #openstack-dev | 02:16 | |
| dolphm_reallyafk | bknudson: ++ | 02:16 |
| dolphm_reallyafk | morganfainberg: what was wrong with bknudson's solution? | 02:16 |
| morganfainberg | dolphm_reallyafk, it's still a race for things running before devstack starts running (stack.sh) | 02:17 |
| *** dolphm_reallyafk is now known as dolphm | 02:17 | |
| morganfainberg | dolphm_reallyafk, same issue, just a little less possible | 02:17 |
| morganfainberg | the correct fix (imo) is to make the images d-g instances run on use the correct port range | 02:17 |
| *** hartsocks has joined #openstack-dev | 02:18 | |
| dolphm | morganfainberg: this is a rare transient issue though -- right? no need to rush in a fix? | 02:18 |
| bknudson | It's just going to lead to other problems... if you have a script or something that assumes 35357 since that's what IANA assigned. | 02:18 |
| ayoung | lets use 443 and do it in HTTPD | 02:18 |
| morganfainberg | bknudson, please jump in and comment. | 02:18 |
| bknudson | I know we have all sorts of examples. | 02:18 |
| dolphm | bknudson: that *was* kind of the point of getting an IANA assignment | 02:18 |
| *** sarob has joined #openstack-dev | 02:18 | |
| morganfainberg | dolphm, +++ | 02:18 |
| *** ArxCruz has quit IRC | 02:18 | |
| *** erkules has quit IRC | 02:19 | |
| morganfainberg | lol tempest doesn't like it. wonder if it makes assumptions about the port(s) | 02:19 |
| *** epim has quit IRC | 02:19 | |
| dolphm | morganfainberg: probably lol | 02:20 |
| morganfainberg | dolphm, i think infra is looking for a fix sooner vs later that will help keep gate churn down. | 02:20 |
| *** michchap has joined #openstack-dev | 02:21 | |
| ayoung | morganfainberg, than they should drop Neutron testing | 02:21 |
| ayoung | This is minor | 02:21 |
| morganfainberg | ayoung, i don't have an argument to that logic | 02:21 |
| bknudson | could we change keystone to wait until the port is available? | 02:21 |
| morganfainberg | bknudson, i'm sure we could. | 02:22 |
| bknudson | it's pretty common to have servers wait until networking is available (on os's where networking might not be up yet) | 02:22 |
| *** rongze has joined #openstack-dev | 02:22 | |
| *** moted has joined #openstack-dev | 02:22 | |
| *** aditirav has joined #openstack-dev | 02:22 | |
| ayoung | we could stop doing random ports and keep the web on 443 | 02:23 |
| dolphm | bknudson: i'd rather fail fast there | 02:23 |
| morganfainberg | ayoung, i think 443 is fine, but it's silly we can't use our IANA assigned port if we want to. | 02:24 |
| *** michchap_ has quit IRC | 02:24 | |
| *** boris-42 has joined #openstack-dev | 02:24 | |
| morganfainberg | if we can't use the port, whats the point of getting one registered? | 02:24 |
| *** kdbrierly has quit IRC | 02:25 | |
| ayoung | jamielennox, aside from breaking the tests in Keystone client, do you follow my logic in the client review? | 02:25 |
| ayoung | "Refreshing of Tokens and enforcement of that workflow is done on the server side. The client should be able to request a token for a token, Trust or no, as that new token cannot get a longer lifespan than the original. With the upcoming changes in "revoke" by expires_at, we will also makes sure they are all revoked together." | 02:25 |
| dolphm | morganfainberg: you mean, what's the point of IANA assigning an unusable port? | 02:26 |
| *** novas0x2a|laptop has quit IRC | 02:26 | |
| jamielennox | ayoung: 1 minute | 02:26 |
| *** aditirav has quit IRC | 02:27 | |
| morganfainberg | dolphm, well considering IANA said here are the new recommendations and linux maintainers said "oh well that doesn't give us enough ephemeral ports" (circa 2007) who really is in the right here? ( | 02:27 |
| *** ftcjeff has joined #openstack-dev | 02:27 | |
| bknudson | let's just go to 64-bit ports. | 02:29 |
| *** teran has joined #openstack-dev | 02:29 | |
| morganfainberg | bknudson, TCPv6? | 02:29 |
| * morganfainberg ducks | 02:29 | |
| *** buzztroll has quit IRC | 02:30 | |
| ayoung | lets stop requesting ports for random web services | 02:30 |
| *** sdake_ has joined #openstack-dev | 02:30 | |
| *** buzztroll has joined #openstack-dev | 02:30 | |
| *** alex_xu has joined #openstack-dev | 02:30 | |
| dstanek | bknudson: ah, i see what you are going in that review - you run another set of test *if* the env vars are set | 02:31 |
| morganfainberg | ayoung, ++ | 02:31 |
| bknudson | dstanek: right. | 02:31 |
| dstanek | bknudson: i like it! | 02:31 |
| bknudson | dstanek: maybe I need to clarify the commit message or comments. | 02:32 |
| *** loq_mac has quit IRC | 02:32 | |
| jamielennox | ayoung: ok - was finishing a strongly worded letter to the upstream HTTPretty dev to get his arse into gear | 02:32 |
| ayoung | morganfainberg, I was at an Open Stack meetup lastnight, and an Ubuntu sales engineer was demoing Juju. He shoew the OS set up, and they had the Auth Url explicitly on 443. | 02:33 |
| morganfainberg | ayoung, HTTP? or eventlet? | 02:33 |
| bknudson | ayoung: did you get icecream? | 02:33 |
| ayoung | morganfainberg, he was cloooless about Open Stack. I didn't tormet him with detailed questions...I was just kindof annoyed at getting a sales pitch....not at him, but at the organizers | 02:34 |
| ayoung | bknudson, no, Pizza and a chocolate chip cookie | 02:34 |
| jamielennox | ayoung: so the problem is that it's not to do with token refreshing, it's to do with the fact that tokens given to the client are treated as if there only use is doing token/endpoint authentication | 02:34 |
| *** buzztroll has quit IRC | 02:34 | |
| morganfainberg | ayoung, meetup + sales pitch, isn't that the norm? | 02:34 |
| morganfainberg | /s | 02:34 |
| jamielennox | ayoung: the client has code there to handle auth_url/token auth but somewhere along the way it got lost | 02:34 |
| *** tonyfy has joined #openstack-dev | 02:35 | |
| jamielennox | he and i discussed this at summit and again on irc the other day and i said i'd get something together to handle auth_url/token auth | 02:35 |
| bknudson | I think canonical was giving out ice cream at the summit... I didn't get to their booth. | 02:35 |
| ayoung | jamielennox, I need more context | 02:36 |
| jamielennox | unfortunately i didn't remember to put a -1 on it | 02:36 |
| ayoung | jamielennox, I should have waited for you. | 02:36 |
| ayoung | So what happend? What broke> | 02:37 |
| morganfainberg | bknudson, i think i only stumbled into booths to get shirts...and didn't even look at any other things | 02:37 |
| jamielennox | i don't know from the keystone side | 02:37 |
| ayoung | I got ice cream at the Summit, but not sure if it was Cannonical that handed it out. | 02:37 |
| *** bdpayne has quit IRC | 02:37 | |
| bknudson | ayoung: did you get to pick the toppings, etc? | 02:37 |
| ayoung | jamielennox, my new favorite client is curl and jq anyway | 02:37 |
| *** dolphm has quit IRC | 02:38 | |
| jamielennox | ayoung: only cause keystoneclient is ugly to work with | 02:38 |
| *** jhesketh__ has quit IRC | 02:38 | |
| bknudson | we need keystoneclient-tng | 02:38 |
| bknudson | (the next generation) | 02:38 |
| ayoung | jamielennox, and because jq is kindof cool | 02:38 |
| jamielennox | i've no idea what jq is | 02:38 |
| ayoung | I need to do one more blog post showing how to use the web API with trusts | 02:39 |
| *** alop has quit IRC | 02:39 | |
| jamielennox | bknudson: there was a thing on dev list about glance trying to bump to 1.0 and people being upset and suggesting it needs to be glanceclient2 | 02:39 |
| morganfainberg | jamielennox, seriously?! | 02:39 |
| jamielennox | maybe we could get to the point of keystoneclient-nnnnng | 02:39 |
| ayoung | jamielennox, it is a native utility for querying JSON. Very elegant. | 02:40 |
| bknudson | ayoung: like xpath? | 02:40 |
| jamielennox | is there a nice way to figure out the lists.openstack.org url of an email from the actual email? | 02:40 |
| ayoung | bknudson, yeah...used it in my last blog post: | 02:40 |
| *** danwent has quit IRC | 02:40 | |
| ayoung | curl -s -H"X-Auth-Token:$TOKEN" http://localhost:35357/v3/roles | jq '.roles[] | select( contains({name: "usermanager"})) | {id}[] ' | 02:40 |
| morganfainberg | jamielennox, about the service catalog reorg | 02:41 |
| bknudson | ayoung: does it have a python api? | 02:41 |
| morganfainberg | jamielennox, why the use of @classmethods? | 02:41 |
| morganfainberg | are you calling them on the class non-instantiated? | 02:41 |
| *** rm_work is now known as rm_work|away | 02:41 | |
| ayoung | .roles[] is the main collection there, presentaed as an array/list and then select means "print this object" with contains being a boolean to check that name == usermanager...finally it just pulls out the id field | 02:41 |
| *** jhesketh__ has joined #openstack-dev | 02:42 | |
| morganfainberg | jamielennox, RE: mailing list, i google search the subject usually >.< | 02:42 |
| jamielennox | no, but they aren't instance related methods - they relate to something that is inherit to the way the class works | 02:42 |
| *** yamahata_ has quit IRC | 02:42 | |
| jamielennox | thus classmethod | 02:42 |
| ayoung | bknudson, I'm not sure if there is a native library version, I've only used it from Bash | 02:42 |
| morganfainberg | jamielennox, if they are expected to be overridded @abc.abstractmethod, if they are not being called externally, they could be extracted to functions in the module | 02:43 |
| ayoung | bknudson, https://github.com/stedolan/jq.git | 02:43 |
| ayoung | looks like he had an libjq at some oint | 02:44 |
| ayoung | point | 02:44 |
| jamielennox | they could be functions in the module. I wasn't thinking about abc because it isn't being used - it's also not something that anyone would ever want to write their own implementation of so i'm not sure we need abc | 02:44 |
| ayoung | bknudson, but jq as installed on my machine is just a single elf exe | 02:45 |
| morganfainberg | i'm going to say we should move them to functions because it doesn't seem like they are really @classmethods, they are private and not meant to be called externally with class specific data. if they weren't private they could be @staticmethods it looks like | 02:45 |
| *** dolphm has joined #openstack-dev | 02:45 | |
| *** tonyfy has quit IRC | 02:45 | |
| *** dkranz has joined #openstack-dev | 02:45 | |
| morganfainberg | jamielennox, so in short, since no one would really override / implement new versions, module fuctions sounds like the best bet | 02:46 |
| jamielennox | glance 1.0 http://lists.openstack.org/pipermail/openstack-dev/2013-November/019911.html | 02:46 |
| *** anniec has quit IRC | 02:46 | |
| morganfainberg | new solution, make all clients 1.0 as of tomorrow. make major versions turn into like glanceclient2 | 02:47 |
| jamielennox | yes, they could be staticmethods - i read a thing at one point about how staticmethods were a bad idea and it is more pythonic to use classmethod to get the cls reference and i've just used classmethod by default ever since | 02:47 |
| morganfainberg | jamielennox, right. i have 1 use case for static really, and that was the v3_to_v2 user method | 02:48 |
| dstanek | classmethods are great for factory functions | 02:48 |
| *** paragan has joined #openstack-dev | 02:48 | |
| *** paragan has quit IRC | 02:48 | |
| *** paragan has joined #openstack-dev | 02:48 | |
| ayoung | jamielennox, I think it is one of those things that people just decide one is more Pythonic than the rest, whereas reality is they are both just doing roughly the same thing semantically. I see no reason to favor one or the other | 02:48 |
| morganfainberg | jamielennox, i wouldn't make something a class method unless it needs reference to cls. (yep factor functions is a great example) | 02:49 |
| ayoung | dstanek, morganfainberg but since Python doesn't do type checking, even there it really is meaningless | 02:49 |
| jamielennox | morganfainberg: that would be the same logic behind not making these methods because they don't reference self | 02:50 |
| jamielennox | anyway, i'm really not worried on this | 02:50 |
| dstanek | ayoung: classmethods? | 02:50 |
| morganfainberg | ayoung, heh. | 02:50 |
| jamielennox | i'll change them to regular methods if that is wanted | 02:50 |
| morganfainberg | jamielennox, i think it's just cleaner if they aren't turned into class methods, personal opinion. i really was more curious at the choice since you didn't need the cls reference | 02:51 |
| morganfainberg | (the factory method aside that is) | 02:51 |
| jamielennox | so staticmethod then? | 02:51 |
| *** guohliu has joined #openstack-dev | 02:51 | |
| morganfainberg | jamielennox, not heartset on either. i'll let it go as is, like i said, was curious more than anything else. (not going to -1 for that) | 02:52 |
| *** paragan has quit IRC | 02:52 | |
| jamielennox | ok, no-one around here is shy with the -1 so i'll take it as a fix if i have to do another pass | 02:52 |
| clarkb | ayoung: morganfainberg: if keystone sits behind apache it runs as a wsgi app? and when doing that only one port is needed instead of the standalone 2? | 02:52 |
| clarkb | finally if that is the preferred solution why isn't it the devstack default? | 02:53 |
| morganfainberg | clarkb, i'll defer to ayoung on why it isn't the default (not sure). but yes it can use a single port, the catalog needs ot be aware of the correct pathing information (e.g. /main vs /admin for V2) | 02:54 |
| clarkb | switching to apache doesn't solve bknudson's concerns either | 02:54 |
| morganfainberg | no it doesn.t | 02:54 |
| ayoung | momentum | 02:54 |
| ayoung | its like the Bananas and the gorillas. | 02:55 |
| *** sarob has quit IRC | 02:55 | |
| clarkb | (I am happy to use apache, just trying to understand what that means) | 02:55 |
| ayoung | what concerns does it not solve? It will get us inner peace and harmony. | 02:55 |
| morganfainberg | ayoung, momentum and example scripts | 02:56 |
| *** sarob has joined #openstack-dev | 02:56 | |
| clarkb | ayoung: bknudon's concerns are that he and probably others have tools built to use 35357 already | 02:56 |
| clarkb | switching to apache by default is just as disruptive to that as using a different port | 02:56 |
| *** achampion has joined #openstack-dev | 02:56 | |
| clarkb | maybe more so if the API changes | 02:56 |
| ayoung | clarkb, true. But then, in this case we can claim we are rectifying a mistake | 02:56 |
| *** sdake_ has quit IRC | 02:56 | |
| morganfainberg | clarkb, i think it's as disruptive but not as bad unless we abandon IANA port assignment | 02:57 |
| ayoung | https://wiki.openstack.org/wiki/URLs | 02:57 |
| *** zhikunliu has quit IRC | 02:58 | |
| ayoung | So, ports are not only dumb, they actually avoid the real problem, which is that admin and public needed to be on different INTERFACES not ports. | 02:58 |
| ayoung | and Keystone didn't actually provide for that | 02:58 |
| *** d34dh0r53 has joined #openstack-dev | 02:59 | |
| *** thomasem has joined #openstack-dev | 02:59 | |
| *** RajM has quit IRC | 02:59 | |
| dolphm | ayoung: paste is configurable enough to achieve that | 02:59 |
| ayoung | dolphm, I know...but it was more code that had to be written. It was just....poorly considered | 03:00 |
| dolphm | ayoung: true | 03:00 |
| *** sarob has quit IRC | 03:00 | |
| clarkb | we should also consider that we probably won't be enable to use apache with grizzly and havana, but the test load there is probably low enough that we can deal with the annoying port in use problem | 03:01 |
| *** thomasem has quit IRC | 03:01 | |
| morganfainberg | clarkb, i'm not super worried about stable/* at the moment. for the reason you stated | 03:01 |
| *** ngoracke has joined #openstack-dev | 03:02 | |
| *** jasdeepH has joined #openstack-dev | 03:02 | |
| *** teran has quit IRC | 03:03 | |
| ayoung | clarkb, so the short of it is that we should be deploying pretty much everything inside of Apache. Eventlet is (a proper noun not usable in polite society) | 03:03 |
| ayoung | especially for our needs | 03:03 |
| ayoung | It kindof makes sense for the Messagin side of things, but even there...shudder | 03:03 |
| ayoung | THe security team has come out with a recommendation that we do SSL everywhere | 03:04 |
| ayoung | SSL in Python? I sure hope not | 03:04 |
| *** dolphm has quit IRC | 03:04 | |
| *** dvarga has quit IRC | 03:04 | |
| *** sdake_ has joined #openstack-dev | 03:04 | |
| *** renlt has quit IRC | 03:04 | |
| morganfainberg | ayoung, sadly, apache fronting the services doesn't solve SSL inter-service and pythons woeful encryption support | 03:04 |
| morganfainberg | but the public interfaces at least are much better. | 03:04 |
| ayoung | it solves it for anything doing HTTP. | 03:05 |
| ayoung | there are still client issues | 03:05 |
| morganfainberg | compute process talking to keystone? python still has to use the client -> pure python -> SSL -> service | 03:05 |
| clarkb | if someone wants to propose a change to devstack to make apache the default I would be on board with that | 03:05 |
| *** stevemar has joined #openstack-dev | 03:05 | |
| clarkb | (and can abandon my change) | 03:05 |
| *** matsuhas_ has joined #openstack-dev | 03:05 | |
| *** xarses has joined #openstack-dev | 03:05 | |
| ayoung | clarkb, I propose to make HTTPD the default in Devstack. | 03:06 |
| morganfainberg | i am ok with that as it's been our "recommended production deployment" for a while. | 03:06 |
| ayoung | since 2012/03 according to my blog | 03:06 |
| ayoung | http://adam.younglogic.com/2012/03/keystone-should-move-to-apache-httpd/ | 03:06 |
| morganfainberg | i think there is likely some work to be done if tempest uses explicitly 35357. | 03:07 |
| morganfainberg | i don't know if it always references service catalog | 03:07 |
| clarkb | morganfainberg: if you look at the failures related to my change it appears that nova may assume things | 03:07 |
| morganfainberg | ah | 03:07 |
| morganfainberg | boo. | 03:07 |
| clarkb | but I didn't dig far enough to double check (something gets a bunch of 401s) | 03:08 |
| *** matsuhashi has quit IRC | 03:08 | |
| morganfainberg | clarkb, might just be a auth_token config issue | 03:09 |
| *** nati_uen_ has joined #openstack-dev | 03:10 | |
| *** paragan has joined #openstack-dev | 03:10 | |
| jamielennox | side note when people aren't busy can someone send me some examples of what that client patched caused to fail. the more i think about it the more i think it shouldn't have caused a problem for tests | 03:11 |
| *** ericwb has quit IRC | 03:12 | |
| *** nati_ueno has quit IRC | 03:14 | |
| morganfainberg | ok i am going home... food calls. | 03:15 |
| bknudson | jamielennox: http://logs.openstack.org/53/57553/1/check/gate-keystone-python27/23d7f1a/console.html | 03:16 |
| clarkb | is https://github.com/openstack-dev/devstack/blob/master/lib/keystone#L111 a noop? | 03:16 |
| bknudson | jamielennox: is that what you were asking for? | 03:16 |
| *** fifieldt has quit IRC | 03:16 | |
| *** twoputt has quit IRC | 03:16 | |
| bknudson | jamielennox: I could recreate it locally. | 03:17 |
| *** twoputt_ has quit IRC | 03:17 | |
| ayoung | tempest should use the values out of open.rc | 03:17 |
| *** fifieldt has joined #openstack-dev | 03:18 | |
| bknudson | is there some way to test that I can't listen on a port that client is using? | 03:18 |
| ayoung | If it does use 35357 explicitly it is a mistake | 03:18 |
| jamielennox | bknudson: if that's the test that was failing that's fine. i could probably recreate it as well, i just assumed it had come up some where on the gate | 03:18 |
| clarkb | ayoung: I don't think keystone is failing as it does use the openrc. I think it is something else | 03:18 |
| bknudson | jamielennox: it was those 2 tests failing. | 03:18 |
| ayoung | bknudson, my immediate thought was "use telnet" as that is pretty much how I connect to any tcp port for debugging | 03:18 |
| clarkb | during the devstack setup of things | 03:18 |
| jamielennox | bknudson: i think at least some of my concerns aren't valid so i'm just wondering what about the change caused the fail | 03:19 |
| bknudson | jamielennox: I didn't look into it... dstanek had a proposed fix. | 03:19 |
| ayoung | clarkb, and everything else should be getting their urls out of the service catalog. Which should be populated via devstack...let me look at the failure | 03:19 |
| *** lbragstad has joined #openstack-dev | 03:19 | |
| bknudson | ok, I was able to recreate pretty easily. | 03:20 |
| ayoung | clarkb, swift | 03:20 |
| ayoung | 2013-11-21 01:39:13.099 | 2013-11-21 01:39:12 ++ sed -ne '/^\[filter:authtoken\]/,/^\[.*\]/ { /^auth_port[ \t]*=/ p; }' /etc/swift/proxy-server.conf | 03:20 |
| ayoung | 2013-11-21 01:39:13.100 | 2013-11-21 01:39:12 + line='auth_port = 35357' | 03:20 |
| ayoung | http://logs.openstack.org/77/57577/2/check/check-tempest-devstack-vm-full/21f7af6/console.html | 03:21 |
| *** sdake_ has quit IRC | 03:21 | |
| ayoung | pretty much all of them | 03:21 |
| clarkb | right so first step is fixing that, then we can switch to apache or anything else | 03:21 |
| bknudson | it even gets stuck in time_wait -- tcp 0 0 127.0.0.1:47083 127.0.0.1:22 TIME_WAIT - | 03:22 |
| jamielennox | bknudson: ok, so it's because the test is overriding the auth_token manually even though it has already done an authentication | 03:22 |
| bknudson | jamielennox: it could definitely be the test is invalid. | 03:22 |
| jamielennox | umm, it's really bad practice but i'd be surprised if the heat or openstack-client guys aren't doing something similar | 03:23 |
| *** fifieldt has quit IRC | 03:23 | |
| ayoung | heh, that review is 57577 | 03:23 |
| *** sn0wb1rd has quit IRC | 03:24 | |
| ayoung | clarkb, http://paste.fedoraproject.org/55602/85004290 | 03:25 |
| *** DennyZhang has joined #openstack-dev | 03:25 | |
| *** Shaan7 has quit IRC | 03:26 | |
| *** renlt has joined #openstack-dev | 03:26 | |
| clarkb | ayoung: the account rc's aren't coming out right? | 03:26 |
| ayoung | hmm...nah, probably need to rerun it...that is a generated file | 03:26 |
| ayoung | disregard | 03:27 |
| dstanek | jamielennox: the fix that i had in mind was must to set the auth_info to None in the auth_token setter | 03:27 |
| ayoung | clarkb, ok, that last mistake shows that it is time for me to turn in. Gnight | 03:28 |
| clarkb | good night | 03:28 |
| dstanek | jamielennox: it fixes the tests, but i don't know if it help with your concerns about expectations | 03:28 |
| *** alex_xu has quit IRC | 03:29 | |
| *** thelorax123 has quit IRC | 03:30 | |
| *** melwitt has quit IRC | 03:30 | |
| *** thelorax123 has joined #openstack-dev | 03:31 | |
| dstanek | morganfainberg: let me know when you have a few minutes | 03:33 |
| lbragstad | keystone party? | 03:34 |
| jamielennox | aint no party like a keystone party | 03:35 |
| lbragstad | ^ +1 | 03:36 |
| jamielennox | dstanek: my concern with the expectations was that if you pass endpoint/token then firstly you don't know where to update the token from, and in the case of using auth_url/token we need to be able to work in the same was as user/pass | 03:37 |
| ayoung | clarkb, 35357 is embedded in the conf files. I just spot checked glance | 03:38 |
| ayoung | etc/glance-api.conf:439:auth_port = 35357 | 03:38 |
| ayoung | etc/glance-registry.conf:88:auth_port = 35357 | 03:38 |
| jamielennox | but i've had a bit more of a look and at least some of that isn't relevant because the other way of retrieving a token still checks for auth_ref which you only get if you go through the authentication process | 03:38 |
| ayoung | swift too | 03:39 |
| ayoung | etc/proxy-server.conf-sample:237:# auth_port = 35357 | 03:39 |
| *** nati_uen_ has quit IRC | 03:39 | |
| jamielennox | i didn't think we handle the token in the auth method either but apparently we do, so much of what shardy did was correct - though maybe for slightly different reasons | 03:39 |
| *** nati_ueno has joined #openstack-dev | 03:40 | |
| *** buzztroll has joined #openstack-dev | 03:40 | |
| dstanek | jamielennox: this command shows how the change breaks keystone: http://paste.openstack.org/show/53715/ | 03:41 |
| *** rods has quit IRC | 03:43 | |
| ayoung | Heh....I was tripping over that dstanek | 03:43 |
| ayoung | and yes, now I really am going to bed | 03:43 |
| *** ayoung is now known as ayoung-zZzZzZ | 03:44 | |
| jamielennox | yea, i can see why | 03:44 |
| *** buzztroll has quit IRC | 03:45 | |
| *** sn0wb1rd has joined #openstack-dev | 03:45 | |
| *** fifieldt has joined #openstack-dev | 03:46 | |
| *** pablosan has joined #openstack-dev | 03:46 | |
| *** alex_xu has joined #openstack-dev | 03:47 | |
| *** loq_mac has joined #openstack-dev | 03:50 | |
| *** giulivo has quit IRC | 03:50 | |
| *** cdub has quit IRC | 03:54 | |
| *** CaptTofu has quit IRC | 03:54 | |
| *** CaptTofu has joined #openstack-dev | 03:55 | |
| *** michchap has quit IRC | 03:55 | |
| *** alex_xu has quit IRC | 03:57 | |
| *** aditirav has joined #openstack-dev | 03:57 | |
| *** boris-42 has quit IRC | 03:57 | |
| *** nati_uen_ has joined #openstack-dev | 03:57 | |
| *** boris-42 has joined #openstack-dev | 03:57 | |
| *** nati_uen_ has quit IRC | 03:58 | |
| *** aditirav has quit IRC | 03:59 | |
| *** teran has joined #openstack-dev | 03:59 | |
| *** aditirav has joined #openstack-dev | 03:59 | |
| *** matsuhas_ has quit IRC | 04:00 | |
| *** prekarat has joined #openstack-dev | 04:01 | |
| *** nati_ueno has quit IRC | 04:01 | |
| *** matsuhashi has joined #openstack-dev | 04:02 | |
| *** buzztroll has joined #openstack-dev | 04:02 | |
| *** gordc has joined #openstack-dev | 04:03 | |
| *** ftcjeff has quit IRC | 04:04 | |
| *** mriedem has quit IRC | 04:04 | |
| *** chandankumar has joined #openstack-dev | 04:07 | |
| *** sandywalsh has quit IRC | 04:07 | |
| *** mikeoutland has joined #openstack-dev | 04:08 | |
| *** cyberx has joined #openstack-dev | 04:08 | |
| cyberx | Hello, good morning/evening/afternoon. I am having a bit of an issue getting devstack up and running on centos 6.4. I hope some one may be able to help. | 04:12 |
| cyberx | keystone fails to start | 04:12 |
| cyberx | File "/opt/stack/keystone/bin/keystone-all", line 35, in <module> | 04:12 |
| cyberx | from paste import deploy | 04:12 |
| cyberx | ImportError: cannot import name deploy | 04:12 |
| *** matsuhashi has quit IRC | 04:14 | |
| *** buzztroll has quit IRC | 04:14 | |
| *** buzztroll has joined #openstack-dev | 04:15 | |
| cyberx | running keystone-all manually produces the same error. After I install: yum install python-paste-deploy.noarch I am able to run keystone-all manually | 04:15 |
| cyberx | however, running stack.sh again after installing that package appears to break it again and stack.sh fails again in the same manner | 04:16 |
| *** kaushikc has joined #openstack-dev | 04:19 | |
| clarkb | cyberx: pip should be installing keystone's requirements which include PasteDeploy, is that failing for some reason? | 04:19 |
| *** matsuhashi has joined #openstack-dev | 04:19 | |
| cyberx | it appears so | 04:19 |
| *** buzztroll has quit IRC | 04:19 | |
| clarkb | you should be able to `pip install PasteDeploy` by hand and see if it falls over | 04:20 |
| cyberx | thank you so much for advise let me see where that gets me | 04:20 |
| *** kaushikc has quit IRC | 04:20 | |
| *** coolsvap has joined #openstack-dev | 04:21 | |
| *** DinaBelova has joined #openstack-dev | 04:22 | |
| *** odyssey4me has joined #openstack-dev | 04:23 | |
| *** DinaBelova has quit IRC | 04:23 | |
| *** danwent has joined #openstack-dev | 04:27 | |
| *** coolsvap has quit IRC | 04:27 | |
| *** coolsvap_ has joined #openstack-dev | 04:27 | |
| *** coolsvap_ has left #openstack-dev | 04:28 | |
| *** mikeoutland has quit IRC | 04:31 | |
| *** cfriesen_ has quit IRC | 04:32 | |
| *** teran has quit IRC | 04:32 | |
| *** buzztroll has joined #openstack-dev | 04:33 | |
| *** sumanthns has joined #openstack-dev | 04:35 | |
| *** dkranz has quit IRC | 04:35 | |
| *** loq_mac has quit IRC | 04:35 | |
| *** dkranz has joined #openstack-dev | 04:36 | |
| *** sandywalsh has joined #openstack-dev | 04:37 | |
| *** sdake_ has joined #openstack-dev | 04:38 | |
| *** rdas has joined #openstack-dev | 04:39 | |
| cyberx | ah ok, I see what my problem is. For whatever reason I forced devstack to run using the grizzly branch, it installed many python libraries using pip - albeit older versions. Even though I cleaned out /opt/stack and cloned the latest branch again the older python libs installed during that run are lingering around. Looks like I will be good after I track them all down and uninstall via pip and re-run stack.sh… thank you! | 04:41 |
| clarkb | np | 04:42 |
| clarkb | pip does not upgrade by default you ahve to give it the --upgrade flag | 04:42 |
| *** aditirav_ has joined #openstack-dev | 04:47 | |
| cyberx | :) excellent - thanks again man saved me some time - take care | 04:47 |
| *** aditirav_ has quit IRC | 04:48 | |
| *** aditirav_ has joined #openstack-dev | 04:48 | |
| *** aditirav has quit IRC | 04:48 | |
| *** aditirav_ is now known as aditirav | 04:48 | |
| *** kumaranvram has joined #openstack-dev | 04:49 | |
| *** rushiagr has joined #openstack-dev | 04:54 | |
| *** steven-weston_ has quit IRC | 04:54 | |
| *** nati_ueno has joined #openstack-dev | 04:57 | |
| *** adjohn has joined #openstack-dev | 04:57 | |
| *** DennyZhang has quit IRC | 04:59 | |
| *** michchap has joined #openstack-dev | 05:01 | |
| *** matsuhashi has quit IRC | 05:02 | |
| *** odyssey4me has quit IRC | 05:03 | |
| *** hartsocks has quit IRC | 05:03 | |
| *** renlt has quit IRC | 05:04 | |
| *** afazekas has quit IRC | 05:05 | |
| *** radix_ has quit IRC | 05:05 | |
| *** rraja has joined #openstack-dev | 05:06 | |
| *** renlt has joined #openstack-dev | 05:07 | |
| *** bdpayne has joined #openstack-dev | 05:12 | |
| *** gordc has quit IRC | 05:13 | |
| *** loq_mac has joined #openstack-dev | 05:17 | |
| *** stevemar has quit IRC | 05:18 | |
| *** loq_mac has quit IRC | 05:19 | |
| *** loq_mac has joined #openstack-dev | 05:20 | |
| *** odyssey4me has joined #openstack-dev | 05:20 | |
| *** loq_mac has quit IRC | 05:20 | |
| *** loq_mac has joined #openstack-dev | 05:21 | |
| *** odyssey4me2 has joined #openstack-dev | 05:21 | |
| *** odyssey4me3 has joined #openstack-dev | 05:23 | |
| *** achampion has quit IRC | 05:24 | |
| *** odyssey4me has quit IRC | 05:25 | |
| *** garyk has quit IRC | 05:26 | |
| *** odyssey4me2 has quit IRC | 05:26 | |
| *** afazekas has joined #openstack-dev | 05:27 | |
| *** afazekas has quit IRC | 05:27 | |
| *** dkranz has quit IRC | 05:29 | |
| *** teran has joined #openstack-dev | 05:29 | |
| *** dkranz has joined #openstack-dev | 05:29 | |
| *** bdpayne has quit IRC | 05:31 | |
| *** loq_mac has quit IRC | 05:33 | |
| *** mihgen has joined #openstack-dev | 05:35 | |
| *** zul has joined #openstack-dev | 05:37 | |
| *** sarob has joined #openstack-dev | 05:39 | |
| *** kushal has joined #openstack-dev | 05:40 | |
| *** sarob has quit IRC | 05:43 | |
| *** SergeyLukjanov has joined #openstack-dev | 05:48 | |
| *** rushiagr has quit IRC | 05:49 | |
| *** jasdeepH has quit IRC | 05:51 | |
| *** kaushikc has joined #openstack-dev | 05:54 | |
| *** paragan has quit IRC | 05:54 | |
| *** kaushikc has quit IRC | 05:56 | |
| *** mohits_ has joined #openstack-dev | 05:58 | |
| *** tdruiva has quit IRC | 05:59 | |
| *** danwent has quit IRC | 05:59 | |
| *** achampion has joined #openstack-dev | 05:59 | |
| *** tdruiva has joined #openstack-dev | 05:59 | |
| *** d34dh0r53 has quit IRC | 06:00 | |
| *** teran has quit IRC | 06:03 | |
| *** tdruiva has quit IRC | 06:04 | |
| *** jhesketh has quit IRC | 06:05 | |
| *** jeffrey4l has joined #openstack-dev | 06:14 | |
| *** Mandell has joined #openstack-dev | 06:16 | |
| *** odyssey4me3 has quit IRC | 06:21 | |
| *** loq_mac has joined #openstack-dev | 06:23 | |
| *** egallen has joined #openstack-dev | 06:26 | |
| *** alop has joined #openstack-dev | 06:28 | |
| ekarlso | yo, jd__ ping | 06:30 |
| *** alex_xu has joined #openstack-dev | 06:32 | |
| *** loquacities has joined #openstack-dev | 06:36 | |
| *** loq_mac has quit IRC | 06:36 | |
| *** dcahill has joined #openstack-dev | 06:38 | |
| *** teran has joined #openstack-dev | 06:38 | |
| *** jasdeepH has joined #openstack-dev | 06:39 | |
| *** loquacities has quit IRC | 06:40 | |
| *** loq_mac has joined #openstack-dev | 06:40 | |
| *** paragan has joined #openstack-dev | 06:40 | |
| *** paragan has quit IRC | 06:40 | |
| *** paragan has joined #openstack-dev | 06:40 | |
| *** neeti has joined #openstack-dev | 06:43 | |
| *** asalkeld has joined #openstack-dev | 06:43 | |
| *** michchap has quit IRC | 06:43 | |
| *** asalkeld is now known as angus_presenting | 06:43 | |
| *** michchap has joined #openstack-dev | 06:44 | |
| *** adjohn has quit IRC | 06:44 | |
| *** adjohn_ has joined #openstack-dev | 06:44 | |
| *** macjack has quit IRC | 06:46 | |
| *** rushiagr has joined #openstack-dev | 06:47 | |
| *** nosnos has quit IRC | 06:48 | |
| *** michchap has quit IRC | 06:48 | |
| *** pablosan has quit IRC | 06:48 | |
| *** nosnos has joined #openstack-dev | 06:48 | |
| *** chandankumar_ has joined #openstack-dev | 06:50 | |
| *** steven-weston_ has joined #openstack-dev | 06:50 | |
| *** paragan has quit IRC | 06:50 | |
| *** avishayb has joined #openstack-dev | 06:51 | |
| *** chandankumar has quit IRC | 06:51 | |
| *** jeffrey4l has quit IRC | 06:52 | |
| *** amotoki has joined #openstack-dev | 06:53 | |
| *** angus_presenting has quit IRC | 06:54 | |
| *** dstanek has quit IRC | 06:54 | |
| *** loq_mac has quit IRC | 06:56 | |
| *** adjohn_ has quit IRC | 06:59 | |
| *** reidrac has joined #openstack-dev | 06:59 | |
| *** SergeyLukjanov has quit IRC | 06:59 | |
| *** steven-weston_ has quit IRC | 07:00 | |
| *** macjack has joined #openstack-dev | 07:01 | |
| *** KeithSharp has quit IRC | 07:02 | |
| *** vartom11111 has joined #openstack-dev | 07:05 | |
| *** jeffrey4l has joined #openstack-dev | 07:05 | |
| *** yolanda has joined #openstack-dev | 07:08 | |
| *** ngoracke has quit IRC | 07:08 | |
| *** reidrac has quit IRC | 07:09 | |
| *** mohits_ has quit IRC | 07:12 | |
| *** mrunge has joined #openstack-dev | 07:13 | |
| *** jeffrey4l_ has joined #openstack-dev | 07:14 | |
| *** jcoufal has joined #openstack-dev | 07:14 | |
| *** eglynn has joined #openstack-dev | 07:15 | |
| *** jeffrey4l has quit IRC | 07:16 | |
| *** radix_ has joined #openstack-dev | 07:17 | |
| *** jasdeepH has quit IRC | 07:21 | |
| *** mosulica has joined #openstack-dev | 07:23 | |
| *** avishay has joined #openstack-dev | 07:23 | |
| *** steven-weston has quit IRC | 07:24 | |
| *** steven-weston has joined #openstack-dev | 07:24 | |
| *** evgenyf has joined #openstack-dev | 07:24 | |
| *** vartom11111 has quit IRC | 07:26 | |
| *** nimi has quit IRC | 07:27 | |
| *** eglynn has quit IRC | 07:27 | |
| *** denis_makogon_ has joined #openstack-dev | 07:28 | |
| *** vartom11111 has joined #openstack-dev | 07:30 | |
| *** alop has quit IRC | 07:32 | |
| *** MaxV has joined #openstack-dev | 07:33 | |
| *** gimps has quit IRC | 07:33 | |
| *** gimps has joined #openstack-dev | 07:34 | |
| *** abk has joined #openstack-dev | 07:35 | |
| *** amuller has joined #openstack-dev | 07:38 | |
| *** mindpixel has joined #openstack-dev | 07:38 | |
| *** nati_ueno has quit IRC | 07:40 | |
| *** MaxV has quit IRC | 07:42 | |
| *** nouser1 has joined #openstack-dev | 07:42 | |
| *** nati_ueno has joined #openstack-dev | 07:43 | |
| *** nouser has quit IRC | 07:43 | |
| *** garyk has joined #openstack-dev | 07:45 | |
| *** dstanek has joined #openstack-dev | 07:46 | |
| *** mihgen has quit IRC | 07:49 | |
| *** urulama has joined #openstack-dev | 07:50 | |
| *** garyk has quit IRC | 07:51 | |
| *** dstanek has quit IRC | 07:51 | |
| *** xga_ has joined #openstack-dev | 07:51 | |
| *** xga has joined #openstack-dev | 07:51 | |
| *** xqueralt has joined #openstack-dev | 07:52 | |
| *** jprovazn has joined #openstack-dev | 07:52 | |
| *** hdd has joined #openstack-dev | 07:52 | |
| *** xga_ has quit IRC | 07:54 | |
| *** salv-orlando has joined #openstack-dev | 07:55 | |
| *** e0ne has joined #openstack-dev | 07:55 | |
| *** xga has quit IRC | 07:56 | |
| *** paragan has joined #openstack-dev | 07:56 | |
| *** paragan has quit IRC | 07:56 | |
| *** paragan has joined #openstack-dev | 07:56 | |
| *** rpodolyaka has left #openstack-dev | 07:57 | |
| *** xqueralt has quit IRC | 07:57 | |
| *** buddha_dai has quit IRC | 07:58 | |
| *** mohits_ has joined #openstack-dev | 07:58 | |
| *** urulama has quit IRC | 08:01 | |
| *** pmathews has quit IRC | 08:01 | |
| *** tkammer has joined #openstack-dev | 08:01 | |
| *** tkammer has quit IRC | 08:02 | |
| *** lincai has joined #openstack-dev | 08:02 | |
| *** tkammer has joined #openstack-dev | 08:02 | |
| *** urulama has joined #openstack-dev | 08:03 | |
| *** jtomasek has joined #openstack-dev | 08:05 | |
| *** tonyfy has joined #openstack-dev | 08:07 | |
| *** Mandell has quit IRC | 08:07 | |
| *** reidrac has joined #openstack-dev | 08:07 | |
| *** anniec has joined #openstack-dev | 08:09 | |
| *** anniec has quit IRC | 08:09 | |
| *** fifieldt has quit IRC | 08:12 | |
| *** buzztroll has quit IRC | 08:17 | |
| *** buzztroll has joined #openstack-dev | 08:17 | |
| ttx | notmyname: that would be my robot self applying http://lists.openstack.org/pipermail/openstack-dev/2013-June/010614.html | 08:17 |
| ttx | notmyname: it sets series according to what you put in milestone (and removes prio for non-milestone-targeted stuff). | 08:18 |
| ttx | notmyname: I can remove you from the list if you don't want to be autocleaned that way | 08:19 |
| *** buzztrol_ has joined #openstack-dev | 08:19 | |
| *** buzztroll has quit IRC | 08:19 | |
| *** SergeyLukjanov has joined #openstack-dev | 08:21 | |
| *** flaper87|afk is now known as flaper87 | 08:21 | |
| *** romcheg has joined #openstack-dev | 08:22 | |
| *** bauzas has joined #openstack-dev | 08:23 | |
| *** kumaranvram has quit IRC | 08:23 | |
| *** bauzas has quit IRC | 08:24 | |
| *** bauzas has joined #openstack-dev | 08:24 | |
| *** vills has joined #openstack-dev | 08:25 | |
| *** kumaranvram has joined #openstack-dev | 08:25 | |
| *** teran has quit IRC | 08:29 | |
| *** teran has joined #openstack-dev | 08:29 | |
| *** kumaranvram has quit IRC | 08:30 | |
| *** nosnos_ has joined #openstack-dev | 08:30 | |
| *** mihgen has joined #openstack-dev | 08:30 | |
| *** jgallard has joined #openstack-dev | 08:31 | |
| *** nosnos has quit IRC | 08:33 | |
| *** teran has quit IRC | 08:34 | |
| *** corXi has joined #openstack-dev | 08:36 | |
| *** macjack has quit IRC | 08:36 | |
| *** aditirav has quit IRC | 08:37 | |
| *** aditirav has joined #openstack-dev | 08:37 | |
| *** gilliard has quit IRC | 08:38 | |
| *** akrivoka has joined #openstack-dev | 08:40 | |
| *** faramir1 has quit IRC | 08:41 | |
| *** boris-42 has quit IRC | 08:41 | |
| *** mkerrin has joined #openstack-dev | 08:42 | |
| *** macjack has joined #openstack-dev | 08:44 | |
| *** MaxV has joined #openstack-dev | 08:44 | |
| *** pixelb has joined #openstack-dev | 08:45 | |
| *** jtomasek has quit IRC | 08:45 | |
| *** gilliard has joined #openstack-dev | 08:46 | |
| *** ruhe has joined #openstack-dev | 08:47 | |
| *** jtomasek has joined #openstack-dev | 08:47 | |
| *** jistr has joined #openstack-dev | 08:48 | |
| *** larsbutler has joined #openstack-dev | 08:48 | |
| *** omachace has joined #openstack-dev | 08:49 | |
| *** eglynn has joined #openstack-dev | 08:51 | |
| *** mmagr has joined #openstack-dev | 08:52 | |
| *** yassine has joined #openstack-dev | 08:52 | |
| *** rossella_s has joined #openstack-dev | 08:55 | |
| *** nouser1 has quit IRC | 08:56 | |
| *** kpavel has joined #openstack-dev | 08:57 | |
| *** buzztrol_ has quit IRC | 08:58 | |
| *** buzztroll has joined #openstack-dev | 08:58 | |
| *** derekh has joined #openstack-dev | 08:58 | |
| *** sushils has joined #openstack-dev | 08:59 | |
| *** tonyfy has quit IRC | 08:59 | |
| *** lucasagomes has joined #openstack-dev | 08:59 | |
| *** gilliard has quit IRC | 09:01 | |
| *** sumansn_ has joined #openstack-dev | 09:01 | |
| *** buzztroll has quit IRC | 09:02 | |
| *** boden has joined #openstack-dev | 09:02 | |
| *** sumanthns has quit IRC | 09:03 | |
| *** pschaef has joined #openstack-dev | 09:06 | |
| *** Ng has joined #openstack-dev | 09:06 | |
| *** Ng has quit IRC | 09:06 | |
| *** Ng has joined #openstack-dev | 09:07 | |
| *** belmoreira has joined #openstack-dev | 09:08 | |
| *** claxton has joined #openstack-dev | 09:09 | |
| *** iartarisi has joined #openstack-dev | 09:09 | |
| *** e0ne has quit IRC | 09:12 | |
| *** jeffrey4l_ has quit IRC | 09:14 | |
| *** yamahata_ has joined #openstack-dev | 09:14 | |
| *** AnilV4 has joined #openstack-dev | 09:15 | |
| *** garyk has joined #openstack-dev | 09:17 | |
| *** dosaboy has joined #openstack-dev | 09:17 | |
| *** jpich has joined #openstack-dev | 09:18 | |
| *** tdruiva has joined #openstack-dev | 09:18 | |
| *** dosaboy has quit IRC | 09:19 | |
| *** dosaboy has joined #openstack-dev | 09:19 | |
| *** ruhe has quit IRC | 09:22 | |
| *** skraynev has joined #openstack-dev | 09:23 | |
| *** ljjjustin has joined #openstack-dev | 09:24 | |
| *** jeffrey4l_ has joined #openstack-dev | 09:31 | |
| *** giulivo has joined #openstack-dev | 09:35 | |
| *** belmoreira has quit IRC | 09:37 | |
| *** teran has joined #openstack-dev | 09:37 | |
| *** xqueralt has joined #openstack-dev | 09:37 | |
| *** belmoreira has joined #openstack-dev | 09:39 | |
| *** yamahata_ has quit IRC | 09:40 | |
| *** avishayb has quit IRC | 09:40 | |
| *** vills has quit IRC | 09:42 | |
| *** fbo_away is now known as fbo | 09:42 | |
| *** vills has joined #openstack-dev | 09:43 | |
| *** vills has quit IRC | 09:43 | |
| *** dosaboy has quit IRC | 09:43 | |
| *** vills has joined #openstack-dev | 09:43 | |
| *** dosaboy has joined #openstack-dev | 09:43 | |
| *** athomas has joined #openstack-dev | 09:43 | |
| *** avishay has quit IRC | 09:43 | |
| *** lincai has quit IRC | 09:44 | |
| *** johnthetubaguy has joined #openstack-dev | 09:44 | |
| *** obondarev has joined #openstack-dev | 09:44 | |
| *** rods has joined #openstack-dev | 09:45 | |
| *** avishayb has joined #openstack-dev | 09:46 | |
| *** shang has joined #openstack-dev | 09:48 | |
| *** boris-42 has joined #openstack-dev | 09:48 | |
| *** jtomasek has quit IRC | 09:51 | |
| *** boris-42 has quit IRC | 09:52 | |
| *** martyntaylor has joined #openstack-dev | 09:53 | |
| *** rohitk has joined #openstack-dev | 09:54 | |
| *** shang has quit IRC | 09:54 | |
| *** mkerrin has quit IRC | 09:55 | |
| *** nati_ueno has quit IRC | 09:55 | |
| *** jtomasek has joined #openstack-dev | 09:55 | |
| *** aditirav has quit IRC | 09:56 | |
| *** aditirav has joined #openstack-dev | 09:56 | |
| *** ruhe has joined #openstack-dev | 09:58 | |
| *** buzztroll has joined #openstack-dev | 09:59 | |
| *** SergeyLukjanov is now known as _SergeyLukjanov | 09:59 | |
| *** ndipanov_gone is now known as ndipanov | 10:00 | |
| *** e0ne has joined #openstack-dev | 10:01 | |
| *** danpb has joined #openstack-dev | 10:02 | |
| *** buzztroll has quit IRC | 10:03 | |
| *** xchu has quit IRC | 10:08 | |
| *** jcoufal has quit IRC | 10:09 | |
| *** alex_xu has quit IRC | 10:09 | |
| *** odyssey4me3 has joined #openstack-dev | 10:10 | |
| *** shang has joined #openstack-dev | 10:11 | |
| *** paragan has quit IRC | 10:11 | |
| *** afazekas has joined #openstack-dev | 10:12 | |
| *** rods has quit IRC | 10:12 | |
| *** yatin has joined #openstack-dev | 10:15 | |
| *** ChrisSWS has joined #openstack-dev | 10:16 | |
| dosaboy | dhellman: ping | 10:17 |
| *** bingbu has quit IRC | 10:19 | |
| *** yamahata_ has joined #openstack-dev | 10:21 | |
| *** odyssey4me3 has quit IRC | 10:21 | |
| *** sahid has joined #openstack-dev | 10:21 | |
| *** paragan has joined #openstack-dev | 10:23 | |
| *** rods has joined #openstack-dev | 10:27 | |
| *** shang has quit IRC | 10:29 | |
| *** omachace has quit IRC | 10:29 | |
| *** ChrisSWS has left #openstack-dev | 10:29 | |
| *** ChrisSWS has quit IRC | 10:29 | |
| *** ljjjustin has quit IRC | 10:29 | |
| *** arata has joined #openstack-dev | 10:31 | |
| *** arata has left #openstack-dev | 10:31 | |
| *** odyssey4me3 has joined #openstack-dev | 10:31 | |
| *** omachace has joined #openstack-dev | 10:31 | |
| *** ruhe has quit IRC | 10:31 | |
| *** guohliu has quit IRC | 10:32 | |
| *** yamahata_ has quit IRC | 10:33 | |
| *** vartom11112 has joined #openstack-dev | 10:33 | |
| *** vartom11111 has quit IRC | 10:33 | |
| *** ygbo has joined #openstack-dev | 10:36 | |
| *** tdruiva has quit IRC | 10:36 | |
| *** abk has quit IRC | 10:36 | |
| *** avishayb has quit IRC | 10:36 | |
| *** safchain has joined #openstack-dev | 10:36 | |
| *** tdruiva has joined #openstack-dev | 10:37 | |
| *** abk has joined #openstack-dev | 10:37 | |
| *** nati_ueno has joined #openstack-dev | 10:37 | |
| *** martyntaylor has quit IRC | 10:39 | |
| *** ruhe has joined #openstack-dev | 10:41 | |
| *** tdruiva has quit IRC | 10:41 | |
| *** evgenyf has quit IRC | 10:44 | |
| *** ahale_ has quit IRC | 10:45 | |
| *** e0ne_ has joined #openstack-dev | 10:46 | |
| *** jp_at_hp has joined #openstack-dev | 10:47 | |
| *** e0ne has quit IRC | 10:48 | |
| *** martyntaylor has joined #openstack-dev | 10:49 | |
| *** renlt has quit IRC | 10:52 | |
| *** vills has quit IRC | 10:52 | |
| *** mihgen has quit IRC | 10:52 | |
| *** rongze has quit IRC | 10:53 | |
| *** gszasz has joined #openstack-dev | 10:53 | |
| *** rongze has joined #openstack-dev | 10:53 | |
| *** mihgen has joined #openstack-dev | 10:56 | |
| *** prekarat has quit IRC | 10:56 | |
| *** DinaBelova has joined #openstack-dev | 10:56 | |
| *** prekarat has joined #openstack-dev | 10:56 | |
| *** sahid has quit IRC | 10:59 | |
| *** rongze has quit IRC | 10:59 | |
| *** SergeyLukjanov has joined #openstack-dev | 10:59 | |
| *** buzztroll has joined #openstack-dev | 10:59 | |
| *** aditirav has quit IRC | 11:01 | |
| *** jhesketh__ has quit IRC | 11:02 | |
| *** buzztroll has quit IRC | 11:04 | |
| *** sahid has joined #openstack-dev | 11:04 | |
| *** avishay has joined #openstack-dev | 11:07 | |
| *** arata has joined #openstack-dev | 11:09 | |
| *** arata has quit IRC | 11:09 | |
| *** claxton has quit IRC | 11:10 | |
| *** claxton has joined #openstack-dev | 11:10 | |
| *** aditirav has joined #openstack-dev | 11:10 | |
| *** adalbas has joined #openstack-dev | 11:11 | |
| *** sergmelikyan has joined #openstack-dev | 11:11 | |
| *** odyssey4me3 has quit IRC | 11:11 | |
| *** ArxCruz has joined #openstack-dev | 11:12 | |
| *** sahid has quit IRC | 11:13 | |
| *** jhesketh__ has joined #openstack-dev | 11:14 | |
| jamezpolley | Word of warning about pip --upgrade - it doesn't just upgrade the package you specified, it will recursively upgrade all dependencies as well | 11:15 |
| jamezpolley | if you want a controlled upgrade of packagename to the new version 1.2.3 you can use "pip install packagename==1.2.3" | 11:15 |
| *** avishayb has joined #openstack-dev | 11:16 | |
| jamezpolley | but on the other hand, if your problem is lots of older libs hanging around, it's quite likely that --upgrade will do what you want :) | 11:16 |
| *** teran_ has joined #openstack-dev | 11:17 | |
| *** teran has quit IRC | 11:17 | |
| *** ahale has joined #openstack-dev | 11:18 | |
| *** buzztroll has joined #openstack-dev | 11:19 | |
| *** nosnos_ has quit IRC | 11:20 | |
| *** nosnos has joined #openstack-dev | 11:21 | |
| *** buzztroll has quit IRC | 11:23 | |
| *** kushal has quit IRC | 11:24 | |
| *** evgenyf has joined #openstack-dev | 11:25 | |
| *** nosnos has quit IRC | 11:25 | |
| *** ahale has quit IRC | 11:31 | |
| *** odyssey4me3 has joined #openstack-dev | 11:33 | |
| *** yatin has left #openstack-dev | 11:37 | |
| *** arosen has joined #openstack-dev | 11:37 | |
| *** mihgen has quit IRC | 11:37 | |
| *** ahale has joined #openstack-dev | 11:40 | |
| *** loq_mac has joined #openstack-dev | 11:40 | |
| *** ahale has quit IRC | 11:40 | |
| *** mihgen has joined #openstack-dev | 11:42 | |
| *** neeti has quit IRC | 11:44 | |
| *** boris-42 has joined #openstack-dev | 11:45 | |
| *** thelorax123 has quit IRC | 11:46 | |
| *** neeti has joined #openstack-dev | 11:46 | |
| *** ericw has joined #openstack-dev | 11:47 | |
| *** thelorax123 has joined #openstack-dev | 11:47 | |
| *** loq_mac has quit IRC | 11:48 | |
| *** odyssey4me3 has quit IRC | 11:48 | |
| *** rfolco has joined #openstack-dev | 11:49 | |
| *** ericw has quit IRC | 11:49 | |
| *** yamahata_ has joined #openstack-dev | 11:50 | |
| *** tdruiva has joined #openstack-dev | 11:53 | |
| *** MaxV has quit IRC | 11:54 | |
| *** teran_ has quit IRC | 11:54 | |
| *** MaxV has joined #openstack-dev | 11:54 | |
| *** jgallard has quit IRC | 11:55 | |
| *** urulama has quit IRC | 11:57 | |
| *** yaguang has quit IRC | 11:58 | |
| *** MaxV has quit IRC | 11:59 | |
| *** evgenyf has quit IRC | 11:59 | |
| *** xingchao has quit IRC | 11:59 | |
| *** urulama has joined #openstack-dev | 12:00 | |
| *** avishayb has quit IRC | 12:02 | |
| *** avishay has quit IRC | 12:02 | |
| *** urulama_ has joined #openstack-dev | 12:03 | |
| *** teran has joined #openstack-dev | 12:04 | |
| *** urulama has quit IRC | 12:06 | |
| *** salv-orlando has quit IRC | 12:08 | |
| *** rfolco has quit IRC | 12:08 | |
| *** pschaef has quit IRC | 12:10 | |
| *** urulama_ has quit IRC | 12:10 | |
| *** urulama has joined #openstack-dev | 12:10 | |
| *** paragan has quit IRC | 12:11 | |
| *** urulama has quit IRC | 12:13 | |
| *** ericw has joined #openstack-dev | 12:13 | |
| *** claxton has quit IRC | 12:14 | |
| *** rfolco has joined #openstack-dev | 12:14 | |
| *** ruhe has quit IRC | 12:15 | |
| *** Shaan7 has joined #openstack-dev | 12:15 | |
| *** salv-orlando has joined #openstack-dev | 12:16 | |
| *** morazi has joined #openstack-dev | 12:16 | |
| *** urulama has joined #openstack-dev | 12:17 | |
| *** vills_ has joined #openstack-dev | 12:18 | |
| *** nati_ueno has quit IRC | 12:18 | |
| *** buzztroll has joined #openstack-dev | 12:20 | |
| *** erkules_ is now known as erkules | 12:21 | |
| *** urulama has quit IRC | 12:22 | |
| *** gimps has quit IRC | 12:22 | |
| *** jprovazn has quit IRC | 12:22 | |
| *** urulama has joined #openstack-dev | 12:23 | |
| *** ruhe has joined #openstack-dev | 12:26 | |
| *** MaxV has joined #openstack-dev | 12:26 | |
| *** martyntaylor1 has joined #openstack-dev | 12:26 | |
| *** sahid has joined #openstack-dev | 12:27 | |
| *** urulama has quit IRC | 12:27 | |
| *** martyntaylor has quit IRC | 12:27 | |
| *** michchap has joined #openstack-dev | 12:28 | |
| *** urulama has joined #openstack-dev | 12:29 | |
| *** urulama has quit IRC | 12:30 | |
| *** larsbutler has quit IRC | 12:30 | |
| *** rohitk has quit IRC | 12:33 | |
| *** vkmc has joined #openstack-dev | 12:37 | |
| *** amotoki has quit IRC | 12:40 | |
| *** urulama has joined #openstack-dev | 12:40 | |
| *** jprovazn has joined #openstack-dev | 12:42 | |
| *** lucasagomes is now known as lucas-hungry | 12:44 | |
| *** miziel_r has joined #openstack-dev | 12:44 | |
| *** max_lobur_afk is now known as max_lobur | 12:46 | |
| *** prekarat has quit IRC | 12:46 | |
| *** pcm_ has joined #openstack-dev | 12:47 | |
| *** Akrog has joined #openstack-dev | 12:48 | |
| *** prekarat has joined #openstack-dev | 12:50 | |
| *** rongze has joined #openstack-dev | 12:52 | |
| *** urulama_ has joined #openstack-dev | 12:52 | |
| *** buzztroll has quit IRC | 12:53 | |
| *** mihgen has quit IRC | 12:54 | |
| *** rongze_ has joined #openstack-dev | 12:54 | |
| *** rushiagr has quit IRC | 12:55 | |
| *** aditirav_ has joined #openstack-dev | 12:55 | |
| *** akrivoka has quit IRC | 12:55 | |
| *** aditirav_ has quit IRC | 12:55 | |
| *** mohits_ has quit IRC | 12:56 | |
| *** urulama has quit IRC | 12:56 | |
| *** FunnyLookinHat has joined #openstack-dev | 12:56 | |
| *** rushiagr has joined #openstack-dev | 12:56 | |
| *** urulama_ has quit IRC | 12:57 | |
| *** Shaan7 has quit IRC | 12:57 | |
| *** urulama has joined #openstack-dev | 12:57 | |
| *** rongze has quit IRC | 12:58 | |
| *** MaxV has quit IRC | 12:58 | |
| *** e0ne_ has quit IRC | 12:58 | |
| *** aditirav has quit IRC | 12:59 | |
| *** martyntaylor1 has quit IRC | 13:01 | |
| *** Alexei_987 has joined #openstack-dev | 13:02 | |
| *** martyntaylor has joined #openstack-dev | 13:02 | |
| *** rongze_ has quit IRC | 13:02 | |
| *** martyntaylor has left #openstack-dev | 13:03 | |
| *** akrivoka has joined #openstack-dev | 13:03 | |
| *** rongze has joined #openstack-dev | 13:04 | |
| *** evgenyf has joined #openstack-dev | 13:06 | |
| *** skraynev has quit IRC | 13:06 | |
| *** vills_ has quit IRC | 13:08 | |
| *** kevinconway has joined #openstack-dev | 13:08 | |
| *** teran_ has joined #openstack-dev | 13:10 | |
| *** skraynev has joined #openstack-dev | 13:10 | |
| *** teran__ has joined #openstack-dev | 13:11 | |
| *** ngoracke has joined #openstack-dev | 13:12 | |
| *** ruhe has quit IRC | 13:12 | |
| *** xqueralt has quit IRC | 13:13 | |
| *** markvoelker has quit IRC | 13:13 | |
| *** markvoelker has joined #openstack-dev | 13:14 | |
| *** teran has quit IRC | 13:14 | |
| *** sgordon has joined #openstack-dev | 13:15 | |
| *** sgordon has joined #openstack-dev | 13:15 | |
| *** teran_ has quit IRC | 13:15 | |
| *** ruhe has joined #openstack-dev | 13:17 | |
| *** mriedem has joined #openstack-dev | 13:18 | |
| *** dstanek has joined #openstack-dev | 13:18 | |
| dstanek | shardy: ping | 13:18 |
| *** MaxV has joined #openstack-dev | 13:19 | |
| *** viktors1 has joined #openstack-dev | 13:19 | |
| *** bknudson has quit IRC | 13:21 | |
| *** gordc has joined #openstack-dev | 13:22 | |
| *** viktors1 has left #openstack-dev | 13:23 | |
| *** bswartz has quit IRC | 13:23 | |
| *** tkammer has quit IRC | 13:26 | |
| *** rdas has quit IRC | 13:26 | |
| *** thomasem has joined #openstack-dev | 13:27 | |
| *** e0ne has joined #openstack-dev | 13:29 | |
| *** e0ne_ has joined #openstack-dev | 13:31 | |
| *** nermina has joined #openstack-dev | 13:32 | |
| *** ekhugen has joined #openstack-dev | 13:32 | |
| *** xga has joined #openstack-dev | 13:33 | |
| *** e0ne has quit IRC | 13:33 | |
| *** e0ne has joined #openstack-dev | 13:33 | |
| *** mihgen has joined #openstack-dev | 13:35 | |
| *** miziel_r has quit IRC | 13:35 | |
| *** e0ne_ has quit IRC | 13:37 | |
| *** ekhugen has quit IRC | 13:37 | |
| *** neeti has quit IRC | 13:37 | |
| *** e0ne has quit IRC | 13:37 | |
| *** CaptTofu has quit IRC | 13:39 | |
| *** CaptTofu has joined #openstack-dev | 13:39 | |
| *** dbalog has joined #openstack-dev | 13:40 | |
| *** sandywalsh has quit IRC | 13:40 | |
| *** chandankumar_ has quit IRC | 13:40 | |
| *** marios has quit IRC | 13:41 | |
| *** dave-mcnally has joined #openstack-dev | 13:41 | |
| *** marios has joined #openstack-dev | 13:41 | |
| *** RajM has joined #openstack-dev | 13:43 | |
| *** jergerber has joined #openstack-dev | 13:45 | |
| *** achampion has quit IRC | 13:45 | |
| *** xga_ has joined #openstack-dev | 13:46 | |
| *** dsantos_ has quit IRC | 13:46 | |
| *** dsantos_ has joined #openstack-dev | 13:46 | |
| *** noslzzp has joined #openstack-dev | 13:46 | |
| *** avishayb has joined #openstack-dev | 13:46 | |
| *** bknudson has joined #openstack-dev | 13:47 | |
| *** lucas-hungry is now known as lucasagomes | 13:48 | |
| *** glenng has joined #openstack-dev | 13:48 | |
| *** e0ne has joined #openstack-dev | 13:48 | |
| *** xga has quit IRC | 13:48 | |
| mriedem | so...i've deleted a ML thread that i didn't care about at the time but now want to comment on, does anyone have a trick for getting that back so i can follow up on the thread to continue the history? | 13:48 |
| *** vladikr has joined #openstack-dev | 13:49 | |
| *** buzztroll has joined #openstack-dev | 13:50 | |
| *** thelorax123 has quit IRC | 13:50 | |
| *** zul has quit IRC | 13:50 | |
| *** zul has joined #openstack-dev | 13:51 | |
| *** mindpixel has quit IRC | 13:51 | |
| *** thelorax123 has joined #openstack-dev | 13:51 | |
| *** jergerber has quit IRC | 13:52 | |
| *** burt1 has joined #openstack-dev | 13:52 | |
| *** terriyu has joined #openstack-dev | 13:52 | |
| *** dprince has joined #openstack-dev | 13:53 | |
| *** ayoung-zZzZzZ has quit IRC | 13:53 | |
| *** sandywalsh has joined #openstack-dev | 13:53 | |
| *** ilyashakhat has quit IRC | 13:55 | |
| *** martyntaylor has joined #openstack-dev | 13:55 | |
| *** buzztroll has quit IRC | 13:55 | |
| *** jergerber has joined #openstack-dev | 13:55 | |
| *** rushiagr has quit IRC | 13:56 | |
| *** glenng1 has joined #openstack-dev | 13:57 | |
| *** nermina has quit IRC | 13:57 | |
| *** yamahata_ has quit IRC | 13:57 | |
| *** lbragstad has quit IRC | 13:57 | |
| *** changbl has quit IRC | 13:57 | |
| *** arezmerita has quit IRC | 13:57 | |
| *** yamahata_ has joined #openstack-dev | 13:58 | |
| *** bswartz has joined #openstack-dev | 13:58 | |
| *** mrunge has quit IRC | 13:59 | |
| *** alunduil has quit IRC | 13:59 | |
| *** bpokorny has quit IRC | 13:59 | |
| *** glenng has quit IRC | 14:00 | |
| *** e0ne_ has joined #openstack-dev | 14:00 | |
| *** samuelbercovici has joined #openstack-dev | 14:00 | |
| *** bswartz1 has joined #openstack-dev | 14:01 | |
| *** gimps has joined #openstack-dev | 14:01 | |
| *** e0ne has quit IRC | 14:02 | |
| *** bswartz has quit IRC | 14:02 | |
| *** bswartz1 is now known as bswartz | 14:02 | |
| *** dkranz has quit IRC | 14:03 | |
| dstanek | bknudson: i comment on that same review explaining the problem as i see it | 14:04 |
| *** yamahata_ has quit IRC | 14:07 | |
| *** briancurtin has joined #openstack-dev | 14:07 | |
| *** yamahata_ has joined #openstack-dev | 14:08 | |
| *** omachace has left #openstack-dev | 14:09 | |
| *** DinaBelova has quit IRC | 14:09 | |
| *** Shaan7 has joined #openstack-dev | 14:09 | |
| *** mfer has joined #openstack-dev | 14:10 | |
| *** herndon_ has joined #openstack-dev | 14:10 | |
| *** grapsus_ has quit IRC | 14:13 | |
| *** martyntaylor has quit IRC | 14:13 | |
| *** dolphm has joined #openstack-dev | 14:14 | |
| *** martyntaylor has joined #openstack-dev | 14:14 | |
| *** dvarga has joined #openstack-dev | 14:14 | |
| *** carl_baldwin has joined #openstack-dev | 14:14 | |
| *** carl_baldwin has quit IRC | 14:15 | |
| *** yaguang has joined #openstack-dev | 14:16 | |
| jd__ | sdague: hu what did you do on https://review.openstack.org/#/c/53889/3 ? | 14:16 |
| *** bpokorny has joined #openstack-dev | 14:16 | |
| *** carl_baldwin has joined #openstack-dev | 14:16 | |
| *** urulama has quit IRC | 14:17 | |
| *** carl_baldwin has quit IRC | 14:17 | |
| *** sthaha has quit IRC | 14:17 | |
| *** carl_baldwin has joined #openstack-dev | 14:18 | |
| sdague | jd__: kicked it out of the gate | 14:18 |
| *** martyntaylor has quit IRC | 14:18 | |
| *** vartom11112 has quit IRC | 14:18 | |
| sdague | just did an ammend on the commit message, which means it's a new version, so zuul pulls it from the gate queue | 14:18 |
| sdague | I'm trying to get stuff that's ahead of the actuall fixes out of there, so we can get back to normal faster | 14:19 |
| *** topol has joined #openstack-dev | 14:19 | |
| *** joesavak has joined #openstack-dev | 14:20 | |
| *** DinaBelova has joined #openstack-dev | 14:20 | |
| *** kbringard has joined #openstack-dev | 14:21 | |
| * jd__ gets down on his knees and screams WHYYYYYY | 14:21 | |
| jd__ | sdague: ok fair enough, but now the commit message looks weird :) | 14:22 |
| *** sagar_nikam has joined #openstack-dev | 14:22 | |
| mhu | Hi, looks like there's something funny going on with keystone, the test keystone.tests.test_keystoneclient.KcMasterTestCase.test_tokens_after_user_update_passwd keeps failing for everybody at the check | 14:22 |
| *** markmc has joined #openstack-dev | 14:22 | |
| *** martyntaylor has joined #openstack-dev | 14:22 | |
| skraynev | hello all | 14:23 |
| bknudson | mhu: should be fixed by reverting https://review.openstack.org/#/c/57583/ | 14:23 |
| topol | bknudson, do you know if there is something wrong with Jenkins. It is -1 my patch and some test cases are failing but all I did was update a keystone doc page | 14:24 |
| jd__ | sdague: bah it screwed the bug too by changing ownership; I wonder if next time it wouldn't be better to ask a core reviewer to -2 it temporarily | 14:24 |
| topol | what mhu said :-) | 14:24 |
| *** rongze_ has joined #openstack-dev | 14:24 | |
| mhu | topol, see above :) | 14:24 |
| skraynev | sdague: Hi. I be very happy, if you could spend a little bit time for answer on my comment https://review.openstack.org/#/c/52070/. | 14:25 |
| topol | great minds think alike | 14:25 |
| dstanek | topol: it was a bug in keystoneclient that I reverted | 14:25 |
| *** jruzicka has joined #openstack-dev | 14:25 | |
| danpb | does anyone know of any good docs describing requirements to make a python package that's installable via 'pip install' | 14:26 |
| topol | so if I try recheck no bug again it will now work? | 14:26 |
| mhu | dstanek, bknudson I'll wait for the revert to take effect then, thx | 14:26 |
| roaet | when devstack errors start producing 'Cannot allocate memory' errors, what is the proper solution other than recreating my instance? | 14:26 |
| shardy | dstanek: Hi, sorry you pinged? | 14:26 |
| bknudson | topol: the revert needs to merge first | 14:26 |
| shardy | dstanek: Sorry for the issues my patch is causing :( | 14:26 |
| bknudson | shardy: really not your fault. | 14:26 |
| *** lbragstad has joined #openstack-dev | 14:26 | |
| *** xqueralt has joined #openstack-dev | 14:27 | |
| dstanek | shardy: it's not a big deal; what bknudson said :-) | 14:27 |
| topol | OK, wasnt sure if it was merged or not. Last time I did recheck no bug was at 3am and then went to bed | 14:27 |
| *** nermina has joined #openstack-dev | 14:27 | |
| dstanek | shardy: i added a comment to the revert review about the problem | 14:27 |
| shardy | dstanek, bknudson: OK, cool :) | 14:27 |
| shardy | dstanek: Yeah just reading that | 14:27 |
| *** rongze has quit IRC | 14:27 | |
| topol | oh come a little a guilt. Hey shardy Im trying to go on thanksgiving vacation and the last thing I need to do was get my patch merged :-) | 14:28 |
| dstanek | shardy: i don't have much experience in that code so i can't really suggest a good fix right now | 14:28 |
| topol | I finally succumbed to all of gyees meticulous wording changes :-) | 14:29 |
| shardy | topol: Haha - To be fair, the patch had been up for nearly a month and nobody left any negative feedback | 14:30 |
| *** rongze has joined #openstack-dev | 14:30 | |
| shardy | dstanek: I think jamielennox has some ideas but has not left any comments on the review | 14:30 |
| *** yamahata_ has quit IRC | 14:30 | |
| shardy | Essentially having tokens stored in two places inside keystoneclient seems wrong and is bound to result in obscure bugs where you're not using the token you expect | 14:31 |
| dstanek | shardy: ok, if you need help let me know | 14:31 |
| *** yamahata_ has joined #openstack-dev | 14:31 | |
| topol | shardy why didnt jenkins catch it before it merged just curiously? | 14:31 |
| bknudson | shardy: it might actually be the test in keystone that's not doing the right thing | 14:32 |
| dstanek | topol: the test failures were in Keystone | 14:32 |
| shardy | topol: because the client integration tests are in the keystone tree, not tempest | 14:32 |
| dstanek | topol: keystone grabs client code from master and runs a few tests | 14:32 |
| bknudson | dstanek: it probably runs as many tests as there are in keystoneclient. | 14:32 |
| topol | shardy, gotcha. Has sdague shamed us yet??? :-) | 14:33 |
| *** epopt37 has quit IRC | 14:33 | |
| *** venkatesh has joined #openstack-dev | 14:33 | |
| *** dolphm has quit IRC | 14:33 | |
| shardy | topol: Well we had discussions with sdague re moving client tests into tempest, hopefully this reinforces why that is a good idea ;) | 14:33 |
| *** omachace has joined #openstack-dev | 14:33 | |
| *** rongze_ has quit IRC | 14:33 | |
| bknudson | topol: https://etherpad.openstack.org/p/icehouse-summit-qa-keystone | 14:33 |
| kpavel | dansmith: hi Dan, can you take a look again at https://review.openstack.org/#/c/46649/? | 14:34 |
| topol | shardy, ++ You make a convincing point :-) | 14:34 |
| sdague | shardy: :) | 14:34 |
| dstanek | maybe i'll spend time today learning about tempest :-) wouldn't mind submitting patches for keystone tests | 14:34 |
| *** rkukura has joined #openstack-dev | 14:34 | |
| bknudson | dstanek: shardy is the expert | 14:35 |
| bknudson | since he actually wrote a test for keystonclient in tempest | 14:35 |
| shardy | bknudson: Haha, I've not yet managed to get any tests merged into tempest at all ;) | 14:35 |
| bknudson | shardy: what's the review? | 14:35 |
| * shardy managed to stir up a big discussion tho | 14:36 | |
| sdague | a good one at that | 14:36 |
| topol | bknudson, any chance after dstanek's revert merges can you do a recheck no bug on https://review.openstack.org/#/c/56092/7 ? I really am going on vacation today :-) | 14:36 |
| shardy | bknudson: https://review.openstack.org/#/c/51558/ | 14:36 |
| bknudson | topol: yes, "someone" could go through and recheck all the recent keystone -1s. | 14:37 |
| shardy | bknudson: There are others too https://review.openstack.org/#/q/project:openstack/tempest+owner:shardy,n,z | 14:37 |
| bknudson | shardy: was there a question about where the tests go, or is the directory structure figured out? | 14:37 |
| bknudson | I missed the summit discussion | 14:38 |
| ttx | markmc: you around ? | 14:38 |
| topol | who submitted https://etherpad.openstack.org/p/icehouse-summit-qa-keystone ? Right now they look like a prophetic genius. I want them to pick my lottery numbers for this week :-) | 14:38 |
| bknudson | topol: we've known about this for some time... hasn't gotten to the top of anyone's work queue | 14:39 |
| *** peristeri has joined #openstack-dev | 14:39 | |
| topol | bknudson, K makes sense. | 14:39 |
| bknudson | and now we're getting complaints about key distribution... | 14:39 |
| shardy | bknudson: Planning to revive the OS-TRUST API surface test soon, when we resolve bug #1246941 | 14:40 |
| uvirtbot | Launchpad bug 1246941 in keystone "Trust Auth is broken when Content Type is XML" [Undecided,Invalid] https://launchpad.net/bugs/1246941 | 14:40 |
| markmc | ttx, yep? | 14:40 |
| *** neelashah has joined #openstack-dev | 14:40 | |
| topol | gotta run | 14:40 |
| ttx | markmc: almost done with rootwrap standalonification, but apparently the git rebase --committer-date-is-author-date did not really work (initial commit has wrong date) | 14:40 |
| *** jruzicka has quit IRC | 14:40 | |
| *** changbl has joined #openstack-dev | 14:40 | |
| ttx | markmc: it runs and applies and all, but CommitDate is still wrong | 14:40 |
| markmc | ttx, it doesn't take AuthorDate and stick it in CommitDate? | 14:41 |
| markmc | ttx, oh, you mean the initial commit | 14:41 |
| ttx | WIP result at https://github.com/ttx/oslo.rootwrap - 736568f81c23e73e24b8be1997e39f4ae006d774 has wrong date | 14:41 |
| ttx | markmc: apparently not | 14:41 |
| *** jecarey has joined #openstack-dev | 14:41 | |
| *** topol has quit IRC | 14:41 | |
| shardy | sdague: What was the final decision re location for client tests, just create more scenario tests? | 14:41 |
| markmc | ttx, well, rebase isn't editing the first commit | 14:42 |
| ttx | markmc: ah, that explains it | 14:42 |
| *** achampion has joined #openstack-dev | 14:42 | |
| *** AlanClark has joined #openstack-dev | 14:42 | |
| markmc | ttx, hmm, don't have a quick idea | 14:43 |
| markmc | ttx, maybe filter-branch | 14:43 |
| danpb | mordred: am I right in thinking that if we need to be 'pip install' compatible, we need to use setuptools and not distutils ? | 14:43 |
| ttx | markmc: should I just ignore that and be happy that at least /some/ history is preserved ? | 14:43 |
| sdague | shardy: I think for right now more scenario tests are fine, honestly, it's a travel week for me, so a bit scattered (and seeing if I can get this grenade fix into the gate) | 14:43 |
| *** marun has joined #openstack-dev | 14:43 | |
| markmc | ttx, yeah, it's not a big deal | 14:43 |
| mordred | danpb: yes | 14:43 |
| danpb | mordred: (context: i'm trying to setup libvirt python as a pip installable package) | 14:43 |
| mordred | danpb: w00t! | 14:43 |
| markmc | happy dance! :) | 14:43 |
| mordred | danpb: I recommend just using pbr | 14:43 |
| danpb | ok, guess i have to throw out my lovingly crafted distutils based setup.py then :-) | 14:43 |
| *** vijendar has joined #openstack-dev | 14:43 | |
| ttx | markmc: ok, will do a quite pass over filter-tree but if not just ignore error | 14:43 |
| mordred | danpb: but if you don't want to go that far, then yes, setuptools | 14:43 |
| danpb | what is pbr ? | 14:44 |
| shardy | sdague: Ok, thanks, I'll revive those expired reviews moved under scenario then | 14:44 |
| mordred | danpb: it's the thing we use in openstack that makes it so that you have a stub setup.py and everything else goes into setup.cfg :) | 14:44 |
| mordred | danpb: actually - does libvirt depend on other python packages? | 14:44 |
| danpb | no, none at all | 14:44 |
| danpb | and i'd like deps to be at a minimum | 14:45 |
| mordred | ok. then actually distutils is fine, tbh | 14:45 |
| mordred | you need to use setuptools if you need to communicate to pip that you have external depends | 14:45 |
| mordred | HOWEVER | 14:45 |
| danpb | is there anything special i need to do to make pip install happy then | 14:45 |
| mordred | use setuptools anyway | 14:45 |
| mordred | :) | 14:45 |
| mordred | nope. as long as python setup.py install works, pip should work with it | 14:45 |
| mordred | you can test locally by doing "virtualenv foo ; foo/bin/pip install ." | 14:46 |
| danpb | because 'pip install' currently fails bizarrely http://paste.fedoraproject.org/55712/44323138 | 14:46 |
| mordred | and "python setup.py sdist ; virtualenv foo ; foo/bin/pip install dist/*tar.gz" | 14:46 |
| danpb | even though 'python setup.py install' works | 14:46 |
| *** sergmelikyan has quit IRC | 14:46 | |
| mordred | danpb: hrm. I may be lying to you then - you got a tree somewhere I can pull and try? | 14:46 |
| *** epopt37 has joined #openstack-dev | 14:47 | |
| danpb | mordred: try this one git://github.com/cardoe/libvirt-python.git | 14:47 |
| * mordred is MORE than happy to directly help with this one | 14:47 | |
| *** schwicht has quit IRC | 14:47 | |
| mordred | danpb: oh wow. yeah - you should not move to pbr - you need to do many things I don't account for | 14:48 |
| danpb | the joy of native code interfaces | 14:49 |
| *** yamahata_ has quit IRC | 14:50 | |
| *** CaptTofu has quit IRC | 14:50 | |
| *** buzztroll has joined #openstack-dev | 14:50 | |
| *** ngoracke has quit IRC | 14:51 | |
| mordred | danpb: fyi - having pkg-config but not libvirt-dev|libvirt-devel fails withugly error | 14:51 |
| *** dkranz has joined #openstack-dev | 14:52 | |
| *** yamahata_ has joined #openstack-dev | 14:52 | |
| mordred | danpb: ok. two things | 14:52 |
| danpb | mordred: you get the ugly "distutils.errors.DistutilsExecError: command '/usr/bin/pkg-config' failed with exit status 1" message ? | 14:52 |
| *** jobewan has joined #openstack-dev | 14:52 | |
| mordred | danpb: yeah. and a traceback. that's fine, I don't really care, just thought I'd share | 14:53 |
| mordred | :) | 14:53 |
| mordred | danpb: http://paste.openstack.org/show/53743 | 14:53 |
| mordred | danpb: that should fix your pip problem | 14:54 |
| mordred | I get this though: /usr/bin/ld: cannot find -lvirt-lxc | 14:54 |
| danpb | what version of libvirt do you have ? | 14:54 |
| *** giulivo has quit IRC | 14:54 | |
| mordred | 1.1.1-0ubuntu8.1 | 14:54 |
| danpb | odd, that library was added in 1.0.2. i'll investigate that | 14:55 |
| danpb | perhaps ubuntu have it in a separate package | 14:55 |
| *** kpavel has quit IRC | 14:55 | |
| mordred | zul: ^^ ? | 14:55 |
| mordred | I have /usr/lib/libvirt-qemu.so | 14:55 |
| mordred | /usr/lib/libvirt.so | 14:56 |
| *** RajM has quit IRC | 14:56 | |
| *** raj_m has joined #openstack-dev | 14:56 | |
| zul | mordred: hmm? | 14:57 |
| danpb | mordred: you should have a libvirt-lxc.so too ! | 14:57 |
| mordred | zul: where is libvirt-lxc? | 14:57 |
| zul | morazi: should be there | 14:57 |
| mordred | I have /usr/lib/libvirt-lxc.so.0.1001.1 | 14:58 |
| *** venkatesh has quit IRC | 14:58 | |
| mordred | in libvirt0 | 14:58 |
| zul | mordred: is this from the cloud archive? which version | 14:58 |
| *** bswartz1 has joined #openstack-dev | 14:58 | |
| mordred | but I do not have the libvirt-lxc symlink in libvirt-dev | 14:58 |
| danpb | mordred: hmm, so removing that 'package_dir' directive means that 'python setup.py install' doesn't find the libvirt.py in the build/ directory | 14:58 |
| *** markmcclain has quit IRC | 14:58 | |
| danpb | so it only installs the .so part of the module :-( | 14:58 |
| mordred | danpb: hrm | 14:58 |
| mordred | ok | 14:58 |
| mordred | danpb: new idea... | 14:58 |
| *** venkatesh has joined #openstack-dev | 14:59 | |
| *** gargya has joined #openstack-dev | 14:59 | |
| *** glenng1 has quit IRC | 14:59 | |
| mordred | danpb: create a build dir | 15:00 |
| *** carlp has joined #openstack-dev | 15:00 | |
| *** glenng has joined #openstack-dev | 15:00 | |
| *** bswartz has quit IRC | 15:00 | |
| *** ftcjeff has joined #openstack-dev | 15:00 | |
| *** bswartz1 is now known as bswartz | 15:00 | |
| danpb | mordred: hehe, that might work | 15:01 |
| *** sumansn_ has quit IRC | 15:01 | |
| * danpb wonders what exact commands pip is actually running | 15:01 | |
| ttx | markmc: OK, I think I fixed it | 15:02 |
| *** FunnyLookinHat has quit IRC | 15:02 | |
| mordred | danpb: it should be running Complete output from command python setup.py egg_info: | 15:02 |
| mordred | gah | 15:02 |
| mordred | danpb: hrm that's interesting | 15:03 |
| mordred | dstufft: ^^ | 15:03 |
| markmc | ttx, cool, looks good - how did you do it ? | 15:03 |
| *** venkatesh has quit IRC | 15:03 | |
| *** grapsus_ has joined #openstack-dev | 15:03 | |
| danpb | mordred: ok, thanks for the suggestions. i think i've got enough to go an play some more and hopefully solve it | 15:03 |
| mordred | awesome! | 15:03 |
| ttx | markmc: filter-branch with GIT_COMMITTER_DATE=$GIT_AUTHOR_DATE git commit-tree "$@" | 15:04 |
| mordred | danpb: I am very excited about your work | 15:04 |
| markmc | ttx, ah, cool | 15:04 |
| mordred | zul: this is not from cloud-archive | 15:04 |
| mordred | zul: this is from stock saucy | 15:04 |
| *** ndipanov has quit IRC | 15:04 | |
| zul | mordred: cool ill have a look at it after this vuds session | 15:04 |
| *** kushal has joined #openstack-dev | 15:05 | |
| mordred | zul: cool. I _think_ this would clasify as "major packaging bug" | 15:05 |
| zul | mordred: i would agree | 15:05 |
| ttx | mordred, markmc: would welcome your sanity check on https://github.com/ttx/oslo.rootwrap -- in particular around the addition of required packaging files | 15:05 |
| *** giulivo has joined #openstack-dev | 15:05 | |
| *** cfriesen_ has joined #openstack-dev | 15:05 | |
| mordred | ttx: you don't need this line: https://github.com/ttx/oslo.rootwrap/blob/master/setup.cfg#L23 | 15:06 |
| ttx | mordred: then next step, I suppose, is a openstack/config change to add that project ? | 15:06 |
| *** ngoracke has joined #openstack-dev | 15:06 | |
| *** rcleere has joined #openstack-dev | 15:06 | |
| mordred | ttx: nor these lines: https://github.com/ttx/oslo.rootwrap/blob/master/setup.cfg#L27-L29 | 15:06 |
| * ttx blames oslo.config | 15:06 | |
| mordred | otherwise looks great | 15:07 |
| danpb | mordred: if i can get this working our aim is to release it with this month's libvirt release (eg about 1+1/2 week's time) | 15:07 |
| markmc | ttx, do you plan to expose any stable API beyond cmd.main() ? | 15:07 |
| mordred | ttx: yes. next step is openstack-infra/config change. | 15:07 |
| markmc | ttx, e.g. for people to write their own filters which subclass the ones in filters.py ? | 15:08 |
| mordred | danpb: w00t! | 15:08 |
| mordred | danpb: now - question - will python-libvirt build against older libvirts? | 15:08 |
| *** ftcjeff has quit IRC | 15:08 | |
| *** omachace has quit IRC | 15:08 | |
| markmc | ttx, are those classes public APIs, or implementation details of the rootwrap binary | 15:08 |
| ttx | markmc: not at this point. there were talks about moving the rootwrap_config wrapper in a lib that would get consumed by calling projects, but it could belong to processutils just the same | 15:08 |
| ttx | markmc: so those classes are all implementation details | 15:08 |
| markmc | ttx, ok, how to make that clear - e.g. use of __all__ or underscore prefixing the class names or ... | 15:09 |
| mordred | danpb: so, like, if I just have libvirt-devel installed from rhel6, would pip install python-libvirt theoretically work? | 15:09 |
| ttx | i.e. the public interface is through the config file and the shell arguments | 15:09 |
| *** sagar_nikam has quit IRC | 15:09 | |
| danpb | mordred: yes, absolutely - any version back to about 0.9.6 is our goal (ie current nova min required libvirt) | 15:09 |
| *** ruhe has quit IRC | 15:09 | |
| ttx | markmc: I'll have to give it a thought. before tagging the first release at least | 15:10 |
| mordred | w00t | 15:11 |
| mordred | danpb: I will owe you very many beers | 15:11 |
| *** rushiagr has joined #openstack-dev | 15:11 | |
| *** tkammer has joined #openstack-dev | 15:11 | |
| *** ndipanov has joined #openstack-dev | 15:11 | |
| ttx | markmc: there might be some random offenders left still calling those modules directly | 15:12 |
| ttx | like in (ewww..) https://github.com/openstack/neutron/tree/master/bin | 15:12 |
| *** carl_baldwin has quit IRC | 15:12 | |
| ttx | so we might need to kill those first | 15:12 |
| *** jasondotstar has joined #openstack-dev | 15:13 | |
| ttx | although that can just be handled when I get to switch neutron to the standalone rootwrap | 15:14 |
| *** ruhe has joined #openstack-dev | 15:16 | |
| *** kbrierly has joined #openstack-dev | 15:16 | |
| *** jruzicka has joined #openstack-dev | 15:17 | |
| ttx | markmc: not sure what would be the best way to make that clear. __all__ would not really prevent anything, and underscore prefixing would look a bit funny when modules are imported from cmd.py | 15:17 |
| *** DinaBelova has quit IRC | 15:17 | |
| *** iartarisi has quit IRC | 15:17 | |
| ttx | markmc: i'm open to suggestions :) | 15:17 |
| *** gargya has quit IRC | 15:18 | |
| *** iartarisi has joined #openstack-dev | 15:18 | |
| markmc | ttx, yeah, I dunno | 15:19 |
| *** ayoung has joined #openstack-dev | 15:19 | |
| *** jtomasek has quit IRC | 15:19 | |
| markmc | ttx, document it at least - do not import this library | 15:19 |
| ttx | markmc: yeah, will add to README | 15:19 |
| ttx | I plan to add a much more thorough README anyway | 15:19 |
| *** nati_ueno has joined #openstack-dev | 15:20 | |
| *** eharney has joined #openstack-dev | 15:20 | |
| ttx | mordred: about setup.cfg; I should keep packages=oslo and only remove the oslo.rootwrap line ? Or just remove the packages line altogether ? | 15:20 |
| *** yamahata_ has quit IRC | 15:21 | |
| *** cdub has joined #openstack-dev | 15:22 | |
| *** blamar has quit IRC | 15:22 | |
| *** teran__ has quit IRC | 15:23 | |
| *** joesavak has quit IRC | 15:23 | |
| *** teran has joined #openstack-dev | 15:23 | |
| *** tdruiva has quit IRC | 15:24 | |
| *** tdruiva has joined #openstack-dev | 15:24 | |
| *** buzztroll has quit IRC | 15:24 | |
| *** markmcclain has joined #openstack-dev | 15:25 | |
| ekarlso | ayoung: ping | 15:25 |
| ttx | mordred: nm, ignore me | 15:25 |
| ekarlso | is the serviceCatalog included in the V2 API when you do a POST on /tokens ? | 15:25 |
| ayoung | ekarlso, I don't ping anymore. I graduated | 15:25 |
| ayoung | ekarlso, yep | 15:26 |
| ekarlso | hah ayoung : p | 15:26 |
| ekarlso | what's the diff on the V2 / 3 catalog ? | 15:26 |
| ayoung | ekarlso, you mean what is the difference between the token formats? | 15:27 |
| *** jimfehlig has joined #openstack-dev | 15:27 | |
| ekarlso | between the returned thing | 15:27 |
| ayoung | ekarlso, the "returned thing" is the body of the token | 15:27 |
| *** dolphm has joined #openstack-dev | 15:27 | |
| ayoung | ekarlso, if you get a token, you get back the token as a blobn, as well as the data that makes up the body in JSON | 15:28 |
| ayoung | http://docs.openstack.org/developer/keystone/api_curl_examples.html#id3 | 15:28 |
| *** steven-weston has quit IRC | 15:28 | |
| *** tdruiva has quit IRC | 15:28 | |
| ayoung | ekarlso, same thing that gets returned when you validate a token | 15:28 |
| ayoung | ekarlso, and the V3 API is documented here: https://github.com/openstack/identity-api/blob/master/openstack-identity-api/v3/src/markdown/identity-api-v3.md#authentication-responses | 15:30 |
| ekarlso | ayoung: in the middleware is X-Service-Catalog always set for both v2.0 + v3 ? | 15:32 |
| ayoung | ekarlso, yep...but don't take my word on it. I lie. | 15:32 |
| ayoung | The code never does | 15:32 |
| *** changbl has quit IRC | 15:33 | |
| *** jtomasek has joined #openstack-dev | 15:33 | |
| *** avishayb has quit IRC | 15:33 | |
| *** steven-weston has joined #openstack-dev | 15:34 | |
| *** DinaBelova has joined #openstack-dev | 15:35 | |
| *** CaptTofu has joined #openstack-dev | 15:35 | |
| *** paragan has joined #openstack-dev | 15:36 | |
| *** datsun180b has joined #openstack-dev | 15:36 | |
| ayoung | dolphm, morganfainberg, WTF is this test supposed to be checking anyway? https://github.com/openstack/keystone/blob/master/keystone/tests/test_keystoneclient.py#L1037 | 15:37 |
| *** evgenyf has quit IRC | 15:37 | |
| *** rongze_ has joined #openstack-dev | 15:37 | |
| *** hartsocks has joined #openstack-dev | 15:37 | |
| *** rongze has quit IRC | 15:37 | |
| bknudson | ayoung: I looked at it and I couldn't figure it out. | 15:37 |
| ayoung | bknudson, glad it is not just me. Let me see If git blame can help | 15:38 |
| *** pschaef has joined #openstack-dev | 15:38 | |
| *** willingc has joined #openstack-dev | 15:39 | |
| *** afazekas has quit IRC | 15:39 | |
| ayoung | bknudson, https://github.com/openstack/keystone/commit/4ab47ad224c422dcd96aa256740945d1e6a8a208 | 15:39 |
| *** jhesketh__ has quit IRC | 15:40 | |
| *** safchain has quit IRC | 15:40 | |
| ayoung | "in addition to changing the users password, all current tokens | 15:40 |
| ayoung | will be cleared (for token backends that support listing) and | 15:40 |
| ayoung | a new token id will be returned." | 15:40 |
| bknudson | ayoung: ah, so it's testing that a new token is returned and can use that token. | 15:40 |
| bknudson | not really testing keystonclient here. | 15:40 |
| ayoung | bknudson, so I am guessing that the python-keystoneclient change broke that by ignoring the new token returned after the change password event | 15:40 |
| *** ericwb has joined #openstack-dev | 15:41 | |
| ayoung | bknudson, well, we did break it | 15:41 |
| dolphm | ayoung: it's testing that the user's token is revoked after a password update | 15:41 |
| *** herndon_ has quit IRC | 15:41 | |
| bknudson | ayoung: so could be changed to use a new client rather than existing client. | 15:41 |
| ayoung | dolphm, but that is not what broke | 15:41 |
| ayoung | bknudson, nope | 15:41 |
| ayoung | it needs to maintain the semantics | 15:41 |
| bknudson | or do we have to preserve the original keystoneclient behavior? | 15:41 |
| ayoung | which is "after changing a password, your client still works, but with a new token" | 15:42 |
| bknudson | (if so, there should be a test in keystoneclient) | 15:42 |
| *** __amotoki__ is now known as amotoki | 15:42 | |
| ayoung | bknudson, there is....this one | 15:42 |
| *** radez_g0n3 is now known as radez | 15:42 | |
| ayoung | it just is not well documented | 15:42 |
| dolphm | did the client break the service tests? | 15:42 |
| ayoung | dolphm, I approved the revoke commit for shardy 's patch | 15:43 |
| *** yassine has quit IRC | 15:43 | |
| *** egallen has quit IRC | 15:43 | |
| dolphm | ayoung: ? revert commit | 15:43 |
| ayoung | this is the test that broke, and the KC patch should have been run against this code... | 15:43 |
| ayoung | dolphm, yeah | 15:43 |
| dolphm | ayoung: got it | 15:43 |
| *** briancurtin has quit IRC | 15:44 | |
| bknudson | ayoung: dolphm: I posted a review to make it easier to test against local keystoneclient: https://review.openstack.org/#/c/57586/ | 15:44 |
| dolphm | ayoung: this test could be updated to do an explicit token validation, rather than just attempting an operation assuming the same token will be used | 15:44 |
| ayoung | dolphm, so ... why wasn't it? It looks like, when we commit a change to the KC, we need to run the Keystone unit tests...or...move the kc client tests to Tempest...lets work on the latter. | 15:44 |
| *** colinmcnamara has joined #openstack-dev | 15:44 | |
| *** yassine has joined #openstack-dev | 15:44 | |
| ayoung | dolphm, actually, this test shows a problem, when you explicitly set the token_id, it was ignored | 15:45 |
| dolphm | bknudson: (without looking at your commit) if you checkout changes into the vendor/python-keystoneclient-master/ repo, the test infrastructure won't overwrite that change with a new checkout | 15:45 |
| *** MoneyMaker has joined #openstack-dev | 15:45 | |
| bknudson | dolphm: that's why it doesn't use vendor/python-keystoneclient-master. | 15:45 |
| ayoung | https://github.com/openstack/keystone/blob/master/keystone/tests/test_keystoneclient.py#L1063 getst the token_id from the change password request | 15:46 |
| ayoung | https://github.com/openstack/keystone/blob/master/keystone/tests/test_keystoneclient.py#L1066 explicitly sets the token | 15:46 |
| ayoung | and then he does a list. | 15:46 |
| *** avishayb has joined #openstack-dev | 15:46 | |
| dolphm | bknudson: i don't follow? my approach accomplishes the same thing as yours without extra code | 15:46 |
| ayoung | shardy's code should probably do the same, if only to keep consistant semantics. | 15:46 |
| *** briancurtin has joined #openstack-dev | 15:47 | |
| *** MoneyMaker has quit IRC | 15:47 | |
| *** egallen has joined #openstack-dev | 15:48 | |
| *** boris-42 has quit IRC | 15:48 | |
| ayoung | dolphm, shardy could do the same thing as this test does: instead of getting a token via change passwrod, he would get it from the trust, and then explicitly set it in the client. | 15:48 |
| bknudson | dolphm: ok, I see how that works, too. | 15:48 |
| *** markmcclain has quit IRC | 15:48 | |
| *** hartsocks has quit IRC | 15:49 | |
| dolphm | bknudson: why does branch default to 'opt'? | 15:49 |
| bknudson | dolphm: just so it's not master. | 15:49 |
| bknudson | and that's the name of the test. | 15:49 |
| dolphm | bknudson: wouldn't it make more sense to just leave the repo alone, and test it as-is? | 15:50 |
| *** jimfehlig has left #openstack-dev | 15:50 | |
| bknudson | dolphm: I think so. That would require more changes to the code that does checkout. | 15:50 |
| dolphm | bknudson: the reason i ask is that i rarely use branches -- for openstack, i'm generally working from an detached HEAD / untracked branch | 15:51 |
| *** paragan has quit IRC | 15:52 | |
| bknudson | dolphm: http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/core.py#n121 | 15:52 |
| dstanek | dolphm: so your saying don't do a checkout in this case and just change the path to the client directory? | 15:53 |
| bknudson | it does clone / checkout master / pull / checkout branch | 15:53 |
| *** ndipanov has quit IRC | 15:53 | |
| dolphm | dstanek: bknudson: yeah, so just skip the checkout / pull / checkout | 15:53 |
| *** ruhe has quit IRC | 15:53 | |
| dolphm | cd into my repo and test it | 15:53 |
| dolphm | (if it actually needs to cd at all?) | 15:53 |
| *** buzztroll has joined #openstack-dev | 15:54 | |
| *** markmcclain has joined #openstack-dev | 15:54 | |
| bknudson | dolphm: should not be difficult. | 15:54 |
| dstanek | dolphm: no cd; just sys.path.append | 15:54 |
| *** jimjiang has joined #openstack-dev | 15:54 | |
| *** buzztroll has quit IRC | 15:54 | |
| *** buzztroll has joined #openstack-dev | 15:55 | |
| dstanek | dolphm: http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/test_keystoneclient.py#n54 | 15:55 |
| *** SergeyLukjanov has quit IRC | 15:55 | |
| *** sergmelikyan has joined #openstack-dev | 15:56 | |
| *** prekarat has quit IRC | 15:56 | |
| dstanek | dolphm, bknudson: that makes the common case much easier so i won't object, but it does make what i was going last night a little harder | 15:56 |
| *** tdruiva has joined #openstack-dev | 15:56 | |
| dolphm | dstanek: what were you doing last night? | 15:57 |
| dstanek | i was switching back and forth between revisions (like git bisect) to find breakage | 15:57 |
| dstanek | that's probably much less common than a dev testing their keystone client changes | 15:57 |
| *** amuller has quit IRC | 15:58 | |
| *** evgenyf has joined #openstack-dev | 15:58 | |
| *** tanisdl has joined #openstack-dev | 15:58 | |
| *** rraja has quit IRC | 15:58 | |
| *** afazekas has joined #openstack-dev | 16:00 | |
| *** edmund has joined #openstack-dev | 16:00 | |
| *** mosulica has quit IRC | 16:00 | |
| *** Ruetobas has quit IRC | 16:01 | |
| *** DinaBelova has quit IRC | 16:01 | |
| *** hartsocks has joined #openstack-dev | 16:01 | |
| *** egallen has quit IRC | 16:02 | |
| *** jruzicka has quit IRC | 16:02 | |
| *** Ruetobas has joined #openstack-dev | 16:03 | |
| *** atiwari has joined #openstack-dev | 16:03 | |
| *** datsun180b_ has joined #openstack-dev | 16:04 | |
| *** reidrac has quit IRC | 16:04 | |
| *** devoid has joined #openstack-dev | 16:04 | |
| *** datsun180b_ has quit IRC | 16:04 | |
| *** hartsocks has quit IRC | 16:04 | |
| *** datsun180b has quit IRC | 16:04 | |
| *** datsun180b has joined #openstack-dev | 16:04 | |
| *** YorikSar has quit IRC | 16:04 | |
| *** ifarkas has quit IRC | 16:04 | |
| *** paragan has joined #openstack-dev | 16:05 | |
| ayoung | shardy, bknudson, https://bugs.launchpad.net/python-keystoneclient/+bug/1244675/comments/6 does that make sense to you? | 16:05 |
| uvirtbot | Launchpad bug 1244675 in python-keystoneclient "keystoneclient sometimes uses a stale token" [Medium,In progress] | 16:05 |
| *** afazekas is now known as _afazekas | 16:05 | |
| *** ndipanov has joined #openstack-dev | 16:06 | |
| *** xingchao has joined #openstack-dev | 16:06 | |
| *** Ruetobas has quit IRC | 16:07 | |
| shardy | ayoung: I'm not sure, doesn't it always use auth_token_from_user, even if it's not valid? | 16:07 |
| *** rushiagr has quit IRC | 16:08 | |
| *** raj_m has quit IRC | 16:08 | |
| shardy | ayoung: As discussed in the comments on my original review? | 16:08 |
| *** rossella_s has quit IRC | 16:08 | |
| *** hartsocks has joined #openstack-dev | 16:08 | |
| shardy | ayoung: I just think doing authenticate() should always request a new token, and if you've got a new token, the client should use it | 16:08 |
| *** afazekas has joined #openstack-dev | 16:09 | |
| *** blamar has joined #openstack-dev | 16:10 | |
| *** devoid has quit IRC | 16:11 | |
| *** irimi has joined #openstack-dev | 16:11 | |
| *** xingchao has quit IRC | 16:11 | |
| *** danwent has joined #openstack-dev | 16:11 | |
| *** irimi has quit IRC | 16:11 | |
| dstanek | shardy: if that's the case wouldn't you be better of clearing the auth_token_from_using from authenticate()? | 16:12 |
| *** JordanP has joined #openstack-dev | 16:13 | |
| *** FunnyLookinHat has joined #openstack-dev | 16:13 | |
| shardy | dstanek: Yes that could be one approach, or store the user-provided token in auth_ref so it gets overwritten after the authenticate() | 16:13 |
| *** Ruetobas has joined #openstack-dev | 16:13 | |
| *** colinmcnamara has quit IRC | 16:13 | |
| *** jmontemayor has joined #openstack-dev | 16:14 | |
| shardy | dstanek: Anything which doesn't result in us requesting, and getting, a new token, but then using the old one forever | 16:14 |
| *** jmontemayor has quit IRC | 16:14 | |
| shardy | dstanek: From a user perspective that is *really* non-obvious | 16:14 |
| shardy | dstanek: If, as ayoung states, it's desired behavior, then authenticate() should fail to request a new token | 16:14 |
| *** mosulica has joined #openstack-dev | 16:16 | |
| *** gszasz has quit IRC | 16:16 | |
| *** gszasz has joined #openstack-dev | 16:16 | |
| *** CaptTofu has quit IRC | 16:17 | |
| *** CaptTofu has joined #openstack-dev | 16:17 | |
| *** carl_baldwin has joined #openstack-dev | 16:17 | |
| *** rongze_ has quit IRC | 16:17 | |
| *** haomaiwang has quit IRC | 16:17 | |
| *** CaptTofu has quit IRC | 16:18 | |
| *** jasdeepH has joined #openstack-dev | 16:19 | |
| *** CaptTofu has joined #openstack-dev | 16:19 | |
| *** avishayb has quit IRC | 16:19 | |
| *** jmontemayor has joined #openstack-dev | 16:19 | |
| *** jmontemayor has quit IRC | 16:19 | |
| *** mrodden has joined #openstack-dev | 16:20 | |
| *** devoid has joined #openstack-dev | 16:20 | |
| *** devoid1 has joined #openstack-dev | 16:20 | |
| *** devoid1 has quit IRC | 16:20 | |
| *** xarses has quit IRC | 16:21 | |
| *** akrivoka has quit IRC | 16:22 | |
| *** rongze has joined #openstack-dev | 16:23 | |
| *** CaptTofu has quit IRC | 16:23 | |
| *** devoid has quit IRC | 16:24 | |
| *** radez is now known as radez_g0n3 | 16:25 | |
| *** evgenyf has quit IRC | 16:26 | |
| *** mihgen has quit IRC | 16:26 | |
| *** akrivoka has joined #openstack-dev | 16:27 | |
| *** haomaiwang has joined #openstack-dev | 16:27 | |
| *** pmathews has joined #openstack-dev | 16:27 | |
| *** CaptTofu has joined #openstack-dev | 16:28 | |
| *** dkranz has quit IRC | 16:29 | |
| *** mlavalle has joined #openstack-dev | 16:29 | |
| *** markmcclain has quit IRC | 16:29 | |
| *** pablosan has joined #openstack-dev | 16:29 | |
| *** pete5 has joined #openstack-dev | 16:29 | |
| *** alop has joined #openstack-dev | 16:30 | |
| *** gszasz has quit IRC | 16:30 | |
| *** viktors1 has joined #openstack-dev | 16:30 | |
| *** teran has quit IRC | 16:30 | |
| *** ruhe has joined #openstack-dev | 16:31 | |
| *** ruhe has quit IRC | 16:31 | |
| *** amotoki is now known as amotoki_zzz | 16:33 | |
| *** branen_ has joined #openstack-dev | 16:34 | |
| *** belmoreira has quit IRC | 16:35 | |
| *** branen_ has quit IRC | 16:35 | |
| *** belmoreira has joined #openstack-dev | 16:35 | |
| *** willingc has quit IRC | 16:36 | |
| *** arezmerita has joined #openstack-dev | 16:37 | |
| *** willingc has joined #openstack-dev | 16:37 | |
| *** colinmcnamara has joined #openstack-dev | 16:38 | |
| *** jruzicka has joined #openstack-dev | 16:39 | |
| *** belmoreira has quit IRC | 16:39 | |
| *** ndipanov has quit IRC | 16:40 | |
| *** zbitter has joined #openstack-dev | 16:40 | |
| *** willingc has quit IRC | 16:41 | |
| *** mikeoutland has joined #openstack-dev | 16:42 | |
| *** abk has quit IRC | 16:43 | |
| *** zaneb has quit IRC | 16:43 | |
| *** colinmcnamara has quit IRC | 16:44 | |
| *** mikeoutland has quit IRC | 16:44 | |
| *** arezmerita has quit IRC | 16:44 | |
| *** colinmcn_ has joined #openstack-dev | 16:44 | |
| *** mosulica has quit IRC | 16:44 | |
| *** mosulica has joined #openstack-dev | 16:45 | |
| *** e0ne_ has quit IRC | 16:46 | |
| *** jmontemayor has joined #openstack-dev | 16:47 | |
| *** jmontemayor has quit IRC | 16:47 | |
| *** yassine has quit IRC | 16:47 | |
| *** yassine has joined #openstack-dev | 16:48 | |
| *** colinmcn_ has quit IRC | 16:49 | |
| *** galstrom_zzz is now known as galstrom | 16:50 | |
| *** steven-weston has quit IRC | 16:51 | |
| *** mosulica has quit IRC | 16:51 | |
| *** samuelbercovici has quit IRC | 16:51 | |
| *** mosulica has joined #openstack-dev | 16:51 | |
| *** paragan has quit IRC | 16:52 | |
| *** mikeoutland has joined #openstack-dev | 16:52 | |
| *** sumanthns has joined #openstack-dev | 16:53 | |
| *** jmontemayor has joined #openstack-dev | 16:53 | |
| *** ndipanov has joined #openstack-dev | 16:53 | |
| *** yassine has quit IRC | 16:54 | |
| *** garyk has quit IRC | 16:54 | |
| *** mikeoutland has quit IRC | 16:55 | |
| *** pmathews has quit IRC | 16:55 | |
| *** cfriesen_ has quit IRC | 16:55 | |
| *** jmontemayor has quit IRC | 16:55 | |
| *** mikeoutland has joined #openstack-dev | 16:57 | |
| *** danwent has quit IRC | 16:57 | |
| *** mriedem1 has joined #openstack-dev | 16:58 | |
| *** mrodden1 has joined #openstack-dev | 16:59 | |
| *** danwent has joined #openstack-dev | 16:59 | |
| *** mriedem has quit IRC | 16:59 | |
| *** jmontemayor has joined #openstack-dev | 16:59 | |
| *** dkranz has joined #openstack-dev | 17:00 | |
| *** changbl has joined #openstack-dev | 17:00 | |
| *** pablosan has quit IRC | 17:01 | |
| *** mriedem1 has quit IRC | 17:01 | |
| *** mriedem has joined #openstack-dev | 17:01 | |
| *** rushiagr has joined #openstack-dev | 17:01 | |
| *** mrodden has quit IRC | 17:01 | |
| *** mrodden1 has quit IRC | 17:01 | |
| *** akrivoka has quit IRC | 17:02 | |
| *** neelashah1 has joined #openstack-dev | 17:03 | |
| *** mrodden has joined #openstack-dev | 17:03 | |
| *** jmontemayor has quit IRC | 17:03 | |
| *** neelashah has quit IRC | 17:04 | |
| *** athomas has quit IRC | 17:05 | |
| *** rudrarugge has joined #openstack-dev | 17:05 | |
| *** mmagr has quit IRC | 17:06 | |
| *** jmontemayor has joined #openstack-dev | 17:06 | |
| *** willingc has joined #openstack-dev | 17:07 | |
| *** evgenyf has joined #openstack-dev | 17:08 | |
| *** safchain has joined #openstack-dev | 17:09 | |
| *** Akrog has quit IRC | 17:10 | |
| *** xarses has joined #openstack-dev | 17:11 | |
| *** markwash has quit IRC | 17:11 | |
| *** prekarat has joined #openstack-dev | 17:12 | |
| *** rudrarugge has quit IRC | 17:12 | |
| *** reed has joined #openstack-dev | 17:13 | |
| *** reed has quit IRC | 17:13 | |
| *** reed has joined #openstack-dev | 17:13 | |
| *** willingc has quit IRC | 17:16 | |
| *** boris-42 has joined #openstack-dev | 17:16 | |
| *** reidrac has joined #openstack-dev | 17:16 | |
| *** nouser has joined #openstack-dev | 17:17 | |
| *** aditirav has joined #openstack-dev | 17:17 | |
| *** pmathews has joined #openstack-dev | 17:17 | |
| *** jpich has quit IRC | 17:18 | |
| *** mosulica has quit IRC | 17:19 | |
| *** safchain has quit IRC | 17:20 | |
| *** tmclaugh[work] has joined #openstack-dev | 17:20 | |
| *** jp_at_hp has quit IRC | 17:21 | |
| *** SergeyLukjanov has joined #openstack-dev | 17:22 | |
| *** Artpicre has quit IRC | 17:23 | |
| *** nermina has quit IRC | 17:23 | |
| *** pmathews has quit IRC | 17:24 | |
| *** angdraug has joined #openstack-dev | 17:24 | |
| *** markmcclain has joined #openstack-dev | 17:24 | |
| *** zbitter is now known as zaneb | 17:25 | |
| *** tanisdl has quit IRC | 17:26 | |
| mordred | saschpe: ping | 17:26 |
| *** Artpicre has joined #openstack-dev | 17:26 | |
| *** corXi has quit IRC | 17:27 | |
| *** schwicht has joined #openstack-dev | 17:27 | |
| dsantos_ | hi, does someone got keystone (and/or other services) running in apache2? I followed some tutorials but I didn't get... After all the configurations I try to restart apache2 and I receive an error "(98)Address already in use: make_sock: could not bind to address 0.0.0.0:5000" | 17:27 |
| *** xavious has joined #openstack-dev | 17:28 | |
| sgran | stop the python service ? | 17:28 |
| *** garyk has joined #openstack-dev | 17:28 | |
| *** cyberx has left #openstack-dev | 17:28 | |
| *** neelashah1 has quit IRC | 17:28 | |
| *** ekarlso has quit IRC | 17:30 | |
| *** fbo is now known as fbo_away | 17:31 | |
| *** evgenyf has quit IRC | 17:31 | |
| *** teran has joined #openstack-dev | 17:31 | |
| *** neelashah has joined #openstack-dev | 17:34 | |
| *** danpb has quit IRC | 17:34 | |
| *** gmoro has quit IRC | 17:34 | |
| *** mriedem1 has joined #openstack-dev | 17:35 | |
| *** bdpayne has joined #openstack-dev | 17:36 | |
| *** iartarisi has quit IRC | 17:36 | |
| *** teran_ has joined #openstack-dev | 17:36 | |
| *** mrodden1 has joined #openstack-dev | 17:36 | |
| *** MaxV has quit IRC | 17:36 | |
| *** dolphm is now known as dolphm_afk | 17:37 | |
| *** MaxV has joined #openstack-dev | 17:37 | |
| *** ekarlso has joined #openstack-dev | 17:37 | |
| *** mriedem has quit IRC | 17:37 | |
| *** teran has quit IRC | 17:38 | |
| *** athomas has joined #openstack-dev | 17:38 | |
| *** Ryan_Lane has joined #openstack-dev | 17:38 | |
| *** thelorax123 has quit IRC | 17:38 | |
| *** cdub_ has joined #openstack-dev | 17:38 | |
| *** isd has joined #openstack-dev | 17:38 | |
| *** ruhe has joined #openstack-dev | 17:39 | |
| dstanek | dsantos_: do you have apache and keystone running on port 5000? | 17:39 |
| *** martyntaylor has quit IRC | 17:39 | |
| *** byeager has joined #openstack-dev | 17:39 | |
| *** mrodden has quit IRC | 17:39 | |
| *** thelorax123 has joined #openstack-dev | 17:39 | |
| *** mfer has quit IRC | 17:41 | |
| *** jistr has quit IRC | 17:41 | |
| *** MaxV has quit IRC | 17:41 | |
| dsantos_ | dstanek: I tried to change the port... I configured the virtualhost to listen on port 5001 but I got the same error saying that apache could not bind to 0.0.0.0:5001 | 17:42 |
| *** RajM has joined #openstack-dev | 17:42 | |
| dstanek | dsantos_: what is listening on it already? | 17:42 |
| *** clayb has joined #openstack-dev | 17:42 | |
| dsantos_ | dstanek: nothing | 17:42 |
| *** mihgen has joined #openstack-dev | 17:42 | |
| *** vkmc has quit IRC | 17:42 | |
| *** devoid has joined #openstack-dev | 17:42 | |
| *** devoid has left #openstack-dev | 17:43 | |
| dsantos_ | dstanek: actually what I need is to configure SSL for the endpoints and I read here http://docs.openstack.org/security-guide/content/ch020_ssl-everywhere.html that I should use some web server | 17:43 |
| *** eglynn has quit IRC | 17:43 | |
| dstanek | dsantos_: have you tried 'sudo netstat -lp' to make sure? | 17:43 |
| *** tdruiva has quit IRC | 17:44 | |
| *** tdruiva has joined #openstack-dev | 17:44 | |
| dsantos_ | dstanek: yup | 17:44 |
| *** viktors1 has left #openstack-dev | 17:44 | |
| *** reed has quit IRC | 17:45 | |
| *** gmoro has joined #openstack-dev | 17:45 | |
| *** kushal has quit IRC | 17:47 | |
| *** markwash has joined #openstack-dev | 17:47 | |
| dstanek | dsantos_: what tutorial are you following? | 17:47 |
| *** romcheg has quit IRC | 17:49 | |
| *** ygbo has quit IRC | 17:49 | |
| *** thelorax123 has quit IRC | 17:49 | |
| mriedem1 | new tempest CI bug for neutron: https://bugs.launchpad.net/neutron/+bug/1253730 | 17:50 |
| uvirtbot | Launchpad bug 1253730 in tempest "test_routers fails with: TearDownException: %(num)d cleanUp operation failed" [Undecided,New] | 17:50 |
| *** hemnafk is now known as hemna_ | 17:50 | |
| *** thelorax123 has joined #openstack-dev | 17:50 | |
| *** nermina has joined #openstack-dev | 17:51 | |
| *** galstrom is now known as galstrom_zzz | 17:51 | |
| *** CaptTofu has quit IRC | 17:52 | |
| dsantos_ | dstanek: I'm following these two tutorials https://keystone-voms.readthedocs.org/en/latest/requirements.html and http://adam.younglogic.com/2012/04/keystone-httpd/ | 17:54 |
| *** zodiak has quit IRC | 17:55 | |
| *** devlaps has joined #openstack-dev | 17:55 | |
| ericwb | i've configured keystone to run in apache, but on port 443 | 17:56 |
| *** joesavak has joined #openstack-dev | 17:57 | |
| *** abk has joined #openstack-dev | 17:57 | |
| *** alunduil has joined #openstack-dev | 17:58 | |
| *** herndon has joined #openstack-dev | 17:58 | |
| *** SumitNaiksatam has joined #openstack-dev | 17:58 | |
| *** colinmcnamara has joined #openstack-dev | 17:58 | |
| *** mattf has left #openstack-dev | 17:58 | |
| *** sumanthns has quit IRC | 17:59 | |
| *** sarob has joined #openstack-dev | 18:00 | |
| *** abk has quit IRC | 18:01 | |
| *** danwent has quit IRC | 18:01 | |
| *** adjohn has joined #openstack-dev | 18:02 | |
| *** derekh has quit IRC | 18:02 | |
| *** danielbruno has joined #openstack-dev | 18:04 | |
| *** arosen has quit IRC | 18:05 | |
| *** boris-42 has quit IRC | 18:05 | |
| *** melwitt has joined #openstack-dev | 18:05 | |
| *** pablosan has joined #openstack-dev | 18:06 | |
| dstanek | dsantos_: you don't have keystone running do you? | 18:06 |
| *** sarob has quit IRC | 18:07 | |
| *** danwent has joined #openstack-dev | 18:07 | |
| dsantos_ | dstanek: I have a devstack installation... keystone is running | 18:07 |
| *** epim has joined #openstack-dev | 18:07 | |
| *** sarob has joined #openstack-dev | 18:07 | |
| *** twoputt has joined #openstack-dev | 18:08 | |
| *** twoputt_ has joined #openstack-dev | 18:08 | |
| ericwb | some of adam's blog might be out-dated by now. | 18:08 |
| *** zodiak has joined #openstack-dev | 18:08 | |
| ericwb | but it is what i used to start | 18:08 |
| dstanek | dsantos_: if keystone is running already it will be running on ports 5000 and 35357; so Apache can't run there too | 18:08 |
| *** gyee has joined #openstack-dev | 18:08 | |
| *** markmcclain has quit IRC | 18:09 | |
| *** xga_ has quit IRC | 18:09 | |
| dsantos_ | dstanek: I know that... that's why I tried other ports but I got the same error... and also I tried to configure the apache without devstack running but I got the same error again | 18:09 |
| dstanek | dsantos_: then i'm at a loss; i don't know why apache would try to bind twice to the same port | 18:10 |
| *** sarob has quit IRC | 18:12 | |
| dsantos_ | dstanek: I saw this discussion http://ubuntuforums.org/showthread.php?t=1636667 and the problem here is because the cert had a password... I tried what the guy suggested but it didn't work for me | 18:12 |
| sgran | probably keystone.conf still says to use port 5000 ? | 18:12 |
| dstanek | dsantos_: does you default (out of the box) apache config run? | 18:12 |
| dsantos_ | sgran: why I should change that? | 18:12 |
| dsantos_ | dstanek: sorry... what did you mean? | 18:14 |
| ericwb | keystone.conf defaults to 5000 and 35357, but you don't need to change them to run in apache. | 18:14 |
| ericwb | the contents of you /etc/httpd/conf.d/keystone.conf probably have more of an affect on the ports in use | 18:15 |
| *** JordanP has quit IRC | 18:17 | |
| *** boris-42 has joined #openstack-dev | 18:18 | |
| *** abk has joined #openstack-dev | 18:19 | |
| *** boris-42 has quit IRC | 18:19 | |
| sgran | grep -r 5000 /etc/apache2/conf* - how many Listen directives do you hit? | 18:21 |
| *** mlavalle has quit IRC | 18:23 | |
| *** afazekas has quit IRC | 18:23 | |
| *** Mandell has joined #openstack-dev | 18:25 | |
| *** sahid has quit IRC | 18:25 | |
| *** prekarat has quit IRC | 18:26 | |
| *** rnirmal has joined #openstack-dev | 18:27 | |
| *** markmcclain has joined #openstack-dev | 18:29 | |
| *** aditirav has quit IRC | 18:29 | |
| *** mfer has joined #openstack-dev | 18:30 | |
| dsantos_ | ericwb: I'm not using "httpd"... I'm using apache2 | 18:33 |
| *** adjohn has quit IRC | 18:33 | |
| *** MaxV has joined #openstack-dev | 18:33 | |
| *** Shaan7 has quit IRC | 18:33 | |
| dsantos_ | sgran: I've no directives | 18:34 |
| *** tkammer has quit IRC | 18:34 | |
| dsantos_ | when I run the command you said | 18:34 |
| ericwb | dsantos_: by httpd I do mean Apache 2.2.x | 18:35 |
| *** hartsocks1 has joined #openstack-dev | 18:35 | |
| *** hartsocks1 has quit IRC | 18:35 | |
| *** hartsocks has quit IRC | 18:35 | |
| dsantos_ | ericwb: ok... I don't have the keystone.conf file into /etc/apache2/conf.d | 18:36 |
| *** adjohn has joined #openstack-dev | 18:37 | |
| dsantos_ | ericwb: I followed ayoung's post http://adam.younglogic.com/2012/04/keystone-httpd/... I just set the conf file in the wsgi file | 18:37 |
| ayoung | dsantos_, there is a more recent one...I'll get you the link | 18:37 |
| dsantos_ | ayoung: I would appreciate that | 18:38 |
| ayoung | dsantos_, http://andymc-stack.co.uk/2013/06/apache2-mod_wsgi-openstack-pt1-keystone/#comment-28 | 18:38 |
| *** armax has joined #openstack-dev | 18:38 | |
| ayoung | dsantos_, he did a follow on article for glance and Nova | 18:39 |
| ayoung | http://andymc-stack.co.uk/2013/07/apache2-mod_wsgi-openstack-pt-2-nova-api-os-compute-nova-api-ec2/ | 18:39 |
| ayoung | http://andymc-stack.co.uk/2013/07/apache2-mod_wsgi-openstack-pt-3-glance-api-glance-registry/ | 18:40 |
| *** shinylasers has joined #openstack-dev | 18:40 | |
| *** shinylasers has quit IRC | 18:40 | |
| ayoung | dsantos_, 4-6 (cinder, ceilometer and neutron) are all on the front page of his blog still: http://andymc-stack.co.uk/ | 18:40 |
| *** rongze has quit IRC | 18:40 | |
| dsantos_ | ayoung: nice... I'm gonna read these articles and try again :) | 18:43 |
| *** yaguang has quit IRC | 18:45 | |
| morganfainberg | allo | 18:46 |
| dstanek | morganfainberg: hi | 18:47 |
| ekarlso | jd__: you here for some question? | 18:48 |
| *** vartom11112 has joined #openstack-dev | 18:49 | |
| ayoung | Morgans morgan. | 18:49 |
| *** krtaylor has quit IRC | 18:50 | |
| ayoung | dsantos_, BTW, httpd is the name of the Apache Web server. We use that name to distinguish between the web server and all of the other projects that fall under the Apache project. | 18:50 |
| *** mrodden1 has quit IRC | 18:51 | |
| *** mriedem1 has quit IRC | 18:51 | |
| dsantos_ | ayoung: I understand... sometimes I just get lost with the paths... xD | 18:52 |
| dstanek | morganfainberg: should this test work? keystone.tests.test_backend_sql.SqlIdentity.test_cache_layer_role_crud | 18:55 |
| *** macjack has quit IRC | 18:55 | |
| *** rm_work|away is now known as rm_work | 18:56 | |
| *** YorikSar has joined #openstack-dev | 18:56 | |
| *** adjohn has quit IRC | 18:56 | |
| *** epim has quit IRC | 18:57 | |
| YorikSar | .цшт 1 | 18:58 |
| *** mrodden has joined #openstack-dev | 18:58 | |
| YorikSar | mordred: Hello. Have a minute? | 18:58 |
| *** mriedem has joined #openstack-dev | 18:58 | |
| YorikSar | mordred: I've found out that some projects moved to using use_develop=True in their tox.ini | 18:59 |
| *** amcrn has joined #openstack-dev | 18:59 | |
| YorikSar | mordred: And it looks dangerous to me. | 18:59 |
| *** mohits_ has joined #openstack-dev | 19:00 | |
| *** mlavalle has joined #openstack-dev | 19:01 | |
| *** epim has joined #openstack-dev | 19:04 | |
| *** pmathews has joined #openstack-dev | 19:04 | |
| *** yatin has joined #openstack-dev | 19:05 | |
| morganfainberg | dstanek, hm? | 19:05 |
| *** sarob has joined #openstack-dev | 19:06 | |
| morganfainberg | dstanek, i hope it works | 19:06 |
| dstanek | if you run it in the suite it works, but if you run it in isolation it doesn't | 19:06 |
| morganfainberg | dstanek, so it has an isolation issue. | 19:06 |
| morganfainberg | dstanek, hrm. let me take a quick look | 19:06 |
| *** yatin has left #openstack-dev | 19:07 | |
| dstanek | morganfainberg: but just looking at it i don't see how it could work | 19:07 |
| *** neelashah has quit IRC | 19:07 | |
| *** dolphm_afk is now known as dolphm | 19:07 | |
| dstanek | morganfainberg: i wouldn't think that this line could pass - http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/test_backend.py#n2682 | 19:08 |
| *** mohits_ has quit IRC | 19:08 | |
| morganfainberg | dstanek, hm.. | 19:09 |
| morganfainberg | dstanek, ok, that works becasue at that point we haven't invalidated the cache | 19:10 |
| morganfainberg | dstanek, we are using the direct call to the driver, circumventing the manager, meaning we read the value from the cache layer, not the underlying sql | 19:11 |
| morganfainberg | https://github.com/openstack/keystone/blob/master/keystone/tests/test_backend.py#L2680 see how we call assignment_api.driver here? | 19:11 |
| *** colinmcnamara has quit IRC | 19:11 | |
| *** rongze has joined #openstack-dev | 19:11 | |
| morganfainberg | dstanek, all caching occurs at the manager layer, not the driver layer. | 19:11 |
| dstanek | morganfainberg: ok, so that means in isolation the cache isn't working? | 19:12 |
| *** neelashah has joined #openstack-dev | 19:12 | |
| morganfainberg | dstanek, it means it is working. | 19:12 |
| morganfainberg | dstanek, unless it is failing at that point | 19:12 |
| dstanek | the assert i mentioned fails :-( | 19:13 |
| morganfainberg | ah | 19:13 |
| *** lucasagomes has quit IRC | 19:13 | |
| morganfainberg | it's because caching isn't enabled then. | 19:13 |
| morganfainberg | when the test is run in isolation | 19:13 |
| dstanek | ok, it looks like it should fail, but it's just misleading because of the cache | 19:13 |
| morganfainberg | dstanek, correct | 19:13 |
| *** galstrom_zzz is now known as galstrom | 19:13 | |
| dstanek | morganfainberg: does the setup explicitly enable the cache? | 19:13 |
| morganfainberg | dstanek, it is enabled in an override config | 19:14 |
| morganfainberg | http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/test_overrides.conf#n19 | 19:14 |
| *** sushils has quit IRC | 19:14 | |
| morganfainberg | dstanek, my guess is if you applied this patch https://review.openstack.org/#/c/53964/ it would skip that test in isolation | 19:16 |
| morganfainberg | or... | 19:16 |
| morganfainberg | dstanek, perhasp there is a race in cache region setup | 19:16 |
| morganfainberg | oh crud. i think i know what is going on | 19:17 |
| morganfainberg | dstanek, http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/core.py#n272 the first test has a bogus cache region | 19:17 |
| morganfainberg | :( | 19:17 |
| morganfainberg | i only run del on the first tearDown | 19:17 |
| morganfainberg | (maybe) | 19:18 |
| morganfainberg | maybe bogus that is | 19:18 |
| *** rongze has quit IRC | 19:18 | |
| *** rushiagr has quit IRC | 19:19 | |
| morganfainberg | dstanek, if the region is already configured http://git.openstack.org/cgit/openstack/keystone/tree/keystone/tests/core.py#n260 will not make a new region. if it is configured with the no-op region at this point, it wont cache anything. | 19:19 |
| morganfainberg | i think what is happening is that the config has already been loaded and the region is configured for the first run of setUp so the cache is failing until tearDown is called at least once | 19:19 |
| dstanek | morganfainberg: interesting...it would have taken me a long time to figure that out | 19:20 |
| morganfainberg | dstanek, heh, the cache implementation was hard to write good tests for | 19:21 |
| morganfainberg | dstanek, i only see it now because some of the headaches from dealing with the cache implementation i ran across when implementing it | 19:21 |
| dstanek | morganfainberg: thanks for the help - i'm going to poke at this for a little bit | 19:23 |
| morganfainberg | dstanek, another oddity is don't reference cache.region.backend directly | 19:24 |
| morganfainberg | it will raise an exception (or it used to) if it wasn't configured | 19:24 |
| *** joesavak has quit IRC | 19:25 | |
| morganfainberg | the correct way of checking if it is configured is "if 'backend' in region._dict__" | 19:25 |
| *** johnthetubaguy has quit IRC | 19:25 | |
| morganfainberg | it's ... silly | 19:25 |
| *** herndon has quit IRC | 19:25 | |
| morganfainberg | dstanek, but look at the keystone.common.cache.core and the way the region is configured for some more insight | 19:26 |
| morganfainberg | dstanek, let me know if you need any more help | 19:26 |
| *** xarses has quit IRC | 19:26 | |
| *** Shaan7 has joined #openstack-dev | 19:26 | |
| dstanek | morganfainberg: will do, thanks again - you have been most helpful | 19:27 |
| *** max_lobur is now known as max_lobur_afk | 19:27 | |
| *** ndipanov is now known as ndipanov_gone | 19:28 | |
| *** jtomasek has quit IRC | 19:28 | |
| *** steven-weston has joined #openstack-dev | 19:28 | |
| dsantos_ | ayoung: I was reading the tutorials but I saw this one is based on a previous version of apache server https://github.com/openstack/keystone/blob/master/doc/source/apache-httpd.rst | 19:32 |
| dsantos_ | ayoung: for example, if I put the keystone.conf file into the directory /etc/apache2/conf.d I get an error | 19:32 |
| dsantos_ | Invalid command '[DEFAULT]', perhaps misspelled or defined by a module not included in the server configuration | 19:33 |
| ayoung | dsantos_, yeah, should not be there. It is keystone config, not apache config, and needs to be out of that directory. SELinux comes in to play, too. | 19:33 |
| dsantos_ | ayoung: what about the file /etc/httpd/conf.d/nss.conf? is it the /etc/apache2/sites-available/default-ssl in apache2? | 19:34 |
| ayoung | dsantos_, nss.conf is apache specific | 19:35 |
| ayoung | dsantos_, nss and openssl are to options for crypto libraries | 19:36 |
| ayoung | RH, due to support for FIPS and Common criteria sensitive organizations, puts a good deal of effort into NSS | 19:36 |
| *** mikeoutland has quit IRC | 19:36 | |
| ayoung | which is why I wrote mine that way | 19:36 |
| ayoung | You an do mod_ssl if you prefer. | 19:37 |
| *** mikeoutland has joined #openstack-dev | 19:39 | |
| *** mikeoutland has quit IRC | 19:40 | |
| dsantos_ | ayoung: ahhh... ok... is just because I tried to configure nss (I've installed libapache2-mod-nss) but I never saw this file | 19:42 |
| *** xarses has joined #openstack-dev | 19:42 | |
| *** gyee has quit IRC | 19:43 | |
| *** amcrn is now known as [a]mcrn | 19:44 | |
| *** rongze has joined #openstack-dev | 19:45 | |
| *** ruhe has quit IRC | 19:45 | |
| *** SumitNaiksatam has left #openstack-dev | 19:45 | |
| *** novas0x2a|laptop has joined #openstack-dev | 19:46 | |
| *** krtaylor has joined #openstack-dev | 19:48 | |
| *** kpavel has joined #openstack-dev | 19:48 | |
| *** mihgen has quit IRC | 19:50 | |
| *** rongze has quit IRC | 19:50 | |
| *** fbo_away is now known as fbo | 19:51 | |
| *** sarob has quit IRC | 19:51 | |
| *** epim has quit IRC | 19:51 | |
| *** sarob has joined #openstack-dev | 19:52 | |
| *** mikeoutland has joined #openstack-dev | 19:52 | |
| *** neelashah has quit IRC | 19:53 | |
| ericwb | question: should a non-admin user be able to revoke his/her own token. i couldn't find a way to modify the policy.json to allow this. | 19:53 |
| *** mihgen has joined #openstack-dev | 19:54 | |
| *** neelashah has joined #openstack-dev | 19:54 | |
| *** yolanda has quit IRC | 19:55 | |
| *** jimfehlig has joined #openstack-dev | 19:55 | |
| *** tdruiva has quit IRC | 19:55 | |
| *** jdbarry has quit IRC | 19:56 | |
| *** tdruiva has joined #openstack-dev | 19:56 | |
| *** sarob has quit IRC | 19:56 | |
| *** pschaef has quit IRC | 19:57 | |
| *** aveiga has joined #openstack-dev | 19:57 | |
| *** alop has quit IRC | 19:57 | |
| *** jdbarry has joined #openstack-dev | 19:58 | |
| *** sarob has joined #openstack-dev | 19:59 | |
| *** sarob has quit IRC | 19:59 | |
| *** bdpayne has quit IRC | 19:59 | |
| *** sarob has joined #openstack-dev | 19:59 | |
| *** byeager has quit IRC | 20:00 | |
| *** bdpayne has joined #openstack-dev | 20:00 | |
| *** ericw has quit IRC | 20:01 | |
| *** adjohn has joined #openstack-dev | 20:04 | |
| *** abk has quit IRC | 20:07 | |
| *** YorikSar has quit IRC | 20:09 | |
| *** byeager has joined #openstack-dev | 20:10 | |
| ayoung | jvrbanac, would Barbican be interested in using Certmaster as a starting point for a CA? | 20:10 |
| ayoung | https://fedorahosted.org/certmaster/ | 20:10 |
| *** herndon has joined #openstack-dev | 20:10 | |
| ayoung | ericwb, I would think so. | 20:10 |
| ayoung | ericwb, the thing is, I don't know if there is enough info at the policy enforcement time to say. | 20:11 |
| *** ykhodork has joined #openstack-dev | 20:12 | |
| *** radix_ has quit IRC | 20:12 | |
| ayoung | ericwb, so the function call is https://github.com/openstack/keystone/blob/master/keystone/auth/controllers.py#L398 | 20:12 |
| ayoung | and that needs to do | 20:12 |
| ayoung | https://github.com/openstack/keystone/blob/master/keystone/common/controller.py#L138 | 20:13 |
| ayoung | get_member_from_driver | 20:13 |
| *** rods`` has quit IRC | 20:13 | |
| *** neelashah has quit IRC | 20:13 | |
| ayoung | https://github.com/openstack/keystone/blob/master/etc/policy.v3cloudsample.json#L6 | 20:13 |
| ayoung | "identity:revoke_token": "rule:admin_or_owner" | 20:14 |
| ayoung | "identity:revoke_token": "rule:admin_or_owner" | 20:14 |
| ayoung | ericwb, so look at the policy.v3cloudsample.json file for how to do it. It really is a better default policy file. | 20:15 |
| ericwb | ok i'll try again. thanks | 20:15 |
| *** vipul is now known as vipul-away | 20:15 | |
| *** vipul-away is now known as vipul | 20:15 | |
| *** vladikr has quit IRC | 20:16 | |
| *** jimfehlig has quit IRC | 20:17 | |
| *** colinmcnamara has joined #openstack-dev | 20:18 | |
| *** mikeoutland has quit IRC | 20:19 | |
| *** gordc has quit IRC | 20:20 | |
| *** jamielennox is now known as jamielennox|away | 20:21 | |
| *** raj_m has joined #openstack-dev | 20:22 | |
| *** colinmcnamara has quit IRC | 20:22 | |
| *** galstrom is now known as galstrom_zzz | 20:22 | |
| jvrbanac | ayoung, I have no idea. The Barbican Tech Lead (John Wood: woodster) and Project Owner (Jarret Raim: jraim) would be the best people to talk to about that. Right now, they have a better picture of the future vision. | 20:23 |
| *** byeager has quit IRC | 20:23 | |
| *** bdpayne has quit IRC | 20:23 | |
| dstanek | morganfainberg: is there any reason that configure_cache_region doesn't remove an existing backend? | 20:26 |
| morganfainberg | dstanek, hm. there is a reason... let me try and remember why | 20:26 |
| *** colinmcnamara has joined #openstack-dev | 20:27 | |
| *** rods` has joined #openstack-dev | 20:27 | |
| *** arezmerita has joined #openstack-dev | 20:28 | |
| *** bdpayne has joined #openstack-dev | 20:30 | |
| dstanek | morganfainberg: looks like it is only used in keystone.services and the tests | 20:30 |
| *** CaptTofu has joined #openstack-dev | 20:31 | |
| *** jecarey has quit IRC | 20:32 | |
| *** mrodden has quit IRC | 20:32 | |
| *** danielbruno has quit IRC | 20:32 | |
| *** ykhodork has quit IRC | 20:34 | |
| *** mriedem1 has joined #openstack-dev | 20:35 | |
| *** mrodden1 has joined #openstack-dev | 20:35 | |
| *** xqueralt has quit IRC | 20:36 | |
| *** vipul is now known as vipul-away | 20:38 | |
| *** vipul-away is now known as vipul | 20:38 | |
| *** mriedem has quit IRC | 20:38 | |
| *** colinmcnamara has quit IRC | 20:39 | |
| *** djoreilly has joined #openstack-dev | 20:39 | |
| *** arezmerita has left #openstack-dev | 20:40 | |
| jvrbanac | ayoung, Sorry I'm not much of a help. | 20:40 |
| notmyname | ttx: ping | 20:41 |
| ayoung | jvrbanac, yeah, but you are actually on IRC. Go bug them to do the same, would you? | 20:44 |
| ayoung | :) | 20:44 |
| *** danielbruno has joined #openstack-dev | 20:44 | |
| *** vartom11112 has quit IRC | 20:45 | |
| *** neelashah has joined #openstack-dev | 20:45 | |
| jvrbanac | ayoung, lol. You can usually find them in #openstack-barbican | 20:45 |
| jvrbanac | we're currently trying to move channels and it's causing a bit of confusion lol | 20:46 |
| *** nermina has quit IRC | 20:47 | |
| *** rongze has joined #openstack-dev | 20:47 | |
| *** markmcclain has quit IRC | 20:47 | |
| *** yeylon_ has quit IRC | 20:47 | |
| *** bknudson has quit IRC | 20:48 | |
| *** xarses has quit IRC | 20:48 | |
| *** jecarey has joined #openstack-dev | 20:48 | |
| *** alop has joined #openstack-dev | 20:48 | |
| *** xarses has joined #openstack-dev | 20:49 | |
| *** markmc has quit IRC | 20:50 | |
| morganfainberg | dstanek, i'll see if i can remember why it was done that way, i need to run though. be back later | 20:50 |
| *** rongze has quit IRC | 20:51 | |
| *** reidrac has quit IRC | 20:52 | |
| *** cdub_ has quit IRC | 20:53 | |
| *** cdub_ has joined #openstack-dev | 20:53 | |
| *** vipul is now known as vipul-away | 20:53 | |
| *** Chicago has joined #openstack-dev | 20:53 | |
| *** djoreilly has left #openstack-dev | 20:54 | |
| *** vipul-away is now known as vipul | 20:54 | |
| *** rfolco has quit IRC | 20:56 | |
| *** markmcclain has joined #openstack-dev | 20:57 | |
| *** bswartz has quit IRC | 20:57 | |
| *** datsun180b has quit IRC | 20:58 | |
| *** danwent has quit IRC | 20:58 | |
| *** bdpayne has quit IRC | 21:00 | |
| *** johnthetubaguy has joined #openstack-dev | 21:00 | |
| *** bdpayne has joined #openstack-dev | 21:01 | |
| *** armax has left #openstack-dev | 21:01 | |
| *** danwent has joined #openstack-dev | 21:02 | |
| mordred | if anyone sees YorikSar ... use_develop is quite on purpose | 21:02 |
| *** SergeyLukjanov has quit IRC | 21:02 | |
| *** dprince has quit IRC | 21:03 | |
| *** jimfehlig has joined #openstack-dev | 21:04 | |
| *** jimfehlig has left #openstack-dev | 21:04 | |
| *** denis_makogon_ is now known as denis_makogon | 21:04 | |
| *** joesavak has joined #openstack-dev | 21:05 | |
| *** bknudson has joined #openstack-dev | 21:05 | |
| *** tdruiva has quit IRC | 21:06 | |
| *** tdruiva has joined #openstack-dev | 21:07 | |
| *** mrodden1 is now known as mrodden | 21:08 | |
| *** tdruiva has quit IRC | 21:08 | |
| *** tdruiva has joined #openstack-dev | 21:09 | |
| *** adjohn has quit IRC | 21:11 | |
| *** dvarga has quit IRC | 21:11 | |
| *** vartom11112 has joined #openstack-dev | 21:11 | |
| *** Underbyte has quit IRC | 21:11 | |
| *** gyee has joined #openstack-dev | 21:12 | |
| *** alunduil has quit IRC | 21:12 | |
| *** tonix has joined #openstack-dev | 21:13 | |
| *** byeager has joined #openstack-dev | 21:14 | |
| *** mikeoutland has joined #openstack-dev | 21:15 | |
| *** f13o has joined #openstack-dev | 21:15 | |
| *** sandywalsh has quit IRC | 21:16 | |
| *** nermina has joined #openstack-dev | 21:18 | |
| *** flaper87 is now known as flaper87|afk | 21:18 | |
| *** adjohn has joined #openstack-dev | 21:19 | |
| *** asalkeld has joined #openstack-dev | 21:19 | |
| *** reed has joined #openstack-dev | 21:20 | |
| *** MarkAtwood has joined #openstack-dev | 21:20 | |
| *** armax has joined #openstack-dev | 21:20 | |
| *** galstrom_zzz is now known as galstrom | 21:21 | |
| *** xavious has quit IRC | 21:21 | |
| *** marun has quit IRC | 21:23 | |
| *** sandywalsh has joined #openstack-dev | 21:25 | |
| *** isd has quit IRC | 21:26 | |
| sdague | mrodden: so can you propose a devstack cleanup to pull the powervm driver as well? | 21:27 |
| sdague | if it's coming out of nova, it should come out of that too | 21:27 |
| mrodden | sdague: right, forgot about that | 21:27 |
| sdague | no worries | 21:28 |
| *** mikeoutland has quit IRC | 21:28 | |
| *** hdd has quit IRC | 21:29 | |
| *** che-arne has joined #openstack-dev | 21:30 | |
| *** steven-weston has quit IRC | 21:30 | |
| *** pcm_ has quit IRC | 21:34 | |
| *** Underbyte has joined #openstack-dev | 21:34 | |
| dstanek | what job actually publishes the keystone docs seen here? http://docs.openstack.org/developer/keystone/ | 21:34 |
| ekarlso | What's teh status of the auth plugin stuff dstanek ? | 21:34 |
| dstanek | ekarlso: the revert commit? | 21:35 |
| ekarlso | dstanek: no as in if it is going anywhere | 21:36 |
| *** danwent has quit IRC | 21:36 | |
| dstanek | ekarlso: oh, i have no idea what's been decided | 21:37 |
| *** vartom11112 has quit IRC | 21:38 | |
| *** RajM has quit IRC | 21:39 | |
| *** raj_m has quit IRC | 21:39 | |
| *** danwent has joined #openstack-dev | 21:44 | |
| *** romcheg has joined #openstack-dev | 21:44 | |
| *** steven-weston has joined #openstack-dev | 21:45 | |
| *** steven-weston has quit IRC | 21:46 | |
| *** steven-weston has joined #openstack-dev | 21:46 | |
| *** byeager has quit IRC | 21:47 | |
| *** rongze has joined #openstack-dev | 21:47 | |
| *** nermina has quit IRC | 21:49 | |
| *** mihgen has quit IRC | 21:51 | |
| *** byeager has joined #openstack-dev | 21:51 | |
| *** rongze has quit IRC | 21:52 | |
| *** CaptTofu has quit IRC | 21:52 | |
| *** CaptTofu has joined #openstack-dev | 21:53 | |
| *** mikeoutland has joined #openstack-dev | 21:54 | |
| *** singhs has joined #openstack-dev | 21:55 | |
| *** mfer has quit IRC | 21:56 | |
| *** byeager has quit IRC | 21:58 | |
| *** adjohn has quit IRC | 21:59 | |
| *** jhesketh has joined #openstack-dev | 21:59 | |
| *** athomas has quit IRC | 22:03 | |
| *** ykhodork has joined #openstack-dev | 22:04 | |
| *** danielbruno has quit IRC | 22:04 | |
| *** clayb has quit IRC | 22:05 | |
| *** ChanServ changes topic to "OpenStack development || Support is in #openstack" | 22:06 | |
| *** f13o has quit IRC | 22:07 | |
| *** gmoro has quit IRC | 22:07 | |
| *** jasondotstar has quit IRC | 22:07 | |
| *** danwent has quit IRC | 22:08 | |
| *** byeager has joined #openstack-dev | 22:08 | |
| *** yamahata_ has joined #openstack-dev | 22:10 | |
| *** danielbruno has joined #openstack-dev | 22:10 | |
| *** radix_ has joined #openstack-dev | 22:10 | |
| *** thelorax123 has quit IRC | 22:11 | |
| *** thelorax123 has joined #openstack-dev | 22:12 | |
| *** sushils has joined #openstack-dev | 22:12 | |
| *** fifieldt has joined #openstack-dev | 22:12 | |
| *** boden has quit IRC | 22:13 | |
| *** mfer has joined #openstack-dev | 22:15 | |
| *** romcheg has left #openstack-dev | 22:15 | |
| *** danwent has joined #openstack-dev | 22:16 | |
| *** changbl has quit IRC | 22:17 | |
| *** sushils has quit IRC | 22:18 | |
| *** sushils has joined #openstack-dev | 22:18 | |
| *** mikeoutland has quit IRC | 22:18 | |
| *** f13o has joined #openstack-dev | 22:18 | |
| *** gmoro has joined #openstack-dev | 22:19 | |
| *** giulivo has quit IRC | 22:19 | |
| *** sgordon has quit IRC | 22:20 | |
| *** dolphm has quit IRC | 22:20 | |
| *** mrodden has quit IRC | 22:20 | |
| *** adjohn has joined #openstack-dev | 22:25 | |
| *** tmclaugh[work] has quit IRC | 22:26 | |
| lifeless | mikal: so, whenever you want to talk through oslo and code copy, I'd be delighted to do so. I think it's hurting us a lot, and this recent breakage - which is broken in H release branches - is a good reason to reevaluate things. | 22:26 |
| mikal | lifeless: sure, if you'd like. However, whilst it might fix some small (and relatively rare) cases, I don't think it helps with the general behindness of imports. | 22:29 |
| bknudson | is the policy to merge often or to only merge when necessary? | 22:29 |
| *** mrodden has joined #openstack-dev | 22:30 | |
| lifeless | mikal: How would it /not/ help ? | 22:30 |
| lifeless | mikal: Perhaps we are talking about different changes? | 22:30 |
| clarkb | lifeless: speaking of broken in H have you proposed your patch to havana? | 22:30 |
| clarkb | lifeless: we should do that before forgetting happens | 22:30 |
| *** loq_mac has joined #openstack-dev | 22:30 | |
| *** fbo is now known as fbo_away | 22:30 | |
| mikal | lifeless: the things which are not in libraries are where they are because their interfaces aren't deemed stable yet | 22:30 |
| *** mfer has quit IRC | 22:30 | |
| lifeless | clarkb: I have not, I was asked to wait, and I believe mtreinish was going to do it. | 22:30 |
| mikal | lifeless: no a new library version with interface changes would require a requirements bump and tweeaks to the calling code | 22:30 |
| mikal | lifeless: which is the bit we're currently poor at | 22:31 |
| clarkb | lifeless: wfm (just want to make sure it doesn't get forgotten) | 22:31 |
| bknudson | "Developers making major changes to incubating APIs in oslo-incubator must be prepared to update the copies in the projects which have previously imported the code. " | 22:31 |
| mikal | lifeless: changing the mechanism of updates wont change their reliability | 22:31 |
| *** aveiga has quit IRC | 22:31 | |
| lifeless | mikal: so I'm proposing we a) have all of oslo /be/ a library, not copied. b) don't set an upper version cap in requirements.txt, c) gate on all changes, d) do the very simple 3-step dance needed when we change things. | 22:32 |
| lifeless | we're currently avoiding CI on oslo, which is bizarre for a project with massive cultural and technical CI investment like us. | 22:32 |
| mikal | lifeless: and I'm saying that wont help | 22:32 |
| mikal | lifeless: but you're not listening to me | 22:33 |
| *** ykhodork has quit IRC | 22:33 | |
| lifeless | mikal: I don't understand your point then. | 22:33 |
| *** alunduil has joined #openstack-dev | 22:33 | |
| lifeless | mikal: because I don't see how it won't help. | 22:33 |
| lifeless | mikal: I am listening - you have my entire focus. | 22:33 |
| mikal | lifeless: the corolocal storage bug from this week is rare, and not the case we should be solving for. I think that's happened once that I can remember. | 22:33 |
| *** buzztroll has quit IRC | 22:33 | |
| mikal | lifeless: the more common case is that the oslo library has moved on (interface changes, extra functionality, refactoring) | 22:34 |
| mikal | lifeless: things in oslo are meant to become libraries when those large changes stop happening | 22:34 |
| mikal | lifeless: so, we're talking about the body of code which is still solidifying | 22:34 |
| *** dkranz has quit IRC | 22:34 | |
| mikal | lifeless: now, if those things are libraries, we would need to lock the version in requirements for the projects that use those libraries | 22:35 |
| mikal | lifeless: and when those libraries did a release, someone would need to walk through all the user libraries and bump the requirements version and test | 22:35 |
| mikal | lifeless: that walk through is the bit we currently suck at | 22:35 |
| mikal | lifeless: and I don't see how changing the mechanism will improve that | 22:35 |
| lifeless | if you keep the axiom of 'we won't do graceful evolution of stuff in olso', yes. | 22:35 |
| lifeless | That axiom is the axiom I'm challenging. | 22:35 |
| *** jhesketh has quit IRC | 22:35 | |
| *** thomasem has quit IRC | 22:36 | |
| mikal | lifeless: i.e. the requirements bump bit needs someone who understands what they're doing, and those people are busy / don't know about all the imports | 22:36 |
| *** mikeoutland has joined #openstack-dev | 22:36 | |
| *** jhesketh has joined #openstack-dev | 22:36 | |
| mikal | lifeless: oslo-incubator is meant to be about rapid change and solidifing things as fast as possible | 22:36 |
| lifeless | The very definition of olso is flawed in that regard, and the multiple reviews, manual tracking of actual change, and sucking at keeping up are symptoms of the problem. | 22:36 |
| mikal | lifeless: I agree we could do something very complicated and resource intensive to maintain backwards compatability | 22:36 |
| mikal | lifeless: but I think that's a poor use of resources and wasn't the intent in the first place | 22:36 |
| lifeless | mikal: I'm not proposing complex, nor resource intensive. | 22:36 |
| lifeless | mikal: There is a meme that doing standard engineering practice - like we do for the clients and for the apis - is super hard | 22:37 |
| mikal | lifeless: I want to see us focus on graceful evolution where it matters (db upgrades, rpc apis) | 22:37 |
| mikal | lifeless: why complicated some tangential thing jsut because in a theoretical sense it could be more polished? | 22:37 |
| *** ykhodork has joined #openstack-dev | 22:37 | |
| lifeless | mikal: Thought experiment: how long should an olso interface take to stablise? | 22:38 |
| mikal | lifeless: why not work on an actually important problem? | 22:38 |
| lifeless | mikal: because it's wasting peoples time right now. | 22:38 |
| mikal | lifeless: that depends on the library... There are (or were last I looked) examples of libraries in oslo-incubator which were just a single cut and paste dump with no attempt to re-integrate them into their sources | 22:38 |
| mikal | lifeless: those will take longer than ones which are actively developed | 22:39 |
| *** michchap has quit IRC | 22:39 | |
| lifeless | I'm going to put some crude stats together | 22:39 |
| mikal | lifeless: I agree we should be graduating more than we have, but dhellman is already working on that | 22:39 |
| *** adjohn has quit IRC | 22:39 | |
| lifeless | I think your fear of overhead is misplaced, but I want to fact check my thoughts on that. | 22:39 |
| *** vijendar has quit IRC | 22:39 | |
| mikal | lifeless: I did ten oslo updates in a day last week, its not onerous if you know what you're doing | 22:40 |
| mikal | lifeless: and that's why the oslo maintainer for the library should do the work | 22:40 |
| mikal | lifeless: not someone random | 22:40 |
| *** markmcclain has quit IRC | 22:40 | |
| *** buzztroll has joined #openstack-dev | 22:40 | |
| *** adjohn has joined #openstack-dev | 22:41 | |
| *** michchap has joined #openstack-dev | 22:42 | |
| lifeless | mikal: you're talking about optimising a process that is a side effect of the way we've chosen to incubate new common deps | 22:42 |
| lifeless | mikal: I'm talking about changing the way we incubate new common deps | 22:42 |
| *** radix_ has quit IRC | 22:42 | |
| *** achampion has quit IRC | 22:42 | |
| ekarlso | awhat library should one use for doing ipaddress checks etc? | 22:43 |
| mikal | lifeless: yes, I know | 22:43 |
| mikal | lifeless: but I am saying your new way doesn't address the problems with the current way | 22:43 |
| *** adjohn has quit IRC | 22:43 | |
| mriedem1 | ttx: ping | 22:43 |
| mikal | lifeless: in fact, I suspect it makes them worse, because oslo-incubator doesn't behave in the way people expect libraries to behave | 22:43 |
| *** asalkeld has quit IRC | 22:44 | |
| lifeless | mikal: but thats the very thing I'm talking about changing | 22:44 |
| lifeless | mikal: I feel like you're the one not listening to me. | 22:44 |
| mikal | lifeless: and frankly, I'd like to see more focus on important problems than random change for theological reasons | 22:44 |
| *** jasdeepH has quit IRC | 22:44 | |
| mikal | lifeless: how does being in a library address forward porting though? | 22:44 |
| clarkb | ekarlso: what kind of checks? | 22:44 |
| *** jamielennox|away is now known as jamielennox | 22:45 | |
| lifeless | it provides signal to the maintainer that the porting hasn't been completed; which means that if the change is incompatible (conceptually, not just syntax or whatever) they find out immediately, not at some removed time) | 22:45 |
| *** buzztroll has quit IRC | 22:45 | |
| *** tdruiva has quit IRC | 22:45 | |
| *** buzztroll has joined #openstack-dev | 22:46 | |
| *** tdruiva has joined #openstack-dev | 22:46 | |
| mikal | lifeless: so how do you change the interface to the library without a very complicated dance? | 22:46 |
| *** danwent has quit IRC | 22:46 | |
| mikal | lifeless: a specific example from last releast -- we changed all the args for periodic tasks from talking 'ticks' to seconds | 22:46 |
| mikal | lifeless: that was easy to land because it was a couple of line code change in each repo | 22:47 |
| mikal | lifeless: but if the library was going to be backward compatible it wouldn't have worked out because it broke a bunch of existing assumptions which had to be hand verified | 22:47 |
| lifeless | mikal: it would have been very easy to do as backwards compat, take a new parameter, convert the old one at the top of the function,a nd issue deprecation warning. | 22:47 |
| mikal | lifeless: now, the oslo library with the chagne gets released | 22:47 |
| mikal | lifeless: and breaks gate for all the users | 22:47 |
| mikal | lifeless: how is that better? | 22:47 |
| lifeless | it doesn't break gate. | 22:47 |
| lifeless | It can't land because it's gated on. | 22:48 |
| bknudson | I would feel better if changes to oslo-incubator got applied to keystone and run through the gate automatically. | 22:48 |
| mikal | lifeless: the only way I can see to do this sensibly with libraries is to bump the major version when you do an interface change, and then cap requirements for users until the forward port is complete | 22:48 |
| *** rongze has joined #openstack-dev | 22:48 | |
| ekarlso | clarkb: like giving a ip, determining stuff about it | 22:49 |
| ekarlso | like ipv4 vs 6 | 22:49 |
| lifeless | mikal: arrgghh. We're clearly failing to communicate. | 22:49 |
| mikal | lifeless: agreed | 22:49 |
| lifeless | mikal: I'm saying 'do backwards compat in incubator' | 22:49 |
| mikal | lifeless: I think you need to go look at some interface changes and see how you would have written a backwards compatible version | 22:49 |
| lifeless | mikal: and 'this isn't hard', 'isn't complex', 'is sustainable' | 22:50 |
| mikal | lifeless: and I'm saying "that's a waste of time" | 22:50 |
| *** adjohn has joined #openstack-dev | 22:50 | |
| mikal | lifeless: yes, and I'm saying "bullshit" | 22:50 |
| notmyname | anyone know what Michael Chapman's IRC nick is? | 22:50 |
| lifeless | mikal: I'm saying that the N different project headaches with the process are symptoms that the assumptions about costs are wrong. | 22:50 |
| mikal | lifeless: its easy with trivial cases, but that's not always true | 22:50 |
| *** tdruiva has quit IRC | 22:50 | |
| *** buzztroll has quit IRC | 22:50 | |
| mikal | lifeless: what actual headaches have the users seen? | 22:50 |
| lifeless | mikal: I've just a little bit of experience running backwards compat across multiple years in python code. | 22:50 |
| mikal | lifeless: apart from one bug this week? | 22:51 |
| lifeless | mikal: so I take your bullshit and raise you a learn to code | 22:51 |
| mikal | lifeless: yes, I know you like to call me a n00b | 22:51 |
| mikal | lifeless: this conversation is over | 22:51 |
| lifeless | mikal: I was about to say the same thing | 22:51 |
| *** jhesketh has quit IRC | 22:53 | |
| *** jhesketh has joined #openstack-dev | 22:54 | |
| *** rongze has quit IRC | 22:54 | |
| *** danwent has joined #openstack-dev | 22:55 | |
| *** loq_mac has quit IRC | 22:56 | |
| *** mlavalle has quit IRC | 22:57 | |
| *** jprovazn has quit IRC | 22:57 | |
| *** AlanClark has quit IRC | 22:58 | |
| lifeless | And sorry to the channel for my poor behaviour there; I should have taken a breather rather than engaging in the escalation. | 23:01 |
| *** thomasem has joined #openstack-dev | 23:01 | |
| *** eharney has quit IRC | 23:02 | |
| *** mlavalle has joined #openstack-dev | 23:02 | |
| michchap | notmyname: Hi | 23:02 |
| *** kevinconway has quit IRC | 23:02 | |
| *** d34dh0r53 has joined #openstack-dev | 23:03 | |
| *** singhs_ has joined #openstack-dev | 23:04 | |
| *** asalkeld has joined #openstack-dev | 23:04 | |
| *** mlavalle has quit IRC | 23:05 | |
| *** singhs has quit IRC | 23:05 | |
| *** singhs_ is now known as singhs | 23:05 | |
| *** singhs has left #openstack-dev | 23:07 | |
| *** kbringard has quit IRC | 23:09 | |
| *** tdruiva has joined #openstack-dev | 23:10 | |
| *** bpokorny has quit IRC | 23:10 | |
| jamielennox | ayoung: ping | 23:10 |
| *** venkatesh has joined #openstack-dev | 23:10 | |
| ayoung | jamielennox, yo | 23:10 |
| jamielennox | ayoung: did you get a chance to look at that client extensions etherpad i wrote? | 23:11 |
| ayoung | some....been a lot to process today, but I did look at it this morning | 23:11 |
| jamielennox | yea, there is a stuff happening everywhere - we need to discuss the kds at some point | 23:11 |
| ayoung | ...which it looks like I just killed in my browser...one sec | 23:11 |
| jamielennox | gyee: you too if you're here | 23:12 |
| ayoung | https://etherpad.openstack.org/p/keystoneclient-extensions2 | 23:12 |
| notmyname | michchap: hi | 23:12 |
| ayoung | jamielennox, so, do we really need extensions *in* the keystoneclient or just extensions that can work *with* the keystoneclient? | 23:13 |
| jamielennox | all i'm really wondering at the moment is how you feel about the aproach | 23:13 |
| *** thomasem has quit IRC | 23:13 | |
| jamielennox | at some point we will absolutely need them in | 23:13 |
| jamielennox | because as much as i try to keep extensions limited to their own path it doesn't always work like that | 23:13 |
| jamielennox | so an extension needs to hook into for example the user object | 23:14 |
| *** burt1 has quit IRC | 23:14 | |
| *** danielbruno has quit IRC | 23:14 | |
| gyee | jamielennox, here | 23:14 |
| ayoung | jamielennox, like how the trusts and oauth inject data into the tokens? | 23:15 |
| gyee | I haven't look at the etherpad yet, been dealing with other issues | 23:15 |
| jamielennox | gyee: that's ok. I went into some python voodoo - i didn't even think about that they might not have much python experience | 23:15 |
| *** tdruiva has quit IRC | 23:15 | |
| jamielennox | ayoung: yes - but at the moment that one's not all that important because we don't really have an auth object yet | 23:16 |
| ayoung | jamielennox, so...lets assume that all an extension needs to do is something like: client.extensions.put(self) | 23:16 |
| jamielennox | but i _really_ want one | 23:16 |
| ayoung | yeah, me too | 23:16 |
| gyee | jamielennox, we are going to implement the framework we discussed at the summit right? | 23:16 |
| gyee | maintaining a global registry of resource to class mappings | 23:17 |
| jamielennox | gyee: yea that part is fine, it's been more about how we allow external extensions to interact with the internal code | 23:17 |
| gyee | have a register_handler() method or discovery via package entry points | 23:17 |
| ayoung | jamielennox, you are talking decorators there on the user object, and it makes sense if we need that behavior. Do we? | 23:17 |
| jamielennox | gyee: ahh - kind of, i'm not looking to *replace* existing resources, i'm looking for a way to have extensions extend them | 23:17 |
| gyee | jamielennox, that's just basic class inheritance right? | 23:18 |
| *** krtaylor has quit IRC | 23:18 | |
| *** galstrom is now known as galstrom_zzz | 23:18 | |
| jamielennox | ayoung: so given the token, trusts and oauth example - there really isn't another choice i think | 23:18 |
| ayoung | jamielennox, that is basically what we do with the pooicy stuff in keystone/common/controllers | 23:18 |
| jamielennox | gyee: not necessarily - we need multiple objects to be able to extend the same resource | 23:19 |
| *** ngoracke has quit IRC | 23:19 | |
| gyee | I am lost | 23:19 |
| ayoung | gyee, https://etherpad.openstack.org/p/keystoneclient-extensions2 line 56 | 23:19 |
| jamielennox | ayoung: no the decorator is more that that. It will patch in to the underlying resource class | 23:20 |
| jamielennox | gyee: say for example we have the trusts and oauth extension code loaded. they need to co-exist without getting in each others way. We can't just have them both extend the User class | 23:20 |
| ayoung | jamielennox, we put the decorator on each of the functions (or find some python magic to do that for us) and then an extension can say "run this function after create_user" or something, right? | 23:21 |
| gyee | ah, interesting | 23:21 |
| gyee | so its like interceptor pattern | 23:21 |
| ayoung | and it will be a stack, so first one called on the way in is the last one called on the way out | 23:21 |
| jamielennox | ayoung: yes. But the way we do that is wrap the base class function and install that onto the class object - just like a decorator but we will be decorating a different class | 23:21 |
| ayoung | https://en.wikipedia.org/wiki/Decorator_pattern | 23:21 |
| gyee | ayoung, isn't that an interceptor as well? | 23:22 |
| jamielennox | so by saying after('create') we are replacing the create on the User class - not on the extension class | 23:22 |
| gyee | either way, interesting idea | 23:22 |
| ayoung | jamielennox, by "wrap" you mean add an attribute lie @extendable or something | 23:23 |
| ayoung | like | 23:23 |
| *** jobewan has quit IRC | 23:23 | |
| gyee | everything is extensible | 23:23 |
| *** kbrierly has quit IRC | 23:23 | |
| ayoung | jamielennox, replace create, or just wrap it? | 23:23 |
| jamielennox | so if we have 3 functions wrapping UserManager.create they will each get called, see if the extension is available on the server and if so have the oportunity to modify the args either before or after the function | 23:23 |
| jamielennox | ayoung: wrap it i think | 23:23 |
| ayoung | jamielennox, pure decorator. Makes sense. | 23:24 |
| *** Chicago has quit IRC | 23:24 | |
| jamielennox | ayoung: a decorator always has the oportunity to short circuit but i think that's a bad idea | 23:24 |
| ayoung | jamielennox, you are thinking chain-of-responsibility | 23:24 |
| ayoung | decorator just adds to the existing | 23:24 |
| *** mrodden has quit IRC | 23:24 | |
| *** reed has quit IRC | 23:24 | |
| ayoung | you're design is good | 23:24 |
| ayoung | your | 23:24 |
| jamielennox | yea, i'm just wary about calling it a decorator because a decorator modifies the current class - in the extension we will be modifying the client's classes | 23:25 |
| *** mrodden has joined #openstack-dev | 23:25 | |
| *** tdruiva has joined #openstack-dev | 23:25 | |
| *** yamahata_ has quit IRC | 23:25 | |
| *** joesavak has quit IRC | 23:25 | |
| *** CaptTofu has quit IRC | 23:25 | |
| gyee | jamielennox, interceptor :) | 23:25 |
| jamielennox | gyee: lol, sure | 23:26 |
| *** CaptTofu has joined #openstack-dev | 23:26 | |
| gyee | aspect programming | 23:26 |
| *** lbragstad has quit IRC | 23:26 | |
| ayoung | gyee Yeah, this really is implementing AOP, but in a python sort of way | 23:26 |
| gyee | ayoung, sounds like an interesting idea | 23:27 |
| *** herndon has quit IRC | 23:27 | |
| * gyee need to read up on the etherpad | 23:27 | |
| *** kpavel has quit IRC | 23:27 | |
| ayoung | jamielennox, so would we pro-actively add the @ to all of the existing methods, or is there some monkeypatch style way to do it instead? | 23:28 |
| *** tdruiva has quit IRC | 23:29 | |
| *** jmontemayor has quit IRC | 23:30 | |
| *** nermina has joined #openstack-dev | 23:30 | |
| *** CaptTofu_ has joined #openstack-dev | 23:30 | |
| *** CaptTofu has quit IRC | 23:31 | |
| ayoung | gyee you make a cameo https://www.youtube.com/watch?v=VA-8K4_4NIg&feature=youtu.be | 23:31 |
| ayoung | around 40 seconds in | 23:31 |
| jamielennox | i've been thinking and i don't know if we need to do anything for the original client | 23:31 |
| ayoung | jamielennox, ok...I would suspect that the only object we need to wrap that way might be the user object. Maybe tokens | 23:32 |
| ayoung | but until we can unpack tokens in the client...its probably academic | 23:32 |
| *** asalkeld has quit IRC | 23:32 | |
| jamielennox | ayoung: so i was wondering if we need to provide like a 'whitelist' of what can be patched | 23:32 |
| jamielennox | ayoung: so have a look at: https://github.com/frasertweedale/elk/blob/master/elk/modifier.py | 23:33 |
| ayoung | jamielennox, around? | 23:34 |
| *** steven-weston has quit IRC | 23:34 | |
| jamielennox | before + after | 23:34 |
| jamielennox | traditional decorator | 23:35 |
| dosaboy | dhellman: ping | 23:35 |
| dosaboy | dhellmann: ping | 23:35 |
| *** neelashah has quit IRC | 23:35 | |
| jamielennox | but more to the point, it just does a getattr, creates a wrapper then sets it back again | 23:35 |
| *** changbl has joined #openstack-dev | 23:35 | |
| *** peristeri has quit IRC | 23:36 | |
| gyee | ayoung, ha, nice video | 23:36 |
| jamielennox | so an extension could hook whatever it likes - which is powerful, but possibly not good | 23:36 |
| ayoung | jamielennox, looks good. If it makes sense to reuse that, I am all for it. I think you are in the right vicinity. | 23:36 |
| jamielennox | then again - this is python. Give the developer everything and trust them not to screw it up | 23:37 |
| jamielennox | ayoung: i don't want to have a dependency on it | 23:37 |
| ayoung | jamielennox, so..on the token thing from yesterday, I assume we all wanted to do the revert, right? I pushed the button on it htis morning, but am still waiting for the merge | 23:37 |
| jamielennox | well we have to do the revert for the tests | 23:37 |
| *** dbalog has left #openstack-dev | 23:37 | |
| *** steven-weston has joined #openstack-dev | 23:38 | |
| ayoung | jamielennox, I am starting to think that the bug is invalid | 23:38 |
| jamielennox | the more i think about it, the more i think shardy's patch is actually fairly close | 23:38 |
| ayoung | we should require the user to either explicitly set the token, or to create a new client object for the trust token | 23:38 |
| jamielennox | i was concerned that it would affect the way token/endpoint was being handled, but there is still a check for that | 23:38 |
| jamielennox | it should be fairly easy to put in a flag or something to keep the current compatability | 23:39 |
| *** FunnyLookinHat has quit IRC | 23:39 | |
| ayoung | jamielennox, for example, Heat is going to need to keep it origianl identity, but constantly do stuff with trusts | 23:39 |
| jamielennox | ayoung: nah, you miss the point - and i don't think it's well put in the patch | 23:39 |
| ayoung | jamielennox, look at the failing test | 23:39 |
| jamielennox | what we have is good for endpoint/token - but it doesn't work for auth_url/token | 23:39 |
| jamielennox | oh, i know what the failure is | 23:40 |
| jamielennox | the assumption that auth_token_from_user is always to be used is good for the endpoint scenario, but it's wrong for the auth_url case | 23:40 |
| ayoung | https://github.com/openstack/keystone/blob/master/keystone/tests/test_keystoneclient.py#L1067 | 23:40 |
| jamielennox | yea, i know - force setting an auth_token should override everything else | 23:41 |
| ayoung | and it should be OK to do, no? | 23:41 |
| ayoung | the question is "what should a user do to consume a trust" | 23:41 |
| *** venkatesh has quit IRC | 23:42 | |
| jamielennox | it's ugly, but i assume the openstack-client will be relying on that too so yes | 23:42 |
| jamielennox | keystone.v3.client(auth_url="..", token="actual_user_token", trust_id='...') | 23:42 |
| *** steven-weston has quit IRC | 23:43 | |
| ayoung | so, right now, if I authenticate with userid/password, and get a token, that token can be traded in for any other token that I am authorized to get. That is kindof a bad design. It would make more sense to have some explicit way of tracking state and scope. | 23:43 |
| ayoung | so I do uid/pw and get a token scoped only to Keystone | 23:43 |
| ayoung | I keep that until expiry | 23:43 |
| jamielennox | so you authenticate with a token you already have (that could be replaces with user/pass) and you when you authenticate i want the token to be trust scoped | 23:43 |
| ayoung | with that token ,I can get tokens scoped to some smaller scope, WHen I am done with them...pop? | 23:43 |
| *** [a]mcrn is now known as amcrn | 23:44 | |
| ayoung | jamielennox, but you don't want to forget about the earlier token | 23:44 |
| ayoung | especially in trusts | 23:44 |
| jamielennox | pop as in discard the new client - sure | 23:44 |
| jamielennox | you don't have to, you just end up with multiple copies of keystoneclient | 23:44 |
| ayoung | I might have a workload that requries two trust based tokens, migrate a vm from user 1 to user 2 or something | 23:44 |
| jamielennox | the original and the trust scoped | 23:44 |
| *** sarob has quit IRC | 23:44 | |
| jamielennox | you use the one that has the priviledges that you need | 23:44 |
| ayoung | right, that, I think, is the right abstraction: whenever I change scope, I get a new object of some sort | 23:45 |
| *** sarob has joined #openstack-dev | 23:45 | |
| jamielennox | i have a way that i can simplify it with auth plugins, but that's not helpful now | 23:45 |
| ayoung | a user might even want to be able to list the tokens that they have in the client or something | 23:45 |
| ayoung | jamielennox, well, if shardy can explicitly create a new client object and hand off the trust token to it, I think his workflow is solved | 23:46 |
| jamielennox | there are no multiple tokens in the client | 23:46 |
| jamielennox | for now 1 token 1 client | 23:46 |
| ayoung | oooh | 23:46 |
| ayoung | so...If I want to remember a token, I need to explicitly read it out of hte client before I request a new token? | 23:46 |
| jamielennox | it can re-authenticate itself, but it's essentially the same token with a new expiry and it will ditch the old one | 23:46 |
| jamielennox | why are you manually requesting a new token? | 23:47 |
| ayoung | new expiry means new token...meaning you authenticated. I'm OK with ditching the token in that case | 23:47 |
| ayoung | jamielennox, from unscoped to scoped, or from regular to trust | 23:47 |
| *** sn0wb1rd has quit IRC | 23:47 | |
| *** MaxV has quit IRC | 23:48 | |
| jamielennox | base_client = client.v3.Client(user="user", pass="pass", domain="xxx") | 23:48 |
| jamielennox | trust_id = base_client.trusts.list()[0] | 23:48 |
| jamielennox | trust_scoped_client = client.v3.Client(auth_url="...", trust_id=trust_id, token=base_client.auth_token) | 23:49 |
| jamielennox | so the token that the second client gets is trust scoped, and if you want to use that token then you use the second client | 23:49 |
| *** cdub has quit IRC | 23:49 | |
| jamielennox | (obviously the syntax is all wrong) | 23:49 |
| *** atiwari has quit IRC | 23:49 | |
| *** gyee has quit IRC | 23:49 | |
| *** sarob has quit IRC | 23:49 | |
| jamielennox | (and i haven't explicitly tried any of this, but it's my understanding of the way this should work now) | 23:50 |
| *** gyee has joined #openstack-dev | 23:50 | |
| ayoung | jamielennox, can you do a proof of concept of that and post it? I think that is what shardy needs | 23:50 |
| *** rongze has joined #openstack-dev | 23:50 | |
| *** atiwari has joined #openstack-dev | 23:50 | |
| ayoung | I was going to do the same thing, but using curl and jq direct to the API | 23:50 |
| *** cdub has joined #openstack-dev | 23:50 | |
| *** jgrimm has quit IRC | 23:51 | |
| ayoung | OK, I need to head home...I'm at the office, and want to catch the kids before bed time. I'll be back on in a few hours. | 23:51 |
| jamielennox | well that, and he needs to be able to say that in the trust_scoped_client if the token expires it should be refreshed which doesn't happen now (the initial problem) | 23:51 |
| jamielennox | ok, cya | 23:51 |
| *** carl_baldwin has quit IRC | 23:52 | |
| *** jhesketh has quit IRC | 23:53 | |
| ayoung | jamielennox, it shouldn't be able to be refereshed directly | 23:53 |
| *** willingc has joined #openstack-dev | 23:53 | |
| *** sn0wb1rd has joined #openstack-dev | 23:53 | |
| ayoung | you need to authenticate again | 23:53 |
| *** rcleere has quit IRC | 23:54 | |
| ayoung | use client1 which has uid/pw to re-execute the trust. If all you have is a token, you can't extend expiry | 23:54 |
| *** rnirmal has quit IRC | 23:54 | |
| *** pete5 has quit IRC | 23:54 | |
| *** danwent has quit IRC | 23:54 | |
| ayoung | jamielennox, anyway, if you can code up a python sample that shows this, the world will thank you. Blog post is appropriate | 23:55 |
| *** rongze has quit IRC | 23:56 | |
| *** michchap has quit IRC | 23:58 | |
| *** michchap has joined #openstack-dev | 23:58 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!