| *** praneshp_ has joined #openstack-oslo | 00:00 | |
| *** wendar has joined #openstack-oslo | 00:01 | |
| *** sigmavirus24b has joined #openstack-oslo | 00:01 | |
| *** gus_ has joined #openstack-oslo | 00:06 | |
| *** russellb_ has joined #openstack-oslo | 00:06 | |
| *** jraim_ has joined #openstack-oslo | 00:07 | |
| *** hartsocks has joined #openstack-oslo | 00:08 | |
| *** hartsocks is now known as Guest32290 | 00:08 | |
| *** mgagne has joined #openstack-oslo | 00:08 | |
| *** mgagne is now known as Guest72739 | 00:09 | |
| *** jroll|dupe has joined #openstack-oslo | 00:09 | |
| *** jraim has quit IRC | 00:13 | |
| *** jecarey has quit IRC | 00:13 | |
| *** praneshp has quit IRC | 00:13 | |
| *** AAzza_afk has quit IRC | 00:13 | |
| *** wendar_ has quit IRC | 00:13 | |
| *** sileht has quit IRC | 00:13 | |
| *** sigmavirus24_awa has quit IRC | 00:13 | |
| *** jroll has quit IRC | 00:13 | |
| *** hartbot has quit IRC | 00:13 | |
| *** Guest6936 has quit IRC | 00:13 | |
| *** russellb has quit IRC | 00:13 | |
| *** gus has quit IRC | 00:13 | |
| *** jroll|dupe is now known as jroll | 00:13 | |
| *** jraim_ is now known as jraim | 00:13 | |
| *** praneshp_ is now known as praneshp | 00:13 | |
| *** sileht has joined #openstack-oslo | 00:15 | |
| *** AAzza has joined #openstack-oslo | 00:19 | |
| *** AAzza_afk has joined #openstack-oslo | 00:29 | |
| *** gus_ is now known as gus | 00:29 | |
| *** AAzza has quit IRC | 00:29 | |
| *** AAzza_afk is now known as AAzza | 00:29 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 00:29 |
|---|---|---|
| *** AAzza_afk has joined #openstack-oslo | 00:32 | |
| *** AAzza has quit IRC | 00:37 | |
| *** AAzza_afk is now known as AAzza | 00:37 | |
| *** HenryG has quit IRC | 00:37 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 00:42 |
| *** harlowja_at_home has quit IRC | 00:46 | |
| *** dimsum_ has quit IRC | 00:57 | |
| *** dimsum_ has joined #openstack-oslo | 01:13 | |
| *** mriedem has left #openstack-oslo | 01:23 | |
| *** mriedem has quit IRC | 01:23 | |
| *** noelbk has joined #openstack-oslo | 01:43 | |
| *** dimsum_ has quit IRC | 01:51 | |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 02:17 | |
| openstackgerrit | lifeless proposed a change to openstack-dev/pbr: Only consider tags that look like versions. https://review.openstack.org/114403 | 02:36 |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 02:39 | |
| *** harlowja_at_home has joined #openstack-oslo | 03:08 | |
| *** harlowja_at_home has quit IRC | 03:24 | |
| *** harlowja_at_home has joined #openstack-oslo | 03:27 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 03:32 |
| *** praneshp_ has joined #openstack-oslo | 04:12 | |
| *** praneshp has quit IRC | 04:14 | |
| *** praneshp_ is now known as praneshp | 04:14 | |
| *** liusheng has quit IRC | 04:31 | |
| *** liusheng has joined #openstack-oslo | 04:31 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 04:42 |
| *** jraim has quit IRC | 05:10 | |
| *** jraim_ has joined #openstack-oslo | 05:12 | |
| *** k4n0 has joined #openstack-oslo | 05:57 | |
| *** ildikov has joined #openstack-oslo | 06:21 | |
| *** palendae has quit IRC | 06:23 | |
| *** palendae has joined #openstack-oslo | 06:25 | |
| *** flaper87|afk is now known as flaper87 | 06:26 | |
| *** vigneshvar has joined #openstack-oslo | 06:27 | |
| openstackgerrit | Kevin Benton proposed a change to openstack/oslo.db: Provide better details for execution failure https://review.openstack.org/121452 | 06:46 |
| *** ihrachyshka has joined #openstack-oslo | 07:25 | |
| openstackgerrit | Peng Wu proposed a change to openstack/oslo-specs: Add spec for more-gettext-support https://review.openstack.org/121459 | 07:25 |
| *** ihrachyshka has quit IRC | 07:29 | |
| *** andreykurilin_ has joined #openstack-oslo | 07:31 | |
| *** vigneshvar has quit IRC | 07:41 | |
| *** andreykurilin_ has quit IRC | 07:49 | |
| *** praneshp_ has joined #openstack-oslo | 07:52 | |
| *** praneshp has quit IRC | 07:55 | |
| *** praneshp_ is now known as praneshp | 07:55 | |
| *** ihrachyshka has joined #openstack-oslo | 07:59 | |
| *** sileht has quit IRC | 08:05 | |
| *** jraim_ has quit IRC | 08:05 | |
| *** jraim_ has joined #openstack-oslo | 08:05 | |
| *** k4n0 has quit IRC | 08:05 | |
| *** k4n0 has joined #openstack-oslo | 08:05 | |
| *** ajo_ is now known as ajo | 08:07 | |
| *** pblaho has joined #openstack-oslo | 08:16 | |
| *** sileht has joined #openstack-oslo | 08:18 | |
| *** ihrachyshka_ has joined #openstack-oslo | 08:20 | |
| *** ihrachyshka has quit IRC | 08:21 | |
| *** viktors|afk is now known as viktors | 08:31 | |
| *** i159 has joined #openstack-oslo | 08:40 | |
| *** k4n0 has quit IRC | 08:48 | |
| *** ildikov has quit IRC | 08:50 | |
| *** k4n0 has joined #openstack-oslo | 09:07 | |
| openstackgerrit | James Page proposed a change to openstack/oslo.serialization: Support 'built-in' datetime module https://review.openstack.org/121476 | 09:09 |
| openstackgerrit | James Page proposed a change to openstack/oslo.serialization: Support 'built-in' datetime module https://review.openstack.org/121476 | 09:12 |
| *** ihrachyshka_ has quit IRC | 09:15 | |
| *** sileht has quit IRC | 09:19 | |
| i159 | jd__, dhellmann, noelbk, mikal, sileht, haypo, russellb_, bnemec, zyluo, markmc, flaper87: Hi, guys! You had reviewed this patch https://review.openstack.org/110058/, which I'm maintaining now. But you did not put any votes and comments. Please give me some feedback about Stan's work, I need to know how to continue this work. | 09:25 |
| *** sileht has joined #openstack-oslo | 09:33 | |
| *** myatsenko has joined #openstack-oslo | 09:54 | |
| myatsenko | test | 09:55 |
| *** pblaho is now known as pblaho|afk | 09:57 | |
| *** ihrachyshka_ has joined #openstack-oslo | 10:00 | |
| *** markmc has joined #openstack-oslo | 10:07 | |
| *** praneshp has quit IRC | 10:07 | |
| *** abhijeetm has joined #openstack-oslo | 10:15 | |
| openstackgerrit | A change was merged to openstack/oslo.db: Add a check for SQLite transactional state https://review.openstack.org/120148 | 10:18 |
| *** vigneshvar has joined #openstack-oslo | 10:23 | |
| *** pcm_ has joined #openstack-oslo | 10:23 | |
| *** pcm_ has quit IRC | 10:23 | |
| *** pcm_ has joined #openstack-oslo | 10:24 | |
| *** Krast has joined #openstack-oslo | 10:34 | |
| *** Krast has quit IRC | 10:35 | |
| *** tsufiev_ is now known as tsufiev | 10:36 | |
| *** HenryG has joined #openstack-oslo | 10:37 | |
| *** k4n0 has quit IRC | 10:42 | |
| openstackgerrit | Tim Kelsey proposed a change to openstack/oslo.messaging: Adds message security plugins https://review.openstack.org/109806 | 10:43 |
| *** pblaho|afk is now known as pblaho | 10:49 | |
| *** k4n0 has joined #openstack-oslo | 11:03 | |
| *** pblaho is now known as pblaho|afk | 11:08 | |
| *** vigneshvar has quit IRC | 11:09 | |
| *** dimsum_ has joined #openstack-oslo | 11:12 | |
| *** pblaho|afk is now known as pblaho | 11:13 | |
| *** vigneshvar has joined #openstack-oslo | 11:20 | |
| *** dims_ has joined #openstack-oslo | 11:24 | |
| *** dimsum_ has quit IRC | 11:24 | |
| openstackgerrit | Davanum Srinivas (dims) proposed a change to openstack/oslo.vmware: Add a memory based shared cache https://review.openstack.org/116297 | 11:28 |
| *** _amrith_ is now known as amrith | 11:31 | |
| *** dims_ has quit IRC | 11:33 | |
| *** dims has joined #openstack-oslo | 11:33 | |
| openstackgerrit | Ann Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync:add correct server_default check for Enum https://review.openstack.org/120746 | 11:47 |
| *** tedross_ has joined #openstack-oslo | 11:50 | |
| *** jgrimm is now known as zz_jgrimm | 11:51 | |
| *** kgiusti has joined #openstack-oslo | 12:09 | |
| openstackgerrit | Ann Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys https://review.openstack.org/116238 | 12:13 |
| *** pblaho is now known as pblaho|afk | 12:15 | |
| *** pblaho|afk is now known as pblaho | 12:17 | |
| *** dims has quit IRC | 12:18 | |
| *** dims has joined #openstack-oslo | 12:19 | |
| *** ajo has quit IRC | 12:21 | |
| *** ajo has joined #openstack-oslo | 12:36 | |
| *** abhijeetm has quit IRC | 12:40 | |
| *** zz_jgrimm is now known as jgrimm | 12:41 | |
| *** jeckersb_gone is now known as jeckersb | 12:41 | |
| *** gordc has joined #openstack-oslo | 12:43 | |
| ttx | dhellmann: posted an oslo_release.sh utility at: https://review.openstack.org/#/c/121527/ | 12:48 |
| *** Guest32290 is now known as hartsocks | 12:49 | |
| ttx | dhellmann: the idea would be to mark bugs FixReleased when you cut an alpha, but keep them in next-juno. Then when you cut a final tag, we rename the next-juno to the real version | 12:49 |
| ttx | the script helps you do both | 12:49 |
| *** markmc has left #openstack-oslo | 12:51 | |
| *** ajo has quit IRC | 12:52 | |
| *** ajo has joined #openstack-oslo | 12:53 | |
| viktors | dhellmann: hI! Please ping me, when you'll become online | 12:56 |
| *** russellb_ is now known as russellb | 12:59 | |
| *** ajo has quit IRC | 13:00 | |
| *** ajo has joined #openstack-oslo | 13:00 | |
| *** tedross_ has quit IRC | 13:02 | |
| *** ajo has quit IRC | 13:02 | |
| *** ajo has joined #openstack-oslo | 13:03 | |
| *** ajo has quit IRC | 13:05 | |
| *** ajo has joined #openstack-oslo | 13:05 | |
| *** pblaho has quit IRC | 13:06 | |
| *** pblaho has joined #openstack-oslo | 13:06 | |
| *** ihrachyshka_ is now known as ihrachyshka | 13:11 | |
| openstackgerrit | Julien Danjou proposed a change to openstack/oslo-incubator: rpc, notifier: remove deprecated modules https://review.openstack.org/101629 | 13:11 |
| *** k4n0 has quit IRC | 13:13 | |
| *** bknudson has quit IRC | 13:15 | |
| *** jeckersb is now known as jeckersb_gone | 13:17 | |
| *** tedross_ has joined #openstack-oslo | 13:18 | |
| *** pblaho has quit IRC | 13:25 | |
| *** jeckersb_gone is now known as jeckersb | 13:26 | |
| *** sigmavirus24b is now known as sigmavirus24 | 13:31 | |
| *** sigmavirus24 has joined #openstack-oslo | 13:31 | |
| *** mriedem has joined #openstack-oslo | 13:32 | |
| dhellmann | ttx: I guess we don't want a separate milestone for each alpha? | 13:42 |
| dhellmann | ttx: I need to think about how we're going to do planning beyond one milestone if we're only going to have a single "next" milestone for the whole cycle | 13:42 |
| dhellmann | ttx: we did use the j1, j2, etc. milestones this cycle, for example | 13:43 |
| dhellmann | viktors: hi, what's up | 13:43 |
| dhellmann | i159: I'm a little uncomfortable introducing such a big change in behavior this late in the cycle. How critical is that bug? Can we make this a k-1 change? | 13:44 |
| viktors | dhellmann: hi! I supposed to release a new version of oslo.db (finally!) and I just wanted to ensure a new version number | 13:46 |
| dhellmann | viktors: how much testing have you done to ensure this release won't cause issues with unit tests? I see a lot of changes ready for release | 13:48 |
| viktors | dhellmann: I've tested on cinder, glance, heat, ironic, keystone, neutron, nova and sahara. All unittest passes ok wit oslo.db's master | 13:50 |
| viktors | at least, it seems to be :) | 13:50 |
| dhellmann | viktors: ok, that seems promising | 13:51 |
| viktors | dhellmann: :) | 13:51 |
| *** bknudson has joined #openstack-oslo | 13:52 | |
| viktors | dhellmann: so I wanted to clarify, should this be version 0.5.0 or 1.0.0 ? | 13:53 |
| dhellmann | viktors: we were planning to tag 1.0 on thursday, but if you think you're ready you can go ahead today | 13:54 |
| ttx | dhellmann: If you do multiple releases in one cycle, you just recreate a next-juno after tagging a release | 13:54 |
| ttx | or you rename after-next-juno to next-juno | 13:54 |
| viktors | dhellmann: I supposed to do it tomorrow morning (by UTC+3) | 13:55 |
| ttx | next-juno becomes a floating ref to "the next release of the lib" | 13:55 |
| ttx | at the end of a cycle there wouldn't be a next-juno, obviously | 13:55 |
| viktors | dhellmann: about your documentat `Oslo Library Audit` - should I do any actions before release? | 13:55 |
| dhellmann | ttx: right, that's how i was thinking of it working, so having the script always rename the milestone in lp would be good | 13:56 |
| dhellmann | ttx: having it make another next-foo tag would be nice, too, if that's scriptable | 13:56 |
| ttx | dhellmann: the script could even optionally recreate next-juno or rename it from after-next-juno | 13:56 |
| dhellmann | viktors: we'll need to make sure those tasks are finished in kilo | 13:56 |
| ttx | yes it is | 13:56 |
| viktors | dhellmann: got it, thanks | 13:57 |
| ttx | dhellmann: for this round we can just manually handle it, since there won't be a next-juno to recreate | 13:57 |
| dhellmann | ttx: how about the foo-1, foo-2, etc. milestones? should we just not use those? | 13:57 |
| dhellmann | ttx: sure, I'm thinking ahead for how we want to use this in kilo | 13:57 |
| ttx | dhellmann: what would you use those for ? You would create a release after each milestone ? | 13:58 |
| dhellmann | ttx: no, just for scheduling | 13:58 |
| dhellmann | ttx: we still want people to be thinking about those deadlines, because frequently if we don't land a feature in a lib by the right time in a cycle, it isn't going to be used that cycle | 13:59 |
| ttx | dhellmann: hmm | 13:59 |
| ttx | dhellmann: so the trick is, foo-N milestones wouldn't worrk so well if you want to be able to release mid-cycle | 14:00 |
| dhellmann | yeah | 14:00 |
| *** jroll has quit IRC | 14:00 | |
| *** jroll has joined #openstack-oslo | 14:00 | |
| ttx | I prefer the flexibility that next-SERIES gives us | 14:00 |
| ttx | but if you can stick to one-lib-release-at-the-end... then sure | 14:00 |
| ttx | splitting the work to get there into sub-milestones could work | 14:00 |
| ttx | we would have to consolidate them at the end of cycle like we do for integrated release | 14:01 |
| ttx | dhellmann: it really depends if we want to be able to release mid-cycle | 14:01 |
| dhellmann | we definitely need that | 14:01 |
| ttx | then mixing the two systems is a bit weird | 14:02 |
| dhellmann | right now the only way for developer unit tests to use mid-cycle changes to the library is to have a release | 14:02 |
| ttx | maybe doable, but weird | 14:02 |
| dhellmann | yeah | 14:02 |
| ttx | we'll have to think about it | 14:02 |
| ttx | for juno, that script should cover you | 14:02 |
| ttx | we can test it on oslo.rootwrap | 14:02 |
| ttx | as I think it's ready for a new alpha tag which would be the final one | 14:03 |
| dhellmann | how much do we care to have milestones in lp that use the alpha version numbers? | 14:03 |
| ttx | dhellmann: I don't think there is so much value in it | 14:03 |
| ttx | what you need to know is if the fix was released in an alpha (FixReleased) | 14:03 |
| ttx | and in which final release it ended up | 14:04 |
| ttx | the proposed process conveys both pieces of information | 14:04 |
| ttx | dhellmann: got to jum in a meeting, ttyl | 14:04 |
| dhellmann | ok, so then what if we modify the script to take an argument for the milestone (foo-N) and whether or not to close it (defaulting to false) | 14:04 |
| dhellmann | that way during the run up to foo-N, we can release a bunch of separate times, marking bugs as fix released as we go, and at the end of foo-N we can close it | 14:04 |
| dhellmann | ttx; ok, ^^ | 14:05 |
| ttx | dhellmann: but what if you don't do a final release within kilo-1 ? | 14:06 |
| ttx | then you end up with a milestone you need to move all stuff from | 14:06 |
| dhellmann | ttx: then I'll just manually close the milestone | 14:06 |
| ttx | but you ultimately want those BPs and bugs to appear on the milestone page for the release | 14:07 |
| ttx | so that means moving them | 14:07 |
| dhellmann | yes, true | 14:07 |
| ttx | which used to be funny, but with LP timeouts, not so much | 14:07 |
| ttx | i'm trying to avoid this hell | 14:07 |
| dhellmann | yeah | 14:07 |
| dhellmann | ttx: I think just using next milestones is going to make that worse, because it gives us no place to put "future" work except the one open milestone | 14:08 |
| ttx | dhellmann: will have to think about it more | 14:09 |
| ttx | ttyl | 14:09 |
| dhellmann | ttx: me, too | 14:09 |
| *** zzzeek has joined #openstack-oslo | 14:10 | |
| *** jecarey_ has quit IRC | 14:10 | |
| *** cfriesen_ has joined #openstack-oslo | 14:11 | |
| *** yamahata has joined #openstack-oslo | 14:18 | |
| *** jecarey has joined #openstack-oslo | 14:21 | |
| openstackgerrit | Doug Hellmann proposed a change to openstack/oslo-incubator: Do not try to show changes of non-master branches https://review.openstack.org/121584 | 14:25 |
| *** beekneemech is now known as bnemec | 14:26 | |
| dhellmann | ttx: were you going to tag another rootwrap alpha today? | 14:29 |
| openstackgerrit | James Page proposed a change to openstack/oslo.messaging: Enable heartbeating on rabbitmq connections https://review.openstack.org/94656 | 14:37 |
| *** tedross_ has quit IRC | 14:42 | |
| dhellmann | jd__: I'm considering leaving https://review.openstack.org/#/c/105746/ out of the 1.4 release for oslo.messaging, since it's not in a published release yet and so isn't likely to be used by anyone. Thoughts? | 14:47 |
| *** dims_ has joined #openstack-oslo | 14:56 | |
| ttx | dhellmann: yes, I was considering to. The returncode patch won't be ready this week, I think, as it's non-trivial to test, and I don't really want to touch those at this point | 14:58 |
| ttx | so I deferred it to kilo-1 | 14:58 |
| ttx | or next-kilo | 14:58 |
| *** dims__ has joined #openstack-oslo | 14:58 | |
| ttx | so I was considering trying my script | 14:58 |
| *** tsekiyama has joined #openstack-oslo | 14:59 | |
| *** dims has quit IRC | 14:59 | |
| *** dims_ has quit IRC | 15:01 | |
| *** jraim_ has quit IRC | 15:04 | |
| gordc | dims, dhellmann: just an fyi, i cut the 0.1.0 release of oslo.middleware this morning | 15:05 |
| gordc | https://pypi.python.org/pypi/oslo.middleware | 15:06 |
| dhellmann | ttx: ok | 15:06 |
| *** jraim has joined #openstack-oslo | 15:06 | |
| dhellmann | ttx: I did a couple of other alpha releases by hand, to get us ready for thursday | 15:06 |
| dhellmann | gordc: cool, when you announce that make sure it's clear we want people to be planning to adopt for kilo | 15:07 |
| gordc | dhellmann: ok. will do... i'll send a note to list once i update global-req | 15:07 |
| *** vigneshvar has quit IRC | 15:08 | |
| dhellmann | gordc: because of there requirements freeze, we probably won't add that until after juno is done | 15:09 |
| gordc | dhellmann: i see... do you recommend i send note even if not included in requirements? | 15:09 |
| dhellmann | gordc: yes, you've released it, so you should go ahead and announce it | 15:10 |
| gordc | dhellmann: sure thing.. i'll draft it up. | 15:10 |
| ttx | dhellmann: will tag tomorrow morning | 15:12 |
| jd__ | dhellmann: I wouldn't mind | 15:13 |
| dhellmann | jd__: ok, thanks | 15:13 |
| dhellmann | ttx: sounds good | 15:13 |
| dhellmann | zzzeek: can you fill me in on the status of https://bugs.launchpad.net/oslo.db/+bug/1367354 please? | 15:19 |
| zzzeek | dhellmann: well if https://review.openstack.org/#/c/120148/ is merged then i think that resolves | 15:20 |
| dhellmann | zzzeek: ok | 15:20 |
| zzzeek | we also need https://review.openstack.org/#/c/120146/ for keystone | 15:21 |
| zzzeek | also merged | 15:21 |
| dhellmann | zzzeek: I'll add those projects to the bug | 15:21 |
| *** markmcclain has joined #openstack-oslo | 15:22 | |
| *** markmcclain has left #openstack-oslo | 15:23 | |
| *** praneshp has joined #openstack-oslo | 15:23 | |
| dhellmann | harlowja_at_home: please let me know when you have a few minutes to talk about taskflow bugs on https://launchpad.net/oslo/+milestone/next-juno | 15:23 |
| *** zzzeek_ has joined #openstack-oslo | 15:24 | |
| *** zzzeek has quit IRC | 15:24 | |
| *** zzzeek_ is now known as zzzeek | 15:24 | |
| jecarey | dhellmann, cinder is now hitting the mask_passwork coercion with six.text_type failure (https://bugs.launchpad.net/oslo.utils/+bug/1366189). I added more analysis ... would like to discuss when you have time. | 15:25 |
| *** viktors is now known as viktors|afk | 15:26 | |
| dhellmann | jecarey: looking | 15:26 |
| *** harlowja_at_home has quit IRC | 15:27 | |
| dhellmann | jecarey: do you have any idea why we think we need to convert to unicode ourselves before using re? | 15:30 |
| jecarey | The problem is if it is a i18n._Message object | 15:30 |
| jecarey | the re will be against the string it is holding that is the default translation (with replacement) ... so if subsequently translated that gets ignored. | 15:31 |
| jecarey | Could add a check just that particular case ... exposes i18n._Message object though. | 15:32 |
| dhellmann | how about "if not isinstance()" as you propose in the bug? | 15:33 |
| dhellmann | if we have a message object and we convert it to unicode, we shouldn't need the encoding | 15:33 |
| dhellmann | text_type() should map to unicode() which doesn't have an encoding | 15:33 |
| dhellmann | jecarey: do you have a traceback in a pastebin somewhere? | 15:34 |
| jecarey | dhellmann, cinder bug has one: https://bugs.launchpad.net/cinder/+bug/1368527 | 15:34 |
| dhellmann | ugh, they're still using the incubator version of this code? | 15:35 |
| dhellmann | oh, it's a decode error, not an encode error | 15:36 |
| dhellmann | misread that before | 15:36 |
| dhellmann | ok, that does make more sense | 15:36 |
| *** ihrachyshka has quit IRC | 15:37 | |
| dhellmann | jecarey: so it looks like processutils is passing us a byte string instead of decoding it first | 15:39 |
| jecarey | dhellmann, right ... should we add decoding there? | 15:40 |
| dhellmann | probably, but that feels like a big change to be making this late | 15:40 |
| dhellmann | jecarey: how about if we make mask_password() trap the decode error and work on the byte string? | 15:41 |
| dhellmann | jecarey: and fix it properly in kilo, when we have more time to deal with fallout from the bigger change to process utils | 15:41 |
| jecarey | dhellmann, ok ... so I can just eat the decode exception and use message as is. | 15:42 |
| dhellmann | jecarey: do you have a real case where we are passing a message to this function? | 15:42 |
| dhellmann | jecarey: nevermind, I thought you meant _Message but the arg name is message | 15:43 |
| i159 | dhellmann: I don't think it is critical at all, because it has status "undecided". Actually I just want to have discussion about it to be more involved. | 15:43 |
| dhellmann | confusion reigns | 15:43 |
| dhellmann | i159: ok, can it wait until next week? we're trying to focus on release critical bugs this week | 15:44 |
| jecarey | dhellman, right. | 15:44 |
| dhellmann | our cut-off is thursday | 15:44 |
| i159 | dhellmann: sure! thanks | 15:44 |
| dhellmann | jecarey: yes, then, if there's a decode error try just using the message value as it's given | 15:44 |
| jecarey | ok ... I'll get a patch up right away. | 15:45 |
| dhellmann | jecarey: please include a comment that this is a temporary fix and not the best way to handle the problem and we'll be working on a better version for kilo | 15:45 |
| dhellmann | i159: thanks | 15:45 |
| jecarey | dhellmann, will do. | 15:45 |
| *** tedross_ has joined #openstack-oslo | 15:46 | |
| dhellmann | jecarey: I will open a separate bug to fix processutils in kilo | 15:47 |
| dhellmann | jecarey: actually, I'll just use this same bug but target it for kilo for oslo.concurrency | 15:47 |
| *** enykeev has joined #openstack-oslo | 15:48 | |
| jecarey | dhellmann, sounds good. | 15:48 |
| *** enykeev has quit IRC | 15:49 | |
| dhellmann | jecarey: I left a couple of comments in response to your last comment on that bug | 15:52 |
| *** i159 has quit IRC | 15:53 | |
| *** tongli has joined #openstack-oslo | 16:04 | |
| *** harlowja_at_home has joined #openstack-oslo | 16:08 | |
| *** electrichead is now known as redrobot | 16:08 | |
| *** yamahata has quit IRC | 16:12 | |
| *** jraim has quit IRC | 16:17 | |
| *** jraim has joined #openstack-oslo | 16:18 | |
| *** vigneshvar has joined #openstack-oslo | 16:28 | |
| *** praneshp_ has joined #openstack-oslo | 16:42 | |
| *** praneshp has quit IRC | 16:45 | |
| *** praneshp_ is now known as praneshp | 16:45 | |
| *** dims__ has quit IRC | 16:59 | |
| zzzeek | dhellmann: should we be getting kilo added to https://github.com/openstack/oslo-specs/tree/master/specs soon | 16:59 |
| *** dims has joined #openstack-oslo | 16:59 | |
| dhellmann | zzzeek: you could submit a patch to add it, but I don't think we're going to be looking at kilo specs before next week | 16:59 |
| zzzeek | dhellmann: OK, just wondering how to be useful til then… :) | 17:00 |
| zzzeek | i hve plenty of upstream work to do should proablby do that | 17:00 |
| dhellmann | zzzeek: I think there are some untriaged oslo.db bugs that could use some attention, too | 17:00 |
| dhellmann | even just triaging them would be useful | 17:00 |
| zzzeek | we hvae bugs, unbelivable | 17:00 |
| *** dims has quit IRC | 17:04 | |
| openstackgerrit | Doug Hellmann proposed a change to openstack/oslo-specs: Remove specs not completed in juno https://review.openstack.org/120095 | 17:04 |
| *** harlowja has joined #openstack-oslo | 17:04 | |
| *** dims has joined #openstack-oslo | 17:04 | |
| *** e0ne has joined #openstack-oslo | 17:06 | |
| *** vigneshvar has quit IRC | 17:12 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Just use the natural module name instead of aliasing to tt https://review.openstack.org/119186 | 17:19 |
| *** takedakn has joined #openstack-oslo | 17:19 | |
| openstackgerrit | James Carey proposed a change to openstack/oslo-incubator: Enabled mask_password to handle byte code strings https://review.openstack.org/121632 | 17:22 |
| *** vigneshvar has joined #openstack-oslo | 17:25 | |
| *** praneshp has quit IRC | 17:30 | |
| *** takedakn1 has joined #openstack-oslo | 17:31 | |
| *** takedakn has quit IRC | 17:33 | |
| *** bjornar_ has joined #openstack-oslo | 17:33 | |
| *** e0ne has quit IRC | 17:34 | |
| dhellmann | jecarey: do you have an oslo.utils patch for that in the works, too? | 17:48 |
| morganfainberg | zzzeek, i keep coming back around that i greatly dislike python-memcache and want to fork it / fix it / release it from stackforge | 17:50 |
| morganfainberg | zzzeek, once we're past RC i think i might do that. | 17:50 |
| zzzeek | interesting | 17:52 |
| morganfainberg | zzzeek, my thought is it would support the same interface as today (mostly) but be backed to a better interface that allowed for either thread.local or non-thread.local use, a pool construct, and py2/3 compat | 17:52 |
| morganfainberg | so drop-in or a better /propper setup. | 17:52 |
| harlowja | morganfainberg why not use https://pypi.python.org/pypi/pymemcache ?? | 17:53 |
| morganfainberg | harlowja, that was the fork, it's only py3 compat last i saw | 17:53 |
| morganfainberg | harlowja, or only 27 | 17:53 |
| harlowja | gotcha | 17:53 |
| morganfainberg | harlowja, one of those. it's like python + memcache is something hard to do :( | 17:54 |
| harlowja | https://pypi.python.org/pypi/pymemcache shows 26,27,33 | 17:54 |
| morganfainberg | harlowja, hm i'll need to take a look iirc there was some major issue with ti. | 17:54 |
| harlowja | hmmm k | 17:54 |
| harlowja | if u figure it out maybe can just ask them to fix it :-P | 17:54 |
| morganfainberg | harlowja, sure. i know there has been some serious lack of interest in fixing things when it comes to memcache libs :( | 17:55 |
| harlowja | hmmm, well let me know if u find the issue, i'll submit a pull request :-P | 17:55 |
| harlowja | as long as the issue isn't huge, lol | 17:55 |
| harlowja | which then i might defer, ha | 17:55 |
| morganfainberg | harlowja, it's (i think) because memcache largly was replaced in the industry by noSQL and datasets that don't fit in memcached's 1MB slab size (default) | 17:56 |
| harlowja | ok, so thats just an inherent problem to memcache | 17:56 |
| morganfainberg | yeah | 17:56 |
| openstackgerrit | Doug Hellmann proposed a change to openstack/oslo-incubator: Enabled mask_password to handle byte code strings https://review.openstack.org/121632 | 17:56 |
| morganfainberg | and i think thats why development on memcache has kindof stalled | 17:56 |
| morganfainberg | Redis has a similar IO profile (iirc) and less limitations | 17:56 |
| harlowja | depends on what its used for i guess, i've used it quite often in yahoo, although not in years | 17:56 |
| morganfainberg | and then people use mongo etc in the same manner | 17:57 |
| harlowja | used enough of it in my c++ days | 17:57 |
| morganfainberg | harlowja, exactly, most people "haven't used it in years". | 17:57 |
| harlowja | :-P | 17:57 |
| harlowja | well thats mainly cause i got sucked into openstack, lol | 17:57 |
| harlowja | i'm pretty sure its used elsewhere, as long as u don't try to shove > 1MB values in it (which is dumb) | 17:57 |
| morganfainberg | harlowja, a lot of people moved to redis (better overall) | 17:58 |
| harlowja | when u have to put ~1MB values in a cache, u not using a cache right, lol | 17:58 |
| harlowja | agreed | 17:58 |
| morganfainberg | harlowja, eh, there are cases where it's valid to cache >1MB values | 17:58 |
| harlowja | meh | 17:58 |
| harlowja | sure, there are cases, but in majority of things u are caching, u shouldn't do that | 17:58 |
| morganfainberg | harlowja, when i worked on CDN tech we did block based caching for videos and mp3s for streaming | 17:58 |
| morganfainberg | it made a lot of sense to cache more than 1MB | 17:58 |
| harlowja | sure, don't use memcache for that kidn of stuff, lol | 17:59 |
| morganfainberg | (this was back in 2008/2009) | 17:59 |
| morganfainberg | we had a bad custom event-based webserver with a weird allocator | 17:59 |
| morganfainberg | in c++ | 17:59 |
| morganfainberg | i.. lets just say never again | 17:59 |
| harlowja | :-p | 17:59 |
| morganfainberg | :P | 17:59 |
| harlowja | +2 | 17:59 |
| harlowja | ha | 17:59 |
| * morganfainberg might even have some of that code still. | 17:59 | |
| morganfainberg | but i hesitate to look for it cause it makes me cringe | 18:00 |
| harlowja | ya, sounds not so much fun | 18:00 |
| harlowja | http://trafficserver.apache.org/ was released from yahoo -> apache, and it sounds similar | 18:00 |
| harlowja | event based server (used for CDN stuf...) | 18:00 |
| harlowja | supposedly crazy c++ code, lol | 18:00 |
| harlowja | still runs a shit-ton of yahoo though, lol | 18:01 |
| *** dims has quit IRC | 18:06 | |
| *** dims has joined #openstack-oslo | 18:11 | |
| morganfainberg | zzzeek, ah maybe we are looking for pymemcache | 18:16 |
| dhellmann | jd__: is there anything to do for the pylockfile adoption other than the governance change? | 18:16 |
| morganfainberg | zzzeek, once we get through RC it should be easy to move to it (i'll submit the PR for dogpile to add it as a backend) | 18:17 |
| zzzeek | morganfainberg: wow have not seen this before looks promising | 18:17 |
| morganfainberg | zzzeek, exactly | 18:17 |
| zzzeek | wow this is perfect | 18:18 |
| zzzeek | the issues they lay out here are exactly the probs we hve | 18:18 |
| zzzeek | the theradlocal thing, etc | 18:18 |
| *** praneshp has joined #openstack-oslo | 18:18 | |
| morganfainberg | zzzeek, one minor issue, need to specify the serializer clearly | 18:18 |
| *** vigneshvar has quit IRC | 18:18 | |
| morganfainberg | python-memcache uses pickle. we probably need to implement something similar as the default in dogpile for that backend | 18:18 |
| morganfainberg | but allow overriding it | 18:18 |
| zzzeek | morganfainberg: i think we have that | 18:18 |
| zzzeek | morganfainberg: its in the backend, take a look | 18:18 |
| morganfainberg | zzzeek, ah ok | 18:18 |
| morganfainberg | zzzeek, i'll 2x check but this should be easy. | 18:19 |
| morganfainberg | zzzeek, add in configurable pools (actually i want to make a drop-in replacement for whatever lib is used that just layers pooling into the client) i think we'll have solved thread.local | 18:19 |
| zzzeek | yup | 18:20 |
| zzzeek | i can support any kind of pooling as long as there isnt one big scary pool actually in the lib itself :) | 18:20 |
| morganfainberg | zzzeek, yeah | 18:20 |
| zzzeek | b.c. someone else can maintain that :) | 18:20 |
| morganfainberg | hahah | 18:20 |
| morganfainberg | zzzeek, so i think i'm going to be rewriting that memcache pool patch to not use the cleaner thread. | 18:21 |
| morganfainberg | i reallly don't like that cleaner thread. | 18:21 |
| morganfainberg | zzzeek, it seems like overkill under eventlet, and in apache we don't care about thread.local | 18:21 |
| zzzeek | morganfainberg: yeah background threads is never something i want straight in a library, that can be built elsewhere | 18:22 |
| morganfainberg | zzzeek, +1billion | 18:22 |
| *** openstackgerrit has quit IRC | 18:32 | |
| *** openstackgerrit has joined #openstack-oslo | 18:32 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Mark middleware as obsolete https://review.openstack.org/118215 | 18:36 |
| *** vigneshvar has joined #openstack-oslo | 18:39 | |
| *** praneshp has quit IRC | 18:43 | |
| *** tkelsey has joined #openstack-oslo | 18:43 | |
| *** tkelsey has quit IRC | 18:45 | |
| *** vigneshvar has quit IRC | 18:48 | |
| *** vigneshvar has joined #openstack-oslo | 18:48 | |
| *** praneshp has joined #openstack-oslo | 18:49 | |
| *** praneshp has quit IRC | 18:50 | |
| *** e0ne has joined #openstack-oslo | 19:04 | |
| *** e0ne has quit IRC | 19:05 | |
| *** praneshp has joined #openstack-oslo | 19:05 | |
| *** amrith is now known as _amrith_ | 19:11 | |
| openstackgerrit | James Carey proposed a change to openstack/oslo.utils: Enable mask_password to handle byte code strings https://review.openstack.org/121657 | 19:14 |
| *** harlowja has quit IRC | 19:14 | |
| *** harlowja_ has joined #openstack-oslo | 19:14 | |
| *** rpodolyaka1 has joined #openstack-oslo | 19:23 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 19:29 |
| *** GheRivero_pto is now known as GheRivero | 19:30 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 19:33 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 19:35 |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 19:47 |
| *** jecarey has quit IRC | 19:47 | |
| *** harlowja_at_home has quit IRC | 19:57 | |
| *** jraim has quit IRC | 20:00 | |
| *** rpodolyaka1 has quit IRC | 20:00 | |
| *** jraim has joined #openstack-oslo | 20:00 | |
| *** rpodolyaka1 has joined #openstack-oslo | 20:01 | |
| *** kgiusti has quit IRC | 20:02 | |
| *** takedakn1 has quit IRC | 20:03 | |
| *** takedakn has joined #openstack-oslo | 20:04 | |
| *** andreykurilin_ has joined #openstack-oslo | 20:07 | |
| *** rpodolyaka1 has quit IRC | 20:11 | |
| *** noelbk has quit IRC | 20:12 | |
| *** rpodolyaka1 has joined #openstack-oslo | 20:13 | |
| *** noelbk has joined #openstack-oslo | 20:16 | |
| *** tongli has quit IRC | 20:26 | |
| *** dims has quit IRC | 20:34 | |
| *** dims has joined #openstack-oslo | 20:34 | |
| *** dims has quit IRC | 20:39 | |
| *** harlowja_at_home has joined #openstack-oslo | 20:48 | |
| *** harlowja_ is now known as harlowja_away | 20:49 | |
| *** jeckersb is now known as jeckersb_gone | 20:49 | |
| *** yamahata has joined #openstack-oslo | 20:58 | |
| *** yamahata has quit IRC | 20:58 | |
| *** yamahata has joined #openstack-oslo | 20:59 | |
| *** rpodolyaka1 has quit IRC | 21:01 | |
| *** jecarey has joined #openstack-oslo | 21:02 | |
| *** rpodolyaka1 has joined #openstack-oslo | 21:04 | |
| openstackgerrit | James Carey proposed a change to openstack/oslo.utils: Enable mask_password to handle byte code strings https://review.openstack.org/121657 | 21:07 |
| openstackgerrit | James Carey proposed a change to openstack/oslo-incubator: Enabled mask_password to handle byte code strings https://review.openstack.org/121632 | 21:08 |
| *** tedross_ has quit IRC | 21:11 | |
| *** vigneshvar has quit IRC | 21:14 | |
| dhellmann | harlowja_at_home: you around today? | 21:26 |
| *** dims has joined #openstack-oslo | 21:29 | |
| *** dims_ has joined #openstack-oslo | 21:31 | |
| *** dims has quit IRC | 21:34 | |
| *** pcm_ has quit IRC | 21:39 | |
| morganfainberg | zzzeek, so here is the new(est) pool impl: https://review.openstack.org/#/c/119452/7/keystone/common/cache/_memcache_pool.py | 21:40 |
| morganfainberg | zzzeek, it's a lot cleaner without the dedicated thread | 21:40 |
| zzzeek | morganfainberg: is that acquire/release for every get() ? | 21:41 |
| morganfainberg | zzzeek, yeah | 21:41 |
| zzzeek | morganfainberg: ouch | 21:41 |
| morganfainberg | zzzeek, same thing as before, because its in a pool. | 21:41 |
| morganfainberg | zzzeek, it just was hidden a bit more since it wasn't a single class. | 21:42 |
| morganfainberg | basically it is doing a "get a connection perform an action and drop that connection back into the pool" each time you need to do something with memcached | 21:42 |
| morganfainberg | it reuses the connection objects where possible. | 21:43 |
| morganfainberg | zzzeek, the biggest difference is we do the TTL reap on accquire instead of as a side-band process now | 21:44 |
| zzzeek | morganfainberg: i see that | 21:44 |
| dhellmann | we're down to 3 high or critical bugs for our final release this week | 21:44 |
| zzzeek | morganfainberg: you can maybe space out that check by only running it on a modulus of the current time | 21:44 |
| morganfainberg | zzzeek, ++ i'm good with that | 21:44 |
| *** flaper87 is now known as flaper87|afk | 21:46 | |
| *** andreykurilin_ has quit IRC | 21:48 | |
| *** andreykurilin_ has joined #openstack-oslo | 21:48 | |
| morganfainberg | zzzeek, the ttl reap is probably light enough weight that a mod wont buy us much | 21:53 |
| zzzeek | i just stare at code all day where .65 sec vs 1.5 sec for 100K is a big deal…. :) | 21:53 |
| morganfainberg | zzzeek, right but we're only reaping if a connection is in the pool and hasn't been used recently | 21:54 |
| zzzeek | you do this loop: for deaduntil, host in zip(self._server_list,98 | 21:54 |
| zzzeek | conn.servers): | 21:54 |
| zzzeek | for every acquire | 21:54 |
| morganfainberg | ah the deaduntil stuff | 21:54 |
| zzzeek | a call to zip(), etc. | 21:54 |
| morganfainberg | that is a different one. | 21:54 |
| morganfainberg | yeah, hrm. unfortunately we need to catch dead servers sooner vs later | 21:55 |
| morganfainberg | so we don't *ever* send traffic to them. | 21:55 |
| morganfainberg | zzzeek, i kinda wish we could just say "don't use memcached" | 21:56 |
| *** mriedem has quit IRC | 22:01 | |
| *** stevemar has joined #openstack-oslo | 22:05 | |
| stevemar | dhellmann, is there going to be another release of oslotest? or does that not necessarily need to be released on thursday with the others? | 22:06 |
| stevemar | dhellmann, was hoping to get https://review.openstack.org/#/c/120316/ in before another cut of oslotest | 22:06 |
| dhellmann | stevemar: I was not planning another release for juno, but that could go into the first alpha for kilo | 22:07 |
| dhellmann | stevemar: is it blocking you? | 22:08 |
| *** harlowja_at_home has quit IRC | 22:12 | |
| stevemar | dhellmann, not really, would just be a nice to have for keystoneclient | 22:13 |
| dhellmann | stevemar: we can do a kilo alpha release monday, if that works for you | 22:15 |
| stevemar | dhellmann, yep! | 22:15 |
| dhellmann | stevemar: ok, we might actually need more time for the review, but in principle it should be possible to release quickly | 22:16 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:17 | |
| *** takedakn has quit IRC | 22:20 | |
| *** bknudson has quit IRC | 22:21 | |
| openstackgerrit | Jon Grimm proposed a change to openstack/oslo.config: Warn if new and deprecated options are both present https://review.openstack.org/121691 | 22:21 |
| *** r1chardj0n3s is now known as r1chardj0n3s_afk | 22:23 | |
| *** harlowja_away is now known as harlowja_ | 22:24 | |
| openstackgerrit | A change was merged to openstack/oslo-incubator: Enabled mask_password to handle byte code strings https://review.openstack.org/121632 | 22:24 |
| *** rpodolyaka1 has quit IRC | 22:30 | |
| *** rpodolyaka1 has joined #openstack-oslo | 22:33 | |
| *** rpodolyaka1 has quit IRC | 22:36 | |
| *** rpodolyaka1 has joined #openstack-oslo | 22:36 | |
| *** gordc has quit IRC | 22:37 | |
| *** jgrimm is now known as zz_jgrimm | 22:38 | |
| *** bjornar_ has quit IRC | 22:39 | |
| *** r1chardj0n3s_afk is now known as r1chardj0n3s | 22:44 | |
| *** takedakn has joined #openstack-oslo | 22:50 | |
| *** andreykurilin_ has quit IRC | 22:59 | |
| *** harlowja_at_home has joined #openstack-oslo | 23:08 | |
| *** dims_ has quit IRC | 23:09 | |
| *** jeckersb_gone is now known as jeckersb | 23:16 | |
| *** rpodolyaka1 has quit IRC | 23:16 | |
| *** harlowja_at_home has quit IRC | 23:19 | |
| *** tsekiyama has quit IRC | 23:27 | |
| openstackgerrit | Joshua Harlow proposed a change to openstack/taskflow: Get the basics of a process executor working https://review.openstack.org/121280 | 23:34 |
| *** myatsenko has quit IRC | 23:52 | |
| *** AAzza has quit IRC | 23:52 | |
| *** AAzza has joined #openstack-oslo | 23:53 | |
| *** myatsenko has joined #openstack-oslo | 23:53 | |
| *** takedakn has quit IRC | 23:55 | |
| *** dims has joined #openstack-oslo | 23:57 | |
| *** arnaud has joined #openstack-oslo | 23:58 | |
| *** 20WABJQK7 has joined #openstack-oslo | 23:58 | |
| *** rpodolyaka1 has joined #openstack-oslo | 23:59 | |
| *** zzzeek has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!