Monday, 2014-08-25

« Sunday, 2014-08-24 Index Tuesday, 2014-08-26 »
*** harlowja_at_home has joined #openstack-oslo00:00
*** harlowja_at_home has quit IRC00:47
*** harlowja_at_home has joined #openstack-oslo01:30
*** harlowja_at_home has quit IRC01:36
*** harlowja_at_home has joined #openstack-oslo01:37
*** yamahata has joined #openstack-oslo01:44
*** oomichi has joined #openstack-oslo01:52
*** harlowja_at_home has quit IRC02:23
*** SridharG has joined #openstack-oslo02:32
*** dims has quit IRC02:44
*** dims has joined #openstack-oslo02:47
*** dims has quit IRC02:47
openstackgerritA change was merged to openstack/oslo.serialization: Updated from global requirements  https://review.openstack.org/11406503:43
*** harlowja_at_home has joined #openstack-oslo03:44
*** dims has joined #openstack-oslo03:48
*** jecarey has joined #openstack-oslo03:50
*** dims has quit IRC03:53
*** Alexei_987 has joined #openstack-oslo03:59
*** stevemar has quit IRC04:02
*** SridharG has quit IRC04:05
openstackgerritJoshua Harlow proposed a change to openstack/taskflow: A few action engine and action engine runner adjustments  https://review.openstack.org/11654404:17
*** harlowja_at_home has quit IRC04:22
*** amotoki has joined #openstack-oslo04:29
*** YorikSar has quit IRC04:41
*** YorikSar has joined #openstack-oslo04:42
*** dims has joined #openstack-oslo04:53
*** YorikSar has quit IRC04:57
*** dims has quit IRC04:59
openstackgerritJoshua Harlow proposed a change to openstack/taskflow: A few action engine and action engine runner adjustments  https://review.openstack.org/11654405:09
*** flaper87|afk is now known as flaper8705:10
*** jecarey has quit IRC05:15
openstackgerritJoshua Harlow proposed a change to openstack/taskflow: Fix engine iteration 'close-up shop' runtime path  https://review.openstack.org/11654405:18
openstackgerritJoshua Harlow proposed a change to openstack/taskflow: Tweak engine iteration 'close-up shop' runtime path  https://review.openstack.org/11654405:21
openstackgerritOpenStack Proposal Bot proposed a change to openstack/oslo.i18n: Imported Translations from Transifex  https://review.openstack.org/11655306:01
*** markmc has joined #openstack-oslo06:03
*** k4n0 has joined #openstack-oslo06:05
openstackgerritOpenStack Proposal Bot proposed a change to openstack/oslo.vmware: Imported Translations from Transifex  https://review.openstack.org/11473206:07
*** AAzza_afk has joined #openstack-oslo06:17
*** AAzzal has joined #openstack-oslo06:18
*** AAzza has quit IRC06:19
*** AAzzal is now known as AAzza06:19
*** AAzza_afk has quit IRC06:21
*** dtantsur|afk is now known as dtantsur06:23
*** YorikSar has joined #openstack-oslo06:36
*** dims has joined #openstack-oslo06:42
*** dims has quit IRC06:47
*** ihrachyshka has joined #openstack-oslo07:24
*** flaper87 is now known as flaper87|afk07:26
*** jokke__ is now known as jokke_07:31
*** praneshp has quit IRC07:34
*** ildikov has joined #openstack-oslo07:40
*** ihrachyshka has quit IRC07:48
openstackgerritAnn Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys  https://review.openstack.org/11623807:55
*** pblaho has joined #openstack-oslo08:05
*** dims has joined #openstack-oslo08:31
*** ihrachyshka has joined #openstack-oslo08:33
*** dims has quit IRC08:36
*** k4n0 has quit IRC08:39
*** ihrachyshka has quit IRC08:48
*** ihrachyshka has joined #openstack-oslo08:48
*** akamyshnikova has quit IRC08:53
*** k4n0 has joined #openstack-oslo08:53
*** ihrachyshka has quit IRC08:56
*** SridharGaddam has joined #openstack-oslo09:17
*** SridharGaddam has quit IRC09:39
openstackgerritAnn Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys  https://review.openstack.org/11623809:39
*** burgobianco has quit IRC09:42
*** dtantsur is now known as dtantsur|lunch09:43
*** yamahata has quit IRC09:43
*** burgobianco has joined #openstack-oslo09:44
openstackgerritAnn Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys  https://review.openstack.org/11623809:49
*** pblaho is now known as pblaho|afk10:01
*** pcm_ has joined #openstack-oslo10:05
openstackgerritAnn Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys  https://review.openstack.org/11623810:09
*** pcm_ has quit IRC10:09
*** pcm_ has joined #openstack-oslo10:10
openstackgerritAnn Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys  https://review.openstack.org/11623810:26
*** k4n0 has quit IRC10:47
*** flaper87|afk is now known as flaper8710:48
openstackgerritAnn Kamyshnikova proposed a change to openstack/oslo.db: ModelsMigrationsSync: Add check for foreign keys  https://review.openstack.org/11623810:49
*** jecarey has joined #openstack-oslo10:55
*** k4n0 has joined #openstack-oslo10:59
*** pcm_ has quit IRC11:01
*** pcm_ has joined #openstack-oslo11:03
*** pblaho|afk is now known as pblaho11:04
*** pcm_ has quit IRC11:04
*** pcm_ has joined #openstack-oslo11:05
*** yamahata has joined #openstack-oslo11:16
*** pblaho is now known as pblaho|afk11:24
*** dteselkin has quit IRC11:24
*** dteselkin has joined #openstack-oslo11:25
*** dims has joined #openstack-oslo11:29
*** pblaho|afk is now known as pblaho11:40
*** inteq has quit IRC11:50
*** pblaho is now known as pblaho|afk11:57
*** f13o has joined #openstack-oslo12:01
*** flaper87 is now known as flaper87|afk12:04
*** ihrachyshka has joined #openstack-oslo12:07
*** flaper87|afk is now known as flaper8712:07
openstackgerritIhar Hrachyshka proposed a change to openstack/oslo.db: Set utf8 encoding for mysql and postgresql  https://review.openstack.org/11123612:09
*** ihrachyshka has quit IRC12:13
*** ihrachyshka has joined #openstack-oslo12:14
*** kgiusti has joined #openstack-oslo12:20
*** jecarey has quit IRC12:35
*** gordc has joined #openstack-oslo12:39
*** mtanino has quit IRC12:41
*** pblaho|afk is now known as pblaho12:45
*** ildikov has quit IRC13:01
*** dtantsur|lunch is now known as dtantsur13:03
*** dims has quit IRC13:06
*** dims has joined #openstack-oslo13:06
*** ildikov_ has joined #openstack-oslo13:07
openstackgerritJulien Danjou proposed a change to openstack/oslo.messaging: Fix Python 3 testing  https://review.openstack.org/11660613:08
*** miqui has joined #openstack-oslo13:08
*** jeckersb_gone is now known as jeckersb13:16
*** bknudson has quit IRC13:18
*** mriedem has joined #openstack-oslo13:24
*** mtanino has joined #openstack-oslo13:31
*** stevemar has joined #openstack-oslo13:31
*** abhijeetm has joined #openstack-oslo13:32
*** AAzza is now known as AAzza_afk13:33
*** k4n0 has quit IRC13:35
*** bknudson has joined #openstack-oslo13:37
*** yassine has joined #openstack-oslo13:40
openstackgerritDavanum Srinivas (dims) proposed a change to openstack/oslo.vmware: VMWare: Support for pluggable Datastore selection  https://review.openstack.org/11455113:55
openstackgerritJulien Danjou proposed a change to openstack/oslo.messaging: Fix Python 3 testing  https://review.openstack.org/11660614:00
dimsdhellmann: ping14:03
dimsdhellmann: a few things from last week, when you get a chance14:08
dimsdhellmann: 1) Do we need to keep tests for code that already moved to libraries? https://review.openstack.org/#/c/114649/14:09
dimsdhellmann: 2) lifeless was asking about a pbr release14:09
dims3) oslo.serialization is ready for a release14:10
dimsdhellmann: 3) oslo.serialization is ready for a release14:10
dimsdhellmann: 4) do we have a oslo.log repo? who is working on it?14:11
*** oomichi has quit IRC14:12
*** amotoki has quit IRC14:14
mriedemwhy is simplejson in oslo-incubator's test-requirements when it's used in the runtime jsonutils code?14:16
*** jeckersb is now known as jeckersb_gone14:19
*** yamahata has quit IRC14:22
dimsmriedem: it's used in runtime only for python < 2.714:23
mriedemdims: yeah.... https://bugs.launchpad.net/oslo/+bug/136123014:24
mriedemi'm going to be updating global-reqs14:24
mriedemand oslo's test-reqs14:24
mriedemmakes it very difficult for anyone doing CD on RHEL 6 using simplejson to know when their package deps need updating :)14:25
dimsmriedem: bug 1361230 failure is only in python 2.6 right?14:25
dimsack14:26
mriedemdims: so far that's all i've seen it on14:26
mriedemrhel 6.514:26
*** mriedem has left #openstack-oslo14:26
*** mriedem has joined #openstack-oslo14:26
*** jraim__ has joined #openstack-oslo14:29
*** noelbk has quit IRC14:31
*** mgagne_ has joined #openstack-oslo14:32
*** mgagne has quit IRC14:32
*** freyes has quit IRC14:32
*** lbragstad has quit IRC14:32
*** jraim has quit IRC14:32
*** mtreinish has quit IRC14:32
*** jroll has quit IRC14:32
*** russellb has quit IRC14:32
*** therve has quit IRC14:32
*** lbragstad has joined #openstack-oslo14:32
*** mtreinish has joined #openstack-oslo14:32
openstackgerritA change was merged to openstack/oslo-incubator: Remove unused/mutable default args  https://review.openstack.org/11638614:32
*** therve has joined #openstack-oslo14:33
*** jroll has joined #openstack-oslo14:33
*** russellb has joined #openstack-oslo14:33
*** freyes has joined #openstack-oslo14:34
*** tsekiyama has joined #openstack-oslo14:34
*** jeckersb_gone is now known as jeckersb14:36
*** ihrachyshka has quit IRC14:44
*** mtreinish_ has joined #openstack-oslo14:50
*** markmc_ has joined #openstack-oslo14:50
*** lbragstad has quit IRC14:50
*** mtreinish has quit IRC14:50
*** markmc has quit IRC14:50
*** lbragstad has joined #openstack-oslo14:50
*** mtreinish_ is now known as mtreinish14:51
flaper87Have any of you seen this pbr error before? http://logs.openstack.org/65/115265/3/check/check-tempest-dsvm-full/14edab9/logs/devstacklog.txt.gz#_2014-08-25_11_03_22_16214:52
*** abhijeetm has quit IRC14:52
flaper87I remember seeing something about pbr and semver in the mailing list (not long ago)14:52
*** AAzza_afk is now known as AAzza14:53
dhellmannflaper87: I suspect that's caused by the version format being invalid. Try re-tagging as 0.0.1.0a3. (the first 3 values are meant to be pure numbers)14:55
dhellmannflaper87: we don't use alpha releases of pre-1.0 versions in oslo either, fwiw14:55
*** ildikov has joined #openstack-oslo14:55
dhellmanndims: thanks for filling in for me last week, I'm still catching up this morning14:55
flaper87dhellmann: ah, mmh, that explains it!14:55
dhellmanndims: I have pbr, cliff, and oslo.serialization on my list for releases today14:56
*** AAzza is now known as AAzza_afk14:57
dhellmannflaper87: would you file a bug against pbr about this? we should report the error when the sdist is created to avoid having issues installing it14:57
flaper87dhellmann: absolutely, let me do that right away before I forget14:57
dhellmannflaper87: and you'll want to test locally before pushing a new tag, since my answer is a guess :-)14:58
dimsdhellmann: cool14:58
*** ildikov_ has quit IRC14:58
flaper87dhellmann: mmh, that's the thing, I can't replicate it locally. Since that's happening with devstack, I'm setting up a vm w/ devstack. Lets see if I can replicate it, it might be an old-pbr bug14:58
flaper87not sure, just guessing14:59
jd__if you want to unblock oslo.messaging gate here it should be https://review.openstack.org/#/c/116606/ :)14:59
dhellmannflaper87: maybe you and I have a different version of pip than the CI systems?15:00
flaper87dhellmann: likely, let me check if I can get the pip version from the logs15:00
dimspbr==0.10.1.dev8.g81c200015:01
flaper87dims: dhellmann: logs say pip 1.5.4-1 is installed15:01
flaper87I have 1.5.615:01
flaper87ah nevermind15:01
flaper87it's then manually updated anyway15:02
dhellmannjd__: that change looks fine; did you think about using one of the decorators? https://docs.python.org/2/library/unittest.html#skipping-tests-and-expected-failures15:02
jd__dhellmann: I did in one place, I can use it everywhere if that's prefered indeed15:02
dhellmannjd__: ah, I just looked at the first file so far15:03
dhellmannflaper87: ah, indeed, it's a problem with the current master version of pbr15:03
dhellmanndims: nice ^^15:03
flaper87oh ok, then let me file that bug and tag a version w/o alpha15:03
dhellmannflaper87: tagging without the alpha works, as does 0.0.1.0a3 (note that the alpha is in the 4th part of the version)15:04
flaper87dhellmann: yup, I'll just follow what we do for oslo, it makes sense to me.15:05
dhellmannflaper87: ok15:05
openstackgerritJulien Danjou proposed a change to openstack/oslo.messaging: Fix Python 3 testing  https://review.openstack.org/11660615:05
jd__dhellmann: updated, fingers crossed15:05
dhellmannjd__: what changed that makes raising at import time not work? do we need to make that more public to help other projects?15:07
jd__dhellmann: I honestly don't know, I think it's something that got released a few days ago since it used to work, like testrepository or subunit15:07
dhellmannjd__: ok, just wondering15:08
jd__not sure it's a big deal since I don't expect a lot of projects doing that15:08
dhellmannjd__: +2a15:08
jd__nice15:09
dhellmanndims: regarding removing tests; yes, we need to keep the tests with modules in the incubator until we remove those modules15:09
dimsdhellmann: ack will re-instate them15:09
*** markmcclain has joined #openstack-oslo15:10
dhellmanndims: we need to rethink our approach to graduation. :-/15:10
*** AAzza_afk is now known as AAzza15:12
dhellmannflaper87: would you let me know when you've got a new glance.store tagged? I don't want to release pbr if it's going to break you in other environments than our gate.15:14
openstackgerritDavanum Srinivas (dims) proposed a change to openstack/oslo-incubator: Switch oslo-incubator to use oslo.utils  https://review.openstack.org/11464915:16
dimsdhellmann: i hear you :(15:17
*** markmcclain has quit IRC15:19
*** markmcclain has joined #openstack-oslo15:21
*** dtantsur is now known as dtantsur|afk15:23
*** ildikov has quit IRC15:30
openstackgerritDavanum Srinivas (dims) proposed a change to openstack/oslo.vmware: VMWare: Support for pluggable Datastore selection  https://review.openstack.org/11455115:45
*** markmcclain has quit IRC15:46
*** yassine has quit IRC15:48
openstackgerritDoug Hellmann proposed a change to openstack/cliff: Add release notes for 1.7.0  https://review.openstack.org/11665815:48
flaper87dhellmann: it's out: https://pypi.python.org/pypi/glance.store15:50
* flaper87 rechecks that devstack patch15:51
flaper87lets see what happens15:51
dhellmannflaper87: ok, let me know how the tests go and I'll release if they pass15:51
flaper87dhellmann: will do. Thanks!15:51
*** flaper87 is now known as flaper87|afk15:52
openstackgerritJulien Danjou proposed a change to openstack/oslo.messaging: Fix Python 3 testing  https://review.openstack.org/11660615:53
jd__dhellmann: I had to update the patch, skipIf doesn't work on class actually15:54
dhellmannjd__: le sigh15:54
*** enykeev has joined #openstack-oslo15:54
*** zzzeek has joined #openstack-oslo16:03
openstackgerritDoug Hellmann proposed a change to openstack/oslo.serialization: Clean up docs before first release  https://review.openstack.org/11666516:05
dhellmanndims, bnemec : we should merge something like ^^ before releasing oslo.serialization16:06
*** pblaho has quit IRC16:10
*** markmcclain has joined #openstack-oslo16:15
openstackgerritSteve Martinelli proposed a change to openstack/oslotest: Add pdb support to tox with debug helper shell script  https://review.openstack.org/11666916:19
*** praneshp has joined #openstack-oslo16:19
*** markmcclain has quit IRC16:20
*** dims_ has joined #openstack-oslo16:23
*** dims has quit IRC16:24
*** markmc_ has quit IRC16:30
*** celttechie has joined #openstack-oslo16:34
*** ildikov has joined #openstack-oslo16:34
*** dims has joined #openstack-oslo16:35
*** dims_ has quit IRC16:35
openstackgerritSteve Martinelli proposed a change to openstack/oslotest: Add pdb support to tox with debug helper shell script  https://review.openstack.org/11666916:36
*** markmcclain has joined #openstack-oslo16:37
YorikSarzzzeek: ping16:51
zzzeekhey16:51
YorikSarzzzeek: Hi16:51
YorikSarzzzeek: I've found out that you're the author of dogpile.cache as well :)16:51
zzzeekyep16:51
YorikSarzzzeek: So I have a question. Do I get it right: when I use dogpile's memcached backend (in Keystone), it creates new connection for every new thread.16:52
YorikSarzzzeek: Can this behavior be somehow changed?16:52
YorikSarzzzeek: I've looked through code and it looks like it can't be turned off, right?16:53
zzzeekYorikSar: i dont know if that is true.  which backend specifically ?16:53
YorikSarzzzeek: python-memcached is used by default16:54
zzzeekYorikSar: according to https://github.com/linsomniac/python-memcached/blob/master/memcache.py#L129 that client is already a threading.local16:55
YorikSarzzzeek: We ran into problem: under load Keystone ends up with huge number of opened connections to memcached.16:55
YorikSarzzzeek: Ah...16:56
YorikSarWhy?...16:56
zzzeekYorikSar: because a single TCP connection is not threadsafe16:56
YorikSarzzzeek: Ok, thanks. Looks like we need to switch to pylibmc or smth...16:57
YorikSarzzzeek: But we could reuse it in different threads.16:57
zzzeekYorikSar: OK but the best you can do is a pool that is of more limited size16:57
YorikSarzzzeek: Like keep a reference around for just a bit longer than thread lives.16:57
YorikSar*then16:58
YorikSar(or than)16:58
zzzeekYorikSar: dogpile’s backend is also doing threading.local so if you want to use some kind of pool you’d need to adapt a custom backend16:58
*** yamahata has joined #openstack-oslo16:58
YorikSarzzzeek: Yeah... That's really strange for me...16:58
zzzeekYorikSar: when you use regular threads, this is the easiest/most stable/simple way to go16:59
zzzeekYorikSar: but when you are using greenlets, those asssumptions change16:59
zzzeekYorikSar: because now you can have 10000 “threads"16:59
YorikSarzzzeek: I think connection lifetime management should be decoupled from connection itself.16:59
zzzeekYorikSar: so dogpile’s backend is not set up for large scale greenlet use16:59
YorikSarzzzeek: Well, even for usual threads you could benefit from TCP connection reuse.17:00
zzzeekYorikSar: as long as dogpile’s memcached backend doesn’t step on itself when run out of the box in a small application with just ten threads, sure17:00
*** openstackgerrit has quit IRC17:00
YorikSarzzzeek: And for 10 threads that rarely use memcached, 1 TCP connection would be enough.17:01
zzzeekYorikSar: no its not17:01
*** harlowja has joined #openstack-oslo17:01
zzzeekYorikSar: unless you have an efficinet pooling system making sure it is sent correctly17:01
zzzeekYorikSar: havent seen such a pool for memcached connections17:01
zzzeekYorikSar: this would be totally custom backend for dogpile.  not hard to do.   dogpile’s backends are totally extensible.  maybe there is one already17:02
YorikSarzzzeek: But if python-memcached doesn't work like that, there's not much we can do.17:03
zzzeekYorikSar: well pylibmc is much better17:03
zzzeekYorikSar: it has some kind of pools but i dont know that I had good results with them17:03
YorikSarzzzeek: But it's C.17:03
*** AAzza is now known as AAzza_afk17:03
YorikSarzzzeek: And it might block eventlet.17:03
zzzeekYorikSar: i feel like pylibmc uses eventlet already17:04
zzzeekguess not17:04
zzzeekYorikSar: see this?  https://blueprints.launchpad.net/oslo/+spec/eventlet-safe-memcache-client17:05
YorikSarzzzeek: Huh... But that doesn't see to be true anymore - we use python-memcached and under load and no such effects.17:07
zzzeekYorikSar: shrugs17:07
dimsdhellmann: are we doing oslo.log? (anytime soon)17:09
*** mgagne_ has quit IRC17:11
*** mgagne_ has joined #openstack-oslo17:11
*** openstackgerrit has joined #openstack-oslo17:13
YorikSarzzzeek: Ok, it seems I'll have to do smth as ugly as http://stackoverflow.com/a/22520633/238308 with a new dogpile backend for our next Icehouse release...17:13
YorikSarzzzeek: But for upstream we really should find a better way.17:13
*** yamahata has quit IRC17:21
dhellmanndims: zul and I are on tap to do that, but neither of us have had a chance to start it17:21
*** AAzza_afk is now known as AAzza17:22
*** markmcclain has quit IRC17:25
openstackgerritVipin Balachandran proposed a change to openstack/oslo.vmware: Refactoring to reduce noise in log files  https://review.openstack.org/11650417:30
*** mgagne_ is now known as mgagne17:39
*** zzzeek has quit IRC17:51
*** zzzeek has joined #openstack-oslo17:51
openstackgerritDavanum Srinivas (dims) proposed a change to openstack/oslo-incubator: Switch oslo-incubator to use oslo.utils  https://review.openstack.org/11464917:58
dhellmanndims: is the oslo.utils situation with keystone worked out?18:01
dimsdhellmann: thanks for the reminder, so we need oslo.utils in stable branches, there were a few reviews, need to check on them18:02
dhellmanndims: ok, let me know if I need to +2 anything there18:02
dimsdhellmann: yep will do18:02
*** AAzza is now known as AAzza_afk18:26
*** AAzza_afk is now known as AAzza18:37
openstackgerritgordon chung proposed a change to openstack/oslo-specs: graduate-oslo-middleware  https://review.openstack.org/11035318:39
*** dtantsur|afk is now known as dtantsur18:54
*** pblaho has joined #openstack-oslo19:02
*** pblaho has quit IRC19:02
openstackgerritA change was merged to openstack/taskflow: Make greenexecutor not keep greenthreads active  https://review.openstack.org/10592219:06
*** tsekiyama has quit IRC19:14
*** tsekiyama has joined #openstack-oslo19:16
*** tpatil has joined #openstack-oslo19:22
dhellmannzzzeek: have a sec to talk about this test thread on the ML?19:26
zzzeeksure19:26
zzzeekso robert collins reached out to me and said he’d work on a sample integration point19:26
dhellmannI hate to make you do this, but can you summarize the issue(s)?19:26
*** tpatil_ has joined #openstack-oslo19:27
zzzeekwell we need resources that span multiple tests to be managed19:27
zzzeekusually the way I’ve acheived that is by running hooks when the test suite starts up and when it tears down19:27
dhellmannisn't this normally where we would do something like setupClass()?19:27
zzzeekdhellmann: well setupclass is not expansive enough19:27
zzzeekdhellmann: also it seems like we dont necessarily make use of setupclass/teardownclass, does testtools support it ?19:28
dhellmannhow does anything else know when it's safe to reuse vs. recreate the resource?19:28
zzzeekdhellmann: well theres different ways to do this.  robert collins pointed me to testresources19:28
dhellmannI assume it does, but I don't know19:28
zzzeekdhellmann: we can use testresources but i was looking for advise on how to set it up for a set of test runs.   i dont like that it requires an explicit hook19:29
zzzeekdhellmann: robert said on friday he’d work out something but havent heard from him19:29
dhellmannzzzeek: ok, he's in Australia, so time zones are going to be way off19:30
zzzeekdhellmann: initially i emulated someting I saw in other openstack reviews which was to use the hooks in .testr.conf as one way to have a global setup/terdown19:30
dhellmannthis isn't something I've looked into before, where did you find that pattern?19:30
*** tpatil has quit IRC19:30
zzzeekdhellmann: it was in a previous patch submitted by angus lees regarding a similar provisioning feature19:30
zzzeekrobert collins said it wasn’t appropriate.  so that’s kind of a big -1 on that.19:31
*** AAzza is now known as AAzza_afk19:32
dhellmannsure, lifeless did a lot of work on the testing tools we're using, so I'll be interested to see what his input is on this19:32
zzzeekdhellmann: well it looks like we would need to use the load_tests() hook in order to instrument some/all TestSuite objects19:32
zzzeekdhellmann: that’s not a universally recognized hook19:33
zzzeekdhellmann: so i was mostly asking, hey, OK i can’t hardcode to testr, but what test runners *do* we need to support, because neither nose nor pytest support load_tests()19:33
zzzeeke.g. the “hardcode / dont hardcode” thing is starting to frustrate me19:33
dhellmannwe very very very much do not care about other test runners for openstack19:34
dhellmannregardless of its faults, we're using testr for all of the projects19:34
zzzeekdhellmann: ive been told otherwise19:34
dhellmannbecause of its features19:34
dhellmannwhich project?19:34
zzzeekhold on19:34
zzzeekthere were a few when i grepped19:34
dhellmannstevedore has not yet been converted, but I have a branch locally where I've started that work19:35
zzzeekdhellmann: read james E blair on aug 21 9:19 pm at https://review.openstack.org/#/c/110486/19:35
dhellmannit doesn't use the db, so that conversion isn't a blocker19:35
zzzeekdhellmann: dont see how to link to the comment19:35
zzzeek“I agree with lifeless; we've fought hard to avoid requiring a specific test runner, and it's still the case that not all OpenStack projects have migrated to testr yet."19:35
zzzeekdhellmann: also, the way that patch is, you *can* use any runner, its just the long-term provisioning thing wont take effect in those cases19:36
dhellmannok, well, that's news to me, I thought infra was the team that wanted us all using testr19:36
zzzeekdhellmann: i dont see a .testr.conf in horizon19:37
dhellmannthey use the django test runner, I think, but I don't think that's important for this change19:37
zzzeekdhellmann: testresources does a much more elaborate system of maintaining resources over many tests.  it’s cool, but needs a more invasive hook setup19:37
zzzeekdhellmann: e.g. the load_tests(), and i wanted to see if lifeless could provide an easy way to make that work across all tests...19:38
zzzeekdhellmann: also, it changes TestSuite into OptimisingTestSuite19:38
zzzeekdhellmann: meaning, it highly modifes how all tests are run19:38
dhellmannyeah, that doesn't sound like a good thing19:38
zzzeekdhellmann: it essentially tries to reorder tests19:38
zzzeekdhellmann: well its part of the wohle testtools family19:38
dhellmannI thought testr was doing that for us anyway19:38
zzzeekdhellmann: not sure.19:38
zzzeekdhellmann: so seems like, I need to take this series of patches and make a blueprint out of it19:39
zzzeeki was going off the bug report i made: https://bugs.launchpad.net/oslo/+bug/133920619:39
dhellmannhow about if you work on a spec, so we can clarify the goals and get everyone to buy off on them before we make this big of a change. I'd like it to be circulated with the infra team and any other programs that want to have a say, and a spec is going to make it easier to log the decisions we make in that process19:40
zzzeekalso, Alexei Kornienko earlier on this patch series: “I don't like the fact that we do a vendor lock on testr. I don't think that this is something we should depend on. We've had similar problem in ceilometer and it's using a small shell script for the same purpose: https://github.com/openstack/ceilometer/blob/master/setup-test-env.sh”19:40
*** celttechie has quit IRC19:42
tpatil_Hi dhellmann19:53
dhellmannhi, tpatil_19:53
tpatil_Can you please review patch https://review.openstack.org/#/c/103186/? this is a critical issue we are trying to fix from long time.19:53
stevemardhellmann, moved19:58
dhellmanntpatil_: you want markmc, jd__ , sileht, and flaper87|afk to look at that -- they're all on the review list19:59
dhellmanntpatil_: it looks like sileht (Mehdi) has already given some feedback19:59
tpatil_dhellmann: Yes, we had a lengthly discussion, he is not happy with calling gc.collect.20:00
dhellmannstevemar: I'm not sure I understand your objection.20:00
dhellmanntpatil_: the folks I listed above are much more familiar with the implementation of that library, so I'm likely to follow their advice20:01
stevemardhellmann, i wanted to add some notes about future use of oslo_debug_helper.sh, but currently, most of the content on the developer docs is actually generated from the README.rst file in the base directory20:01
stevemardhellmann, it just feels like the wrong place to add to it, do you mind if i muss around and restructure things (in a separate patch)20:02
tpatil_dhellmann:Sure, I will contact them. Thank you for your time.20:02
dhellmannstevemar: I think we follow the pattern of importing the readme in most of the oslo libs, but this one does seem to have more than the usual amount of detail20:02
stevemardhellmann, ah didn't know that was the norm20:03
dhellmannstevemar: well, like I said, usually there are fewer details. we should probably have moved these around, so if you want to do that I'm sure it'll be accepted20:04
stevemardhellmann, sure, i'll probably do it in 2 sets, 1) just add to the mess (add the steps in README.rst), then 2) move things around a bit and reduce the amount you see in README.rst20:05
*** yassine has joined #openstack-oslo20:05
*** kgiusti has left #openstack-oslo20:06
*** dtantsur is now known as dtantsur|afk20:07
*** celttechie has joined #openstack-oslo20:08
openstackgerritJoshua Harlow proposed a change to openstack/taskflow: Raise a runtime error when mixed green/non-green futures  https://review.openstack.org/11648720:10
dhellmannstevemar: that approach seems fine, thanks20:14
openstackgerritSteve Martinelli proposed a change to openstack/oslotest: Restructure oslotest docs  https://review.openstack.org/11671620:31
*** miqui has quit IRC20:33
*** tpatil_ has quit IRC20:40
openstackgerritSteve Martinelli proposed a change to openstack/oslotest: Add documentation for running oslo_debug_heler.sh  https://review.openstack.org/11672220:43
openstackgerritSteve Martinelli proposed a change to openstack/oslotest: Restructure oslotest docs  https://review.openstack.org/11671620:48
*** jeckersb is now known as jeckersb_gone20:50
*** ajo|mac has joined #openstack-oslo21:07
*** ajo|mac is now known as ajo__21:07
openstackgerritSteve Martinelli proposed a change to openstack/oslotest: Add documentation for running oslo_debug_heler.sh  https://review.openstack.org/11672221:19
dhellmannbnemec, dims : are you happy with the docs output in https://review.openstack.org/#/c/116665/ ?21:27
dhellmannbnemec, dims : if so, I'll cut a release tomorrow after it merges21:28
bnemecdhellmann: Yep, +A21:30
dhellmannbnemec: cool21:30
*** markmcclain has joined #openstack-oslo21:30
dhellmannmy cliff doc updates failed on neutron tests :-(21:30
openstackgerritDoug Hellmann proposed a change to openstack/cliff: Add release notes for 1.7.0  https://review.openstack.org/11665821:39
openstackgerritDoug Hellmann proposed a change to openstack/cliff: Fix stable integration tests  https://review.openstack.org/11673121:39
*** mriedem has quit IRC21:42
*** dims has quit IRC21:50
*** dims has joined #openstack-oslo21:50
*** stevemar has quit IRC21:53
*** dims has quit IRC21:55
*** ildikov_ has joined #openstack-oslo21:57
*** dims has joined #openstack-oslo21:59
*** dims_ has joined #openstack-oslo21:59
*** ildikov has quit IRC22:00
*** nealph__ has joined #openstack-oslo22:01
dims_dhellmann: one nit - http://docs-draft.openstack.org/65/116665/1/check/gate-oslo.serialization-docs/bf1b993/doc/build/html/_sources/readme.txt22:02
dhellmanndims_: looking22:02
dims_dhellmann: points to a page with ".. include:: ../../README.rst"22:02
dims_dhellmann: not a show stopper22:02
*** nealph_ has quit IRC22:03
*** dims has quit IRC22:03
dhellmanndims: yeah, that's the source file for the page, isn't it?22:03
dhellmanndims_: and that is accurately reflecting the source of the .rst file22:03
*** ajo__ has quit IRC22:04
*** ajo|mac has joined #openstack-oslo22:05
*** dims has joined #openstack-oslo22:05
*** dims_ has quit IRC22:06
*** tsekiyam_ has joined #openstack-oslo22:06
*** gordc has quit IRC22:06
*** pcm_ has quit IRC22:06
*** ajo|mac is now known as ajo__22:07
*** tsekiyama has quit IRC22:08
*** dims has quit IRC22:08
*** jgrimm has joined #openstack-oslo22:09
*** dims has joined #openstack-oslo22:09
*** dims has quit IRC22:10
*** dims_ has joined #openstack-oslo22:10
*** ajo__ has quit IRC22:12
*** HenryG_ has joined #openstack-oslo22:12
*** sheeprine has quit IRC22:13
*** HenryG has quit IRC22:15
openstackgerritAbhishek Chanda proposed a change to openstack/stevedore: Migrate tox to use testr  https://review.openstack.org/10167322:20
*** markmcclain has quit IRC22:21
openstackgerritA change was merged to openstack/oslotest: Add pdb support to tox with debug helper shell script  https://review.openstack.org/11666922:28
*** shakamunyi has joined #openstack-oslo22:32
*** shakamunyi has quit IRC22:40
*** celttechie has quit IRC22:47
*** shakamunyi has joined #openstack-oslo22:54
*** jgrimm is now known as zz_jgrimm22:54
*** bknudson has quit IRC23:03
*** markmcclain1 has joined #openstack-oslo23:04
*** oomichi has joined #openstack-oslo23:16
*** oomichi is now known as oomichi_away23:17
*** yassine has quit IRC23:22
openstackgerritlifeless proposed a change to openstack-dev/pbr: Raise an error if preversion versions are too low  https://review.openstack.org/10826923:27
openstackgerritlifeless proposed a change to openstack-dev/pbr: Look for and process sem-ver pseudo headers in git  https://review.openstack.org/10827023:27
openstackgerritlifeless proposed a change to openstack-dev/pbr: Refactor oneline changelog walking.  https://review.openstack.org/10827123:27
openstackgerritlifeless proposed a change to openstack-dev/pbr: Only consider tags that look like versions.  https://review.openstack.org/11440323:27
openstackgerritJoshua Harlow proposed a change to openstack/taskflow: Add a futures type that can contain most of our future functionality  https://review.openstack.org/11648923:36
*** dims_ has quit IRC23:48
*** dims has joined #openstack-oslo23:48
*** dims has quit IRC23:52
*** dims has joined #openstack-oslo23:52
*** ildikov_ has quit IRC23:58
« Sunday, 2014-08-24 Index Tuesday, 2014-08-26 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!