* DHE still votes for Unix listen sockets for proxy server as a feature (patch is abandoned on gerrit) | 00:02 | |
*** itlinux has joined #openstack-swift | 00:11 | |
*** itlinux has quit IRC | 00:11 | |
*** mikecmpbll has quit IRC | 00:18 | |
*** gyee has quit IRC | 01:08 | |
*** itlinux has joined #openstack-swift | 02:18 | |
*** itlinux has quit IRC | 04:11 | |
*** itlinux has joined #openstack-swift | 04:43 | |
*** itlinux has quit IRC | 05:11 | |
*** ccamacho has quit IRC | 06:28 | |
*** rcernin has quit IRC | 07:09 | |
*** pcaruana has joined #openstack-swift | 07:12 | |
*** ccamacho has joined #openstack-swift | 08:22 | |
*** godog has joined #openstack-swift | 08:22 | |
*** alecuyer has quit IRC | 08:40 | |
*** alecuyer has joined #openstack-swift | 08:41 | |
*** rledisez has joined #openstack-swift | 08:41 | |
*** mikecmpbll has joined #openstack-swift | 09:02 | |
*** hseipp has joined #openstack-swift | 09:11 | |
*** hseipp has quit IRC | 09:33 | |
*** e0ne has joined #openstack-swift | 09:41 | |
*** hseipp has joined #openstack-swift | 09:43 | |
*** hseipp has quit IRC | 10:22 | |
*** mvkr has quit IRC | 10:24 | |
*** mvkr has joined #openstack-swift | 10:56 | |
*** mvkr has quit IRC | 11:16 | |
*** mvkr has joined #openstack-swift | 11:16 | |
*** NM has joined #openstack-swift | 11:36 | |
*** mikecmpb_ has joined #openstack-swift | 12:00 | |
*** mikecmpbll has quit IRC | 12:02 | |
openstackgerrit | Cyril Roelandt proposed openstack/swift master: Py3: Add working tests to tox.ini https://review.openstack.org/624792 | 12:05 |
---|---|---|
*** hseipp has joined #openstack-swift | 12:32 | |
*** e0ne has quit IRC | 12:51 | |
*** e0ne has joined #openstack-swift | 13:35 | |
*** e0ne has quit IRC | 13:49 | |
*** irclogbot_2 has quit IRC | 13:50 | |
*** NM has quit IRC | 13:56 | |
*** NM has joined #openstack-swift | 13:57 | |
*** e0ne has joined #openstack-swift | 13:57 | |
*** irclogbot_2 has joined #openstack-swift | 14:12 | |
*** irclogbot_2 has quit IRC | 14:19 | |
*** irclogbot_2 has joined #openstack-swift | 14:34 | |
openstackgerrit | Cyril Roelandt proposed openstack/swift master: Py3: Add working tests to tox.ini https://review.openstack.org/624792 | 14:58 |
*** itlinux has joined #openstack-swift | 15:07 | |
openstackgerrit | Cyril Roelandt proposed openstack/swift master: Py3: Add working tests to tox.ini https://review.openstack.org/624792 | 15:43 |
*** ccamacho has quit IRC | 15:52 | |
timburke | kota_: thanks for the pointer on https://review.openstack.org/#/c/610747/ ! you're right, of course. will re-spin that today; i've got some users complaining about not being able to do multipart uploads :-( | 16:27 |
patchbot | patch 610747 - swift - s3api: Stop relying on container listings during m... - 4 patch sets | 16:27 |
timburke | note that https://review.openstack.org/#/c/609860/ is a prerequisite, though | 16:27 |
patchbot | patch 609860 - swift - Add slo_manifest_hook callback - 4 patch sets | 16:27 |
*** e0ne has quit IRC | 16:30 | |
*** e0ne has joined #openstack-swift | 16:30 | |
notmyname | good morning | 16:55 |
notmyname | timburke: perfect updates to the priority review page. thanks | 17:01 |
timburke | should maybe mention https://review.openstack.org/#/c/624516/ and https://review.openstack.org/#/c/624527/, too? meh | 17:02 |
patchbot | patch 624516 - swift (stable/queens) - copy: Stop hanging per-request data on middleware ... - 1 patch set | 17:02 |
patchbot | patch 624527 - swift (stable/pike) - copy: Stop hanging per-request data on middleware ... - 1 patch set | 17:02 |
notmyname | yeah, good to track them there | 17:03 |
timburke | i should also consider backporting the pike patch to ocata, since i've already done the work for post-as-copy | 17:03 |
notmyname | my gut reaction is to *not* port back to post-as-copy | 17:04 |
timburke | too laaaaaaate | 17:04 |
*** notkevin_ has joined #openstack-swift | 17:22 | |
*** hseipp has quit IRC | 17:23 | |
notkevin_ | I am having some issue with getting the s3 middleware working on Queens. When I make the request using Boto, I get an http 403 error. When I look on the proxy server logs I see "You did not specify an auth or a storage token. skipping." I am unsure if this is an error with how I am trying to access swift via the boto client or if it is an issue with a response from keystone | 17:26 |
notmyname | notkevin_: can you put some (redacted?) logs on http://paste.openstack.org ? | 17:27 |
notkevin_ | I hope I redacted everything properly: http://paste.openstack.org/show/737239/ | 17:32 |
notkevin_ | Thank you! | 17:32 |
*** mikecmpb_ has quit IRC | 17:33 | |
notmyname | timburke is our resident s3api expert. I'm looking, but he may see something faster | 17:40 |
notmyname | but yeah, it does seem to be an issue with keystone maybe? | 17:40 |
*** ianychoi has quit IRC | 17:42 | |
notkevin_ | Ok, thank you. Someone else manages our keystone server and I understand more recent versions changed how the ec2 tokens were handled | 17:52 |
notmyname | timburke: do you have any ideas here? | 17:52 |
notkevin_ | I see in the code for s3_token.py that it wants to access https://keystone/v2.0/s3tokens, but I get a 404 error when I try to access that url | 17:54 |
notkevin_ | I wasn't sure if it was an issue with keystone deprecating v2.0 auth | 17:54 |
notmyname | yeah, there was something that changed with that recently | 17:55 |
timburke | try moving s3token to be *after* auth_token (but still before keystoneauth) in the pipeline | 17:57 |
timburke | v3 responses don't include an actual token id anymore (for good reason), and auth_token will strip out the headers s3token adds to say "hey, i know who this is!" | 17:58 |
timburke | at least, that's my first guess for what's going on; i know i've seen that before | 18:00 |
timburke | what version of the middleware are you using? | 18:00 |
notkevin_ | swift3-1.12 | 18:02 |
notkevin_ | Same error with pipeline as pipeline = catch_errors gatekeeper healthcheck proxy-logging cache bulk swift3 authtoken s3token keystone staticweb slo dlo versioned_writes proxy-logging proxy-server | 18:03 |
timburke | huh. judging by the "skipping" message, i figured it would predate https://github.com/openstack/swift3/commit/cd094ee -- but that shoulda been included in 1.12 | 18:03 |
timburke | so is s3token coming from swift3, too, of is that coming from keystonemiddleware? | 18:04 |
notkevin_ | I am not sure I understand the question. I have the rpms for swift3 and keystonemiddleware installed | 18:08 |
*** gyee has joined #openstack-swift | 18:09 | |
timburke | in the [filter:s3token] proxy-server.conf section, there should be a use or paste.filter_factory line that says where the code for that middleware comes from. s3token used to be in keystone, then keystoneclient, then keystonemiddleware, then swift3, and now it lives in swift (along with all the rest of swift3) | 18:10 |
*** e0ne has quit IRC | 18:11 | |
notkevin_ | Ok, the line is paste.filter_factory = keystonemiddleware.s3_token:filter_factory | 18:11 |
notkevin_ | We are using puppet-swift to manage the config, so I believe that is what it is putting in there | 18:12 |
timburke | ah, makes sense. try switching it to `use = egg:swift3#s3token` | 18:13 |
*** e0ne has joined #openstack-swift | 18:15 | |
*** e0ne has quit IRC | 18:16 | |
timburke | when you get to rocky (or later), you'll want to change it again, to point to swift. iirc there weren't many changes to it for rocky, though -- but there'll be some nice improvements in stein | 18:16 |
notmyname | yep, i'm working ont he changlog for 2.20.0 now | 18:16 |
notkevin_ | Ok thank you! | 18:19 |
notkevin_ | It isn't working but I got this errors "Keystone reply error: status=404 reason=Not Found (txn: tx34b0ee700ab746f798e82-005c12a268)" now I can go talk to our keystone admin :) | 18:19 |
timburke | yay! :P | 18:20 |
notkevin_ | And ill have to work on updating puppet | 18:20 |
notmyname | I knew timburke would know the answers! :-) | 18:20 |
notkevin_ | You guys always know the answers! Thank you! | 18:21 |
notmyname | rledisez: you're still interested in https://review.openstack.org/#/c/613987/? can you rebase? | 18:25 |
patchbot | patch 613987 - swift - SSYNC: enable multiple SSYNC connections per job - 2 patch sets | 18:25 |
timburke | notkevin_: proposed //review.openstack.org/625080 | 18:27 |
timburke | er, https://review.openstack.org/625080 | 18:27 |
patchbot | patch 625080 - puppet-swift (stable/queens) - Use s3token proveded by swift3 instead of keystone... - 1 patch set | 18:27 |
notkevin_ | Awesome, thank you! | 18:29 |
*** mvkr has quit IRC | 18:40 | |
*** mikecmpbll has joined #openstack-swift | 18:42 | |
openstackgerrit | John Dickinson proposed openstack/swift master: added value and notes to a sample config file for s3token https://review.openstack.org/625082 | 18:46 |
notmyname | sorrison: ^^ that patch may be of particular interest to you | 18:53 |
*** mvkr has joined #openstack-swift | 19:21 | |
openstackgerrit | Merged openstack/swift master: Py3: Add working tests to tox.ini https://review.openstack.org/624792 | 19:23 |
*** pcaruana has quit IRC | 19:47 | |
*** e0ne has joined #openstack-swift | 19:53 | |
*** e0ne has quit IRC | 19:58 | |
openstackgerrit | Merged openstack/swift master: Python3: fix test_xprofile.py https://review.openstack.org/624791 | 20:33 |
*** psachin has joined #openstack-swift | 20:52 | |
zaitcev | Hmm | 20:54 |
zaitcev | The 624791 passed the CI, but now my tox -e py35 throws in test.unit.common.middleware.test_xprofile.TestProfileMiddleware | 20:55 |
*** ianychoi has joined #openstack-swift | 20:59 | |
*** notkevin_ has quit IRC | 21:05 | |
*** NM has quit IRC | 21:05 | |
openstackgerrit | John Dickinson proposed openstack/swift master: authors/changelog for 2.20.0 release https://review.openstack.org/625116 | 21:09 |
notmyname | first draft of changelog; no renofication yet ^^ | 21:09 |
notmyname | tdasilva: FYI I changed your reported email to gmail instead of redhat in that patch | 21:09 |
tdasilva | notmyname: thanks! | 21:09 |
timburke | zaitcev: still not sure we need ordx -- did you have a chance to try out http://paste.openstack.org/show/736602/? | 21:09 |
timburke | and what do you think about having Request.params populated with WSGI strings? agree that we can't keep decoding as utf8, but that seems like a not-crazy middle ground? probably need to fix up the setter, too... | 21:13 |
*** psachin has quit IRC | 21:27 | |
openstackgerrit | John Dickinson proposed openstack/swift master: authors/changelog for 2.20.0 release https://review.openstack.org/625116 | 21:58 |
DHE | oh cool... | 22:16 |
mattoliverau | morning | 22:18 |
*** rcernin has joined #openstack-swift | 22:21 | |
zaitcev | timburke: I'm working on that ordx. Just did a rebase in 613505 and stepped on the problem that I mentioned. | 22:32 |
*** itlinux has quit IRC | 22:36 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!