Friday, 2014-06-20

*** yamahata has quit IRC00:01
*** yamahata has joined #openstack-qa00:04
*** malini1 has joined #openstack-qa00:23
*** malini1 has quit IRC00:24
*** malini1 has joined #openstack-qa00:24
*** matsuhashi has joined #openstack-qa00:28
openstackgerritAdam Gandelman proposed a change to openstack/tempest: Add new snapshot compute feature flag  https://review.openstack.org/10138100:35
*** leseb has joined #openstack-qa00:36
*** leseb has quit IRC00:41
openstackgerritA change was merged to openstack-dev/devstack: Use oslo-messaging-zmq-receiver to start zeromq receiver  https://review.openstack.org/10113600:45
openstackgerritA change was merged to openstack/tempest: Skip case for fixing incorrect exception assertion  https://review.openstack.org/9634400:45
*** devlaps has joined #openstack-qa00:45
openstackrecheckopenstack-dev/devstack change: https://review.openstack.org/100991 failed because of: gate-grenade-dsvm-partial-ncpu: https://bugs.launchpad.net/bugs/132061700:47
*** matsuhashi has quit IRC00:48
*** yamahata has quit IRC00:49
*** matsuhashi has joined #openstack-qa00:49
*** matsuhashi has quit IRC00:54
*** dims has joined #openstack-qa00:58
*** matsuhashi has joined #openstack-qa01:01
*** dims has quit IRC01:02
*** dims_ has joined #openstack-qa01:05
*** matsuhashi has quit IRC01:09
*** matsuhashi has joined #openstack-qa01:10
openstackgerritA change was merged to openstack-infra/devstack-gate: Use the DEVSTACK_GATE_USE_APACHE flag to disable  https://review.openstack.org/10074401:17
*** nosnos has joined #openstack-qa01:30
openstackgerritGhanshyam Mann proposed a change to openstack/tempest: Verify list-Floating-IP-Bulk Nova API attributes  https://review.openstack.org/10110001:33
*** leseb has joined #openstack-qa01:37
*** yamahata has joined #openstack-qa01:38
*** dims_ has quit IRC01:39
*** Sukhdev has quit IRC01:39
*** TravT has quit IRC01:41
*** leseb has quit IRC01:42
*** bnemec is now known as bnemec_away02:20
*** rwsu has quit IRC02:24
openstackrecheckopenstack/horizon change: https://review.openstack.org/75862 failed gate-tempest-dsvm-full in the gate queue with an unrecognized error02:25
*** nati_ueno has quit IRC02:26
*** adalbas has quit IRC02:30
*** leseb has joined #openstack-qa02:38
*** GMann has quit IRC02:40
*** GMann has joined #openstack-qa02:40
*** leseb has quit IRC02:43
*** harlowja is now known as harlowja_away02:56
*** weshay has quit IRC03:00
*** matsuhashi has quit IRC03:02
*** matsuhashi has joined #openstack-qa03:03
*** dpaterson has quit IRC03:06
*** matsuhashi has quit IRC03:07
*** devlaps has quit IRC03:10
*** Sukhdev has joined #openstack-qa03:14
*** mestery has quit IRC03:17
*** nosnos has quit IRC03:18
openstackgerritYair Fried proposed a change to openstack/qa-specs: Restructure Scenarios  https://review.openstack.org/10123203:30
*** raies has joined #openstack-qa03:38
*** leseb has joined #openstack-qa03:39
openstackgerritOpenStack Proposal Bot proposed a change to openstack/tempest: Updated from global requirements  https://review.openstack.org/9682903:39
*** raies has quit IRC03:39
openstackgerritRavikumar Venkatesan proposed a change to openstack/tempest: Adds "user-password-update" to v3/test_users  https://review.openstack.org/9210003:41
openstackgerritYair Fried proposed a change to openstack/tempest: Convert scenario test tearDown to addCleanup  https://review.openstack.org/6210103:42
*** leseb has quit IRC03:43
openstackgerritYair Fried proposed a change to openstack/tempest: Convert scenario test tearDown to addCleanup  https://review.openstack.org/6210103:45
openstackgerritYair Fried proposed a change to openstack/qa-specs: Restructure Scenarios  https://review.openstack.org/10123203:47
*** Madan1 has joined #openstack-qa03:50
*** matsuhashi has joined #openstack-qa03:53
*** mwagner_lap has joined #openstack-qa03:55
*** Poornima has joined #openstack-qa03:59
*** matsuhashi has quit IRC04:02
*** matsuhashi has joined #openstack-qa04:02
*** matsuhashi has quit IRC04:06
openstackgerritYuiko Takada proposed a change to openstack/tempest: Add Ironic GET validate-node API test  https://review.openstack.org/9028204:08
*** nosnos has joined #openstack-qa04:11
openstackgerritYair Fried proposed a change to openstack/qa-specs: Restructure Scenarios  https://review.openstack.org/10123204:15
openstackgerritYair Fried proposed a change to openstack/qa-specs: Restructure Scenarios  https://review.openstack.org/10123204:16
openstackgerritAnthony Woods proposed a change to openstack-dev/devstack: Adds neutron linuxbridge-agent support to ironic deployment.  https://review.openstack.org/9831704:20
openstackgerritAnthony Woods proposed a change to openstack-dev/devstack: Adds neutron linuxbridge-agent support to ironic deployment.  https://review.openstack.org/9831704:31
*** leseb has joined #openstack-qa04:39
openstackgerritYair Fried proposed a change to openstack/qa-specs: Restructure Scenarios  https://review.openstack.org/10123204:44
*** leseb has quit IRC04:44
*** Madan1 has quit IRC04:49
*** Madan has joined #openstack-qa04:50
*** Longgeek has joined #openstack-qa04:50
*** ildikov has quit IRC04:53
*** raies has joined #openstack-qa04:55
*** openstackgerrit has quit IRC04:57
afazekasmtreinish: my git blame show the example.com added by https://review.openstack.org/#/c/2884/  more than 2 years before, when probably glance did not tried to connect to the location at all05:00
*** Longgeek has quit IRC05:11
*** Longgeek has joined #openstack-qa05:12
*** Sukhdev has quit IRC05:14
*** Longgeek has quit IRC05:14
*** Longgeek has joined #openstack-qa05:14
*** sweston_cloud has joined #openstack-qa05:15
*** coolsvap|afk is now known as coolsvap05:22
*** sweston_cloud has quit IRC05:26
*** sweston_cloud has joined #openstack-qa05:27
*** sweston_cloud has quit IRC05:28
*** sweston_cloud has joined #openstack-qa05:30
*** sweston_cloud has quit IRC05:34
*** sweston has joined #openstack-qa05:34
*** sweston has quit IRC05:38
*** chandan_kumar has joined #openstack-qa05:40
*** leseb has joined #openstack-qa05:40
*** k4n0 has joined #openstack-qa05:43
*** leseb has quit IRC05:45
*** yfried has quit IRC05:50
*** ildikov has joined #openstack-qa06:00
afazekasThe instance failed to reach the Active state not reported properly https://review.openstack.org/#/c/101415/06:02
afazekasMany confusing exception appears on the console log06:02
afazekassdague,dkranz, mtreinish: ^^06:02
afazekasianw: ping06:05
*** nati_ueno has joined #openstack-qa06:09
*** yfried has joined #openstack-qa06:14
*** Longgeek has quit IRC06:30
*** Longgeek_ has joined #openstack-qa06:30
ianwafazekas: hi06:37
*** ghyoc has joined #openstack-qa06:38
*** leseb has joined #openstack-qa06:41
*** SridharG has joined #openstack-qa06:45
*** leseb has quit IRC06:46
*** jlibosva has joined #openstack-qa07:04
*** shardy has quit IRC07:12
*** qba73 has joined #openstack-qa07:21
*** vrovachev has joined #openstack-qa07:41
*** leseb has joined #openstack-qa07:42
*** leseb has quit IRC07:46
*** jpich has joined #openstack-qa07:57
*** ghyoc has quit IRC08:00
*** ghyoc has joined #openstack-qa08:02
*** safchain has joined #openstack-qa08:07
*** ghyoc has quit IRC08:13
*** ghyoc has joined #openstack-qa08:14
eezhovayfried: ping08:17
yfriedeezhova: pon08:17
eezhovayfried: concerning lbaas tests08:17
yfriedeezhova: yeah?08:17
eezhovayfried: I din't think they can be made faster. most of the time is consumed by booting and susepding/resuming an instance08:18
kashyapafazekas, Hi, I as following along these - https://review.openstack.org/#/c/100309/ and https://review.openstack.org/#/c/101110/08:19
yfriedeezhova: you could disable network for an instance if you want to check failover08:19
kashyapafazekas, Where can I read what does "experimental" mean08:20
kashyapafazekas, IIUC, that means, patches won't be gated on F20 jobs, isn't it?08:20
*** zz_johnthetubagu is now known as johnthetubaguy08:20
eezhovaeezhova: I could, but it would still take more than 60 sec. and in light of limited resources of infra and refactoring of lbaas itself we need to hold on with these tests08:21
eezhovayfried: ^ it was to you, I'm sorry08:22
yfriedeezhova: regardless of infra, it's better to simply shut down a network instead of suspending a VM if you simply want it out of the LB cycle08:23
yfriedeezhova: or better yet, set a firewall rule on that VM08:24
*** jordanP has joined #openstack-qa08:25
yfriedeezhova: also, if you boot VMs with a single command (using the NUM_INSTANCE keyword) it should trigger parallel booting instead of serial08:25
yfriedeezhova: but, putting aside the efficiency effort, these tests should be in the tree, even if they don't run for every single patch08:26
eezhovayfried: these are wise peaces of advice, thanks. but it is the decision of lbaas team that these tests shouldn't merge now. I'd gladly return to these tests once I have green light again :)08:26
yfriedeezhova: can you point me to that decision?08:27
yfriedeezhova: yesterday meeting discussed, shortly, "slow tests" gate for these kind of scenarios08:28
eezhovayfried: i discussed it with enikanorov lately08:28
eezhovayfried: if they changed their mind, that's great08:29
eezhovayfried: btw, could you clarify, how can I trigger a parallel booting?08:29
yfriedeezhova: who should I talk to, to get them to change their mind?08:29
yfriedeezhova: you should check the api08:29
yfriedeezhova: CLI has --num-instances08:30
eezhovayfried; ok, I will08:30
yfriedeezhova: don't know, off the top of my head, what the corresponding keyword would be08:30
yfriedeezhova: but this shifts the load of booting over to the server instead of the the client waiting for each VM to go 'ACTIVE'08:31
eezhovayfried: it seems that the lbaas team was also reorganized somehow, so I don't really know who is is in charge there now08:31
eezhovayfried: that sounds promising08:31
afazekaskashyap: you need to explicitly write a 'check experimental' in-order to run the experimental jobs08:39
kashyapafazekas, Write where?08:39
afazekasgerrit comment08:39
kashyapafazekas, Ah I see. Okay08:39
*** leseb has joined #openstack-qa08:42
eezhovaafazekas: Could you please take a look at https://review.openstack.org/#/c/98065/ ? It improves the stability of the load_balancer_basic test08:47
*** leseb has quit IRC08:47
afazekaseezhova: do you know how that bad status line looks like ?08:49
afazekasis it an 504 response ?08:50
afazekaseezhova: I believe it improves the stability, but I would like to better understand what really happening there08:51
*** oomichi has quit IRC08:53
*** oomichi has joined #openstack-qa08:54
eezhovaafazekas: I can's say for sure about the code. But the trace looks like that https://bugs.launchpad.net/tempest/+bug/1326607. It sometimes happens when haproxy receives a response from a member but then suddenly drops the connection08:54
*** oomichi has quit IRC08:55
eezhovaafazekas: usually it is fixed by changing haproxy's configuration (increasing quota for connection), but in any case the situation is still weird because the number of requests has already been decreased to a minimum.08:56
eezhovaafazekas: what's the most important, the problem never comes from a backend - they always give the right response08:58
afazekas505 would not cause 504, so the content probably in invalid http response08:58
afazekaseezhova: is it possible the backend is not ready for new connection ?08:59
afazekasand the HA proxy tries to maintain more connection08:59
eezhovaafazekas: no, as I said, it always is ready. tcpdump showed that backends always operated correctly09:00
afazekaseezhova: is it bug in HaProxy ?09:01
*** leseb has joined #openstack-qa09:01
eezhovaafazekas: I consulted with members of lbaas team and they say that it seems so09:02
eezhovaafazekas: I ran the test with this patch 200 times in a row and it never failed09:02
eezhovaafazekas: so I hope, that it will finally grant the test stability09:03
*** chandan_kumar has quit IRC09:09
*** aimon_ has quit IRC09:11
*** nati_ueno has quit IRC09:13
*** yamahata_ has quit IRC09:13
afazekaswhat was the issue with the inetd version ?09:18
*** leseb has quit IRC09:18
*** gszasz has joined #openstack-qa09:18
*** andreaf_ has joined #openstack-qa09:18
eezhovaafazekas: as it turned out, the case was the same. the only difference is that inetd backend always sends RST+ACK packets in response to SYN09:26
*** davidlenwell has quit IRC09:27
eezhovaafazekas: BadStatusLine just appears regardless of the backend09:27
* afazekas I will check the nc behavior on cirros09:28
*** qba73 has quit IRC09:34
*** amotoki has quit IRC09:34
*** qba73 has joined #openstack-qa09:35
*** qba73 has quit IRC09:35
*** jamielennox is now known as jamielennox|away09:35
*** aimon has joined #openstack-qa09:39
*** GMann has quit IRC09:47
afazekass/505 would not cause 504/505 would not cause BadStatusLine/09:50
*** davidlenwell_ has joined #openstack-qa09:52
*** davidlenwell_ is now known as davidlenwell09:52
*** davidlenwell is now known as davidlenwell_09:52
*** davidlenwell_ is now known as davidlenwell09:52
*** tosky has joined #openstack-qa09:55
*** yfried has quit IRC09:59
*** yjiang|2 has quit IRC10:02
eezhovaafazekas: http://postimg.org/image/o0rf7p1f7/ Here is an example of unusual behavior of HaProxy (see highlighted packets). 10.0.0.2 is a backend, 10.0.0.5 - VIP, 172.24.4.1 - router. the backend sends the VIP a correct reply, then vip sends ACK to the backend, closes connection with it and then closes connection with router. Normally, after sending ACK to an instance, vip should send the response to a router and only then close connections.10:02
afazekasnc with parallel nc client seams stable10:02
*** samuelbercovici has joined #openstack-qa10:05
afazekaseezhova: what is visible in the follow tcp stream ?10:08
*** yfried has joined #openstack-qa10:11
afazekasthe server1 and server2 string is 7 byte, but the content length is 8 byte10:16
*** leseb has joined #openstack-qa10:19
*** coolsvap is now known as coolsvap|afk10:19
afazekaseezhova: can you confirm or deny the content length is incorrect in the test case ?10:22
*** leseb has quit IRC10:23
*** coolsvap|afk is now known as coolsvap10:31
*** cdent has joined #openstack-qa10:32
openstackrecheckopenstack/heat change: https://review.openstack.org/100701 failed because of: gate-tempest-dsvm-neutron: https://bugs.launchpad.net/bugs/125152110:38
*** eezhova has left #openstack-qa10:42
*** eezhova has joined #openstack-qa10:42
eezhovaafazekas: you are right, the content length should be 7 bytes instead of 8 (it's a leftover from a version where a expected response was server1\n, server2\n).10:47
eezhovaafazekas: the stream gives an example of a situation when BadStatusLine exception appears10:48
eezhovaafazekas: have you managed to reproduce this situation?10:49
*** Longgeek_ has quit IRC10:49
afazekaseezhova: I have removed all badstatusline capture and increased the request to 1000, in this I have 50% chance to get the exception (4 try)10:52
afazekasFixing the content-Length does not helps10:52
eezhovaafazekas: content length should be fixed, yes, but it is not relevant here10:53
* afazekas launch break , than wireshark ..10:53
eezhovaafazekas: I can provide you with a script i used to run the test with tcpdump if you wish10:54
*** openstackgerrit has joined #openstack-qa10:55
eezhovaafazekas: http://paste.openstack.org/show/84561/10:58
*** samuelbercovici has quit IRC11:00
openstackgerritVadim Rovachev proposed a change to openstack/tempest: Added cinder and neutron notifications tests for ceilometer  https://review.openstack.org/6716411:01
*** Longgeek has joined #openstack-qa11:02
*** yamahata_ has joined #openstack-qa11:02
*** nosnos has quit IRC11:03
eezhovaafazekas: I'm sorry that was the wrong script. That's the one: http://paste.openstack.org/show/84562/11:06
*** mkollaro has joined #openstack-qa11:17
*** leseb has joined #openstack-qa11:19
*** leseb has quit IRC11:24
*** coolsvap is now known as coolsvap|afk11:31
*** sweston has joined #openstack-qa11:32
*** sweston has quit IRC11:38
openstackgerritSean Dague proposed a change to openstack/qa-specs: add javelin2 specification  https://review.openstack.org/9644511:39
*** sweston has joined #openstack-qa11:43
*** Poornima has quit IRC11:46
*** Madan has quit IRC11:50
openstackgerritSridhar Gaddam proposed a change to openstack/tempest: Utility API to generate EUI-64 IPv6 address  https://review.openstack.org/10149411:56
*** leseb has joined #openstack-qa12:01
*** weshay has joined #openstack-qa12:01
openstackrecheckopenstack/nova change: https://review.openstack.org/91727 failed gate-grenade-dsvm-partial-ncpu in the gate queue with an unrecognized error12:04
*** leseb has quit IRC12:06
*** jaypipes has joined #openstack-qa12:21
*** ChanServ sets mode: +o jaypipes12:21
*** jaypipes is now known as leakypipes12:22
openstackgerritSean Dague proposed a change to openstack/tempest: Remove assertTrue from ssh unit tests  https://review.openstack.org/10151112:22
openstackgerritSean Dague proposed a change to openstack/tempest: Remove assertTrue from ssh unit tests  https://review.openstack.org/10151112:25
sdaguemtreinish: maybe we should have a local hacking rule that prevents the use of assertTrue without the 2nd parameter12:27
*** salv-orlando has quit IRC12:28
*** ekhugen has joined #openstack-qa12:29
*** k4n0 has quit IRC12:30
*** yamahata has quit IRC12:39
mtreinishsdague: pretty much all the asserts basically have non-descript default failure messages. I'm sure you remember 2 != 312:48
mtreinishbut I wouldn't be opposed to a hacking rule enforcing error msgs12:48
sdaguemtreinish: sure12:48
sdaguebut at least you would vaguely know what the 2 values were12:48
sdagueassertTrue is really problematic12:48
sdaguebecause I actually don't have any idea what was in the comparison12:48
sdaguewas it 3.000 < 312:49
*** malini1 has quit IRC12:50
openstackgerritA change was merged to openstack/qa-specs: add javelin2 specification  https://review.openstack.org/9644512:51
*** eezhova_ has joined #openstack-qa13:00
*** leseb has joined #openstack-qa13:02
*** dkorn has joined #openstack-qa13:03
*** qba73 has joined #openstack-qa13:04
*** dkranz has joined #openstack-qa13:04
openstackgerritVadim Rovachev proposed a change to openstack/tempest: Add glance notifications tests for ceilometer  https://review.openstack.org/7125913:06
mtreinishooh tempest bugzilla got spammed last night :)13:06
mtreinisherr lp bugs13:06
*** dpaterson has joined #openstack-qa13:06
*** leseb has quit IRC13:07
openstackgerritAlok Kumar Maurya proposed a change to openstack/tempest: Allowed Address Pair API Tests Enhancements  https://review.openstack.org/10152213:08
afazekaseezhova_: reproduced without neutron: http://fpaste.org/111463/26974914/13:10
eezhova_afazekas: that seems to prove, that it is a bug in haproxy, doesn't it?13:11
afazekaseezhova_: looks like, I tried to found evidences it assumes connection keep-alive , but I did not found13:12
afazekaswith http/1.0 the connection: close  is expected default, I also tryed to add it to the response header13:13
dkranzmtreinish: I sent a message to infra about the spamming earlier13:13
eezhova_afazekas: I feel, that if we want to know for sure what happens there we have study haproxy code. But I have seen a bit of it and to tell you the truth it's not something I really want to debug:(13:16
mtreinishdkranz: yep I just saw that thread13:16
afazekaseezhova_: the only thing what nc does, and normal server would not , it starts responding before the request, and may be even not consumes the request (If you see ot on the wire does not means the app actually read it)13:18
openstackgerritAlok Kumar Maurya proposed a change to openstack/tempest: Allowed Address Pair API Tests Enhancements  https://review.openstack.org/10152213:19
afazekaslets see what happens with an apache back-end13:19
*** dansmith is now known as superdan13:24
*** malini1 has joined #openstack-qa13:26
leakypipesdkranz: hey, just FYI, I filed a bunch of new bugs in tempest. I made the descriptions of the bugs a little bit of a word puzzle, just for you. example: "How The Turn Metric Sewing Can Helpfulness Retrogress Metric Speedily"13:27
leakypipesdkranz: hope you enjoy!13:27
afazekaseezhova_: can you try to create an inet based version in a way which actually consumes the request before responding ?13:30
afazekaswith apache the haproxy seams ok13:30
*** mriedem has joined #openstack-qa13:31
mtreinishleakypipes: are you still at? Because we just got some more :)13:32
*** ghyoc has quit IRC13:33
leakypipesmtreinish: yup. just added a bug about Tempest not effectively losing weight enough. Left a riddle in there for you specifically: "If yes, added ponder is, fuck you e'er proven a metric decline set that is marketed for its slimming effects upon its depletion?"13:34
leakypipesmtreinish: oh... and, you're welcome.13:35
openstackgerritAlok Kumar Maurya proposed a change to openstack/tempest: Allowed Address Pair API Tests Enhancements  https://review.openstack.org/10152213:35
mtreinishleakypipes: heh, then I'm assuming that's really about cinder then :)13:38
leakypipesmtreinish: lol!13:39
leakypipesmtreinish: you just made my day. thx :)13:39
afazekaseezhova_: sh -c " " can be used with the nc, so a more tricky shell script could wait for the request13:42
eezhova_afazekas: Alright, I will try this. Than you for your expertise13:44
* afazekas a version with an added ' sleep 1', seams ok so far .. 13:45
*** rbak has joined #openstack-qa13:49
afazekaseezhova_: you are welcome13:49
*** eezhova_ has quit IRC13:52
* afazekas when the haproxy under strace the simple sleep 1 is not enough13:52
*** sweston has quit IRC13:53
openstackgerritA change was merged to openstack/tempest: Updated from global requirements  https://review.openstack.org/9682913:55
*** sweston has joined #openstack-qa13:56
openstackgerritDavid Shrewsbury proposed a change to openstack/tempest: Add ironic instance rebuild advanced test  https://review.openstack.org/9443914:00
*** gszasz has quit IRC14:03
*** leseb has joined #openstack-qa14:03
*** comstud is now known as bearhands14:04
*** leseb has quit IRC14:07
*** kgriffs|afk is now known as kgriffs14:11
*** jlanoux has joined #openstack-qa14:14
openstackrecheckopenstack/tempest change: https://review.openstack.org/97834 failed gate-tempest-python27 in the gate queue with an unrecognized error14:24
*** qba73 has quit IRC14:28
*** bmwiedemann has joined #openstack-qa14:30
*** mrodden has joined #openstack-qa14:32
*** yamahata has joined #openstack-qa14:32
*** rwsu has joined #openstack-qa14:35
*** nati_ueno has joined #openstack-qa14:37
*** sabeen has joined #openstack-qa14:38
mriedemmtreinish: so looks like that min version check for CLI is getting no stdout back when calling 'nova --version'14:39
mriedemor 'heat --version'14:40
mriedemnot really sure why that is14:40
mriedemthere are cli tests that check stdout right?14:40
mtreinishmriedem: yeah most of them check stdout14:40
mtreinishto confirm formatting for things like lists14:40
mriedemhmm, wondering if it's something with not using a released client package?14:41
mriedemAdding python-novaclient 2.17.0.167.gfa2f38a to easy-install.pth file14:43
mriedemhmm, maybe because i'm not running it with credentials14:47
mriedemdidn't think that was necessary for --version, like --help14:47
mtreinishmriedem: yeah it shouldn't need creds14:47
mtreinishI think that was the bug that prompted adding the cli tests actually14:48
mtreinishit would stack trace on --version if you didn't give it creds14:48
vrovachevHello mtreinish, sdague, afazekas, dkranz, masayukig, please review me: https://review.openstack.org/#/c/71259/14:49
*** bmwiedemann has left #openstack-qa14:52
afazekaseezhova_: this  (the cat part), lets to the ha-proxy to FIN or RST the connection first  http://fpaste.org/111505/40327602/14:54
*** mmaglana has joined #openstack-qa14:55
*** chandan_kumar has joined #openstack-qa14:55
*** atiwari has joined #openstack-qa14:57
*** salv-orlando has joined #openstack-qa14:58
*** malini2 has joined #openstack-qa14:59
*** malini1 has quit IRC15:02
*** leseb has joined #openstack-qa15:04
*** dustins has joined #openstack-qa15:04
afazekaseezhova: ^^15:05
openstackgerritA change was merged to openstack/tempest: make logformat more similar to oslo  https://review.openstack.org/9881915:06
afazekasI am checking which part does not respected the http spec or the common sense, but the 'cat >/dev/null'  at the end makes it stable with 0 (zero)  BadStatusLine exception (I removed all related catch)15:07
*** leseb has quit IRC15:08
*** Longgeek has quit IRC15:09
afazekasvrovachev: at the first look seams ok, I will have another look15:10
vrovachevafazekas: thank you so much :)15:11
openstackgerritSabeen proposed a change to openstack/tempest: Create and delete a stack using a HOT template  https://review.openstack.org/9725715:12
*** mkollaro has quit IRC15:19
*** mkollaro has joined #openstack-qa15:20
*** johnthetubaguy is now known as zz_johnthetubagu15:21
*** tosky has quit IRC15:26
*** zz_johnthetubagu is now known as johnthetubaguy15:26
*** johnthetubaguy is now known as zz_johnthetubagu15:27
*** zz_johnthetubagu is now known as johnthetubaguy15:28
*** johnthetubaguy is now known as zz_johnthetubagu15:34
*** cbader has joined #openstack-qa15:38
afazekaseezhova: The nc was rude by closing both end of the connection (RST) in situation where the clients(proxy) main responsibility to close the connection, however the ha proxy SHOULD tolerate it15:43
*** dkorn has quit IRC15:43
*** aimon has quit IRC15:53
*** aimon has joined #openstack-qa15:54
*** zz_johnthetubagu is now known as johnthetubaguy15:54
*** aimon has quit IRC15:58
*** mriedem has quit IRC15:58
*** ghyoc has joined #openstack-qa16:00
*** vponomaryov has quit IRC16:01
*** johnthetubaguy is now known as mr_oom-pah16:03
*** jlanoux has quit IRC16:03
openstackgerritAlok Kumar Maurya proposed a change to openstack/tempest: Allowed Address Pair API Tests Enhancements  https://review.openstack.org/10152216:04
*** TravT has joined #openstack-qa16:04
*** leseb has joined #openstack-qa16:04
*** mriedem has joined #openstack-qa16:05
afazekassdague: https://review.openstack.org/#/c/101415/16:08
*** jlibosva has quit IRC16:08
*** leseb has quit IRC16:09
*** jogo is now known as flashgordon16:15
*** ildikov has quit IRC16:18
*** aimon has joined #openstack-qa16:18
openstackgerritA change was merged to openstack-dev/devstack: cinder: use non-deprecated notification_driver  https://review.openstack.org/10041816:20
sabeenHi afazekas - when you have a few minutes, can you review my code: https://review.openstack.org/#/c/97257/16:30
*** jpich has quit IRC16:31
flashgordonsdague: I like your hacking cleanup patches16:31
*** BobBall is now known as BobBall_Away16:31
flashgordonsdague: but this needs to go the the ML IMHO16:31
*** cdent has quit IRC16:33
*** cdent has joined #openstack-qa16:34
*** nati_ueno has quit IRC16:36
*** yfried has quit IRC16:40
*** vrovachev has quit IRC16:43
*** cdent has quit IRC16:47
*** mestery has joined #openstack-qa16:53
*** mlavalle has joined #openstack-qa16:53
*** salv-orlando has quit IRC16:54
*** mlavalle_ has joined #openstack-qa16:56
*** mlavalle has quit IRC16:58
*** mlavalle_ is now known as mlavalle16:58
*** jordanP has quit IRC16:59
*** mestery has quit IRC16:59
*** mestery has joined #openstack-qa17:00
*** harlowja_away is now known as harlowja17:00
*** mr_oom-pah is now known as johnthetubaguy17:01
*** mestery has quit IRC17:04
*** leseb has joined #openstack-qa17:05
*** ildikov has joined #openstack-qa17:09
*** leseb has quit IRC17:10
*** nati_ueno has joined #openstack-qa17:17
*** wramthun has quit IRC17:23
*** wramthun has joined #openstack-qa17:23
*** mike77 has joined #openstack-qa17:26
sdagueflashgordon: I'm fine with that :)17:29
sdagueI was wondering how much of a rumble it would be, but figured I'd just write those patches17:29
*** mkollaro has quit IRC17:40
openstackgerritMorgan Fainberg proposed a change to openstack-dev/devstack: Update Apache + mod_wsgi deployment mechanisms  https://review.openstack.org/10161117:40
*** ekhugen has quit IRC17:42
openstackgerritMatthew Treinish proposed a change to openstack-infra/elastic-recheck: Add query for bug 1332512  https://review.openstack.org/10161217:42
openstackrecheckopenstack/cinder change: https://review.openstack.org/60947 failed gate-cinder-python27 in the gate queue with an unrecognized error17:43
*** mspreitz has joined #openstack-qa17:44
sdagueflashgordon: though does that mean there will be an ML thread for any new rules?17:46
*** johnthetubaguy is now known as zz_johnthetubagu17:47
*** mriedem has quit IRC17:47
*** ekhugen has joined #openstack-qa17:48
*** yamahata has quit IRC17:48
*** SridharG has quit IRC17:51
flashgordonsdague: any changes to HACKING.rst that add/remove a rule, yes17:53
*** salv-orlando has joined #openstack-qa17:54
openstackgerritSabeen proposed a change to openstack/tempest: Create and delete a stack using a HOT template  https://review.openstack.org/9725717:56
sdagueI think we should also then put out something any time a rule goes from culture to code17:56
sdaguebecause I think H305 is an instance where that should have been caught earlier17:57
flashgordonsdague: in theory everything in HACKING.rst should be fare game as we should be looking at that carefully, but in reality yeah17:58
afazekasvrovachev: Do you have any idea why it is not working with qpid devstack setups ?17:58
flashgordonsdague: speaking of which you may want to go through hacking.rst and go through the list of things we don't enforce and see what you don't like17:59
sdagueflashgordon: not really, I just want to dial back the auto enforcement of silly stuff18:00
afazekassdague: https://bugs.launchpad.net/tempest/+bug/133241418:00
flashgordonsdague: I think we should drop http://docs.openstack.org/developer/hacking/#dictionaries-lists18:03
flashgordonand possibly Calling MethodsĀ¶18:03
flashgordonas we don't enforce either of those now, and unclear to me if we want to in the future.18:04
flashgordonso hacking the code, just enforces HACKING.rst which is the cannonical source of for hacking rules18:04
sdagueI think the point is we should consider some of these recommendations18:05
sdagueand yell at people that only -1 for them18:06
*** leseb has joined #openstack-qa18:06
sdaguebecause I think the crux is believing that these rules alone make the code more readable. When instead most of the times I'd be -1ing people for these kinds of things it would be because in reality the entire patch is terrible and non understandable. :)18:07
flashgordonsdague: so I agree that hacking doesn't inherently mean the code is readable18:08
flashgordonbut I am hesitant to have recomendations that we don't enforce. I would rather just nix them18:08
sdagueI completely disagree18:08
flashgordonunless we clearly say one shouldn't -1 for those things in bold18:08
sdaguebecause recommendations are good18:08
flashgordonthey are but then we get into bikeshedding in reviews18:08
sdaguethese are examples of better written code18:08
sdagueonly if you let the bikeshedders get away with it18:09
flashgordonat the very least though if you think some sections should only be recomendations and not enforced we should clearly say that18:09
flashgordonsdague: true, but never underestimate our ability to bike shed, see periods in commit messages for example18:10
sdagueyeh, that was silliness18:10
sdagueby an unreasonable reviewer18:10
*** leseb has quit IRC18:10
sdagueand I addressed that in other projects where hacking is not enforced by yelling at them :)18:10
sdagueI think that's effective18:10
flashgordonanyway, while I am not a fan of leaving in sections as only recommendations. I am fine with going down that route (we can always re-evaulate later if needed)18:11
flashgordonyelling is always fun18:11
sdagueCalls to methods 80 characters or longer should format each argument with18:11
sdaguenewlines. This is not a requirement, but a guideline::18:11
openstackgerritMorgan Fainberg proposed a change to openstack-dev/devstack: Update Apache + mod_wsgi deployment mechanisms  https://review.openstack.org/10161118:11
sdaguewe already call out a lot of that explicitly in the file18:11
flashgordonyeah, it does make things easier to read usually18:12
sdagueactually, I find that convention usually screws us18:12
sdaguewe should really pull the first param back down18:12
sdaguewe have a ton of stuff where we are wrapping inside a param because our function openning doesn't start until col 6518:13
flashgordonohh yeah that is wrong IMHO18:13
flashgordonyour right18:13
sdaguebut, again, I would consider that a recommendation18:14
flashgordonso that works for me, but if think that lets update HACKING.rst18:14
flashgordoninstead of keeping that as oral tradition18:14
sdaguethat's fine18:15
openstackgerritMorgan Fainberg proposed a change to openstack-dev/grenade: Ensure upgrade testing tests same deployment methods  https://review.openstack.org/10076418:16
clarkbsdague: you are my hero18:16
sdagueclarkb: for this? :)18:17
clarkbyes H803 is well :(18:17
flashgordonclarkb: H803 is crazy pants18:18
sdagueI know why it's there, I know all the history, I'm ready for us to be past it18:19
openstackgerritMorgan Fainberg proposed a change to openstack-dev/devstack: Enable Keystone under Apache by default  https://review.openstack.org/10074718:20
flashgordonamen18:20
clarkband more generally I think hacking has become more about lets enforce all sorts of things that are controversial to avoid properly dealing with those prolems18:21
clarkbif we haev something that should be enforced lets do it18:21
clarkbbut location of a period?18:22
clarkbmeh18:22
sdagueclarkb: agreed18:22
sdagueI actually think a good policy would be only to adopt up local hacking rules that projects build18:22
flashgordonclarkb: do you have other examples?18:22
clarkbflashgordon: the docstring thing is good18:23
sdaguebecause then we know those rules are really useful for at least one project18:23
flashgordonsdague: a lot of local hacking rules are strange18:23
clarkbthe capitalization rules too18:23
sdagueflashgordon: I'm not saying they all come up18:23
flashgordonclarkb: so those were there from day one in HACKING.rst18:23
sdaguebug the apache license one actually came up through tempest18:23
clarkbflashgordon: I don't care if they are in hacking.rst or not18:23
sdagueas provably useful18:23
flashgordonclarkb: not saying they should be there18:23
clarkbthey can still be there and not be enforced by default18:23
clarkbI actualyl don't care if they are there or not18:24
clarkbbut there is a difference between this is absolutely required and this is the proper way to do it18:24
flashgordonclarkb: so if they are sinply recomendations that we will never -1 on, then leave em in18:24
flashgordonbut we need to document that18:24
sdagueor pretend people aren't robots :)18:25
flashgordonso one of the things on my big list of TODOs is to update HACKING.rst with all the whys behind each rule18:26
sdaguebtw - HP poughkeepsie is rather pleasant today - https://twitter.com/sdague/status/48005417251636019218:27
flashgordonwhere is the other HP poughkeepsie employee?18:28
mtreinishI'm at home18:28
flashgordonhaha18:28
sdagueI invited him over :)18:31
mtreinishsdague, flashgordon: https://review.openstack.org/#/c/101612/ drop a few from the uncategorized list18:33
mtreinishI hit it in the gate18:33
flashgordonsdague: apparently working from home sounded better then going into the office18:33
sdaguemtreinish: there is some weird whitespace at the end of it18:34
sdaguecan you clean that fast?18:34
mtreinisheww, where did that come from18:34
flashgordonmtreinish: does tempest gate on  python26?18:34
sdagueflashgordon: no18:34
openstackgerritMatthew Treinish proposed a change to openstack-infra/elastic-recheck: Add query for bug 1332512  https://review.openstack.org/10161218:35
flashgordonneato18:35
mtreinishflashgordon: yeah tempest doesn't officially support py26 I think it's in the readme :)18:35
flashgordonthats what I thought, just checking18:37
*** markmcclain has joined #openstack-qa18:38
openstackgerritA change was merged to openstack-infra/elastic-recheck: Add query for bug 1332512  https://review.openstack.org/10161218:38
*** mestery has joined #openstack-qa18:40
flashgordonsdague: so the only thing missing from the HP poughkeepsie office is a tall pitcher of ice tea18:42
sdagueI should fix that and set out sun tea18:42
sdagueI'll go do that now18:43
openstackgerritDavid Paterson proposed a change to openstack/qa-specs: Add spec for bp:post-run-cleanup  https://review.openstack.org/9172518:44
*** devananda has quit IRC18:47
*** dustins has quit IRC18:48
*** dpaterson has quit IRC18:49
*** mestery has quit IRC18:50
*** dustins has joined #openstack-qa18:51
*** openstackrecheck has quit IRC18:53
*** openstackrecheck has joined #openstack-qa18:53
*** samuelbercovici has joined #openstack-qa18:54
*** ildikov has quit IRC18:54
*** mike77 has quit IRC18:55
*** harlowja is now known as harlowja_away18:57
openstackgerritmeera-belur proposed a change to openstack/qa-specs: QA spec for bp:add-basic-tests-for-barbican  https://review.openstack.org/9997819:00
*** harlowja_away is now known as harlowja19:01
flashgordonnice19:02
*** d-qoi has joined #openstack-qa19:02
*** d-qoi has left #openstack-qa19:03
*** d-qoi has joined #openstack-qa19:04
*** mspreitz has quit IRC19:06
*** Sukhdev has joined #openstack-qa19:06
*** leseb has joined #openstack-qa19:07
sdaguemrodden: I pushed the bashate rename stuff, as well as some other fixes for doing discovery19:08
sdagueif you want to take a look at some point19:08
*** leseb has quit IRC19:12
openstackgerritmeera-belur proposed a change to openstack/qa-specs: QA spec for bp:add-basic-tests-for-barbican  https://review.openstack.org/9997819:14
mroddensdague: sure was going to look into that today19:19
mroddenfighting some internal fires...19:20
*** samuelbercovici has quit IRC19:20
sdaguemrodden: I think the patch is good, so you just need to review it19:20
sdaguemrodden: good luck with the fires19:20
mroddensdague: just looked it over, i would be fine with merging it now, but do we want to wait on that?19:22
sdaguemrodden: I'm not sure there is any reason to wait unless there are other reviewers. The consumption of bash8/bashate is via pip, so until it releases we won't take the changes.19:25
sdagueI would like to get the discover pieces in before a release is cut, mostly because that will make tox integration much simpler19:25
mroddenright. just haven't renamed a project before so not sure what best practice is19:26
openstackrecheckopenstack/cinder change: https://review.openstack.org/85031 failed because of: gate-tempest-dsvm-full: https://bugs.launchpad.net/bugs/131365519:26
openstackgerritMorgan Fainberg proposed a change to openstack-dev/devstack: Update Apache + mod_wsgi deployment mechanisms  https://review.openstack.org/10161119:30
*** salv-orlando has quit IRC19:33
mroddensdague: commented19:34
mrodden.coveragerc19:34
sdaguecool19:36
*** mestery has joined #openstack-qa19:41
*** mestery_ has joined #openstack-qa19:44
*** mestery has quit IRC19:45
*** mestery_ is now known as mestery19:45
*** mike77 has joined #openstack-qa19:49
*** ildikov has joined #openstack-qa19:51
*** malini2 has quit IRC19:59
*** leseb has joined #openstack-qa20:00
*** salv-orlando has joined #openstack-qa20:01
*** ildikov has quit IRC20:02
openstackgerritMorgan Fainberg proposed a change to openstack-dev/devstack: Enable Keystone under Apache by default  https://review.openstack.org/10074720:03
*** aimon has quit IRC20:12
*** ildikov has joined #openstack-qa20:12
*** mmaglana has quit IRC20:13
*** IngoH has joined #openstack-qa20:20
*** ghyoc has quit IRC20:28
*** mestery has quit IRC20:28
*** markmcclain has quit IRC20:28
*** markmcclain has joined #openstack-qa20:39
*** markmcclain has quit IRC20:54
*** weshay has quit IRC20:54
*** jareking has joined #openstack-qa20:55
*** Sukhdev has quit IRC21:06
*** jamielennox|away has quit IRC21:08
*** Sukhdev has joined #openstack-qa21:12
*** leseb has quit IRC21:16
*** leseb has joined #openstack-qa21:16
*** d-qoi has quit IRC21:17
*** aimon has joined #openstack-qa21:17
*** leseb_ has joined #openstack-qa21:20
*** leseb has quit IRC21:21
*** mrodden has quit IRC21:24
*** cbader has quit IRC21:26
*** ekhugen has quit IRC21:28
*** mike77 has quit IRC21:30
*** mike77 has joined #openstack-qa21:33
openstackgerritHenry Gessau proposed a change to openstack-dev/devstack: Do not install openvswitch-datapath-dkms on Ubuntu 14.04  https://review.openstack.org/10165821:36
*** sabeen has quit IRC21:40
*** jareking has quit IRC21:41
flashgordonsdague: looks like your screen timeout bump worked nicely21:42
flashgordon7 commits away from breaking 500 this week21:42
sdagueyeh21:42
sdagueI have a retry loop that looks like it does it even better21:42
flashgordonif I didn't know better I would say these graphs cannot be right: http://jogo.github.io/gate/21:43
sdaguewell, they are artifacty21:43
sdagueerror bars and all21:43
flashgordonyes, but the overall trend21:43
sdaguealso it's been a pretty slow day21:43
flashgordonyeah it has, which is too bad I want to see us stress the system more21:44
flashgordonwe haven't even hit quota in a while21:44
flashgordonI think one of the takeaways here is that failing jobs really hurts throughput due to all the rechecks21:45
sdagueyep21:45
*** jareking has joined #openstack-qa21:50
openstackrecheckopenstack/keystone change: https://review.openstack.org/99218 failed gate-keystone-python27 in the gate queue with an unrecognized error21:55
*** mlavalle_ has joined #openstack-qa22:00
*** nati_ueno has quit IRC22:01
*** mlavalle has quit IRC22:01
*** mlavalle_ is now known as mlavalle22:01
openstackgerritMohammad Banikazemi proposed a change to openstack-dev/devstack: Adds spport for two config parameters  https://review.openstack.org/10166122:03
*** andreaf_ has quit IRC22:04
*** Sukhdev has quit IRC22:15
*** yamahata has joined #openstack-qa22:22
*** mkollaro has joined #openstack-qa22:23
*** mestery has joined #openstack-qa22:31
*** mestery has quit IRC22:31
*** mestery has joined #openstack-qa22:32
*** mike77_ has joined #openstack-qa22:40
*** morganfainberg is now known as morganfainberg_Z22:41
*** mike77 has quit IRC22:43
*** ianw has quit IRC22:58
openstackgerritDean Troyer proposed a change to openstack-dev/devstack: Move static docs into master branch  https://review.openstack.org/10166823:05
*** leseb_ has quit IRC23:06
*** leseb has joined #openstack-qa23:07
*** Sukhdev has joined #openstack-qa23:11
*** leseb has quit IRC23:11
openstackgerritA change was merged to openstack/tempest: Add branchless tempest section to README.rst  https://review.openstack.org/9783423:13
openstackgerritA change was merged to openstack/tempest: Add README section about API stability  https://review.openstack.org/9823023:14
*** nati_ueno has joined #openstack-qa23:15
*** mkollaro has quit IRC23:16
*** Sukhdev has quit IRC23:18
*** torandu has joined #openstack-qa23:21
*** torandu has quit IRC23:21
*** mestery has quit IRC23:21
*** torandu has joined #openstack-qa23:21
*** torandu has quit IRC23:21
*** torandu has joined #openstack-qa23:21
*** kgriffs is now known as kgriffs|afk23:27
openstackrecheckopenstack-infra/devstack-gate change: https://review.openstack.org/98365 failed because of: gate-grenade-dsvm-partial-ncpu: https://bugs.launchpad.net/bugs/1298472 https://bugs.launchpad.net/bugs/1312199 https://bugs.launchpad.net/bugs/125489023:28
*** TravT has quit IRC23:36
*** nati_ueno has quit IRC23:44
*** mrodden has joined #openstack-qa23:47
*** locke105 has quit IRC23:48
*** mlavalle has quit IRC23:55

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