Wednesday, 2016-08-03

*** gildub has joined #rdo00:00
*** kaminohana has quit IRC00:01
*** rlandy has quit IRC00:12
*** Son_Goku has joined #rdo00:20
*** tumble_ has quit IRC00:22
*** jdennis has quit IRC00:29
number80apevec: ack00:30
apevecnumber80, nack for you not sleeping00:30
number80apevec: well, hallway talks are still running00:31
apevecah then good00:31
apevechallway is the most important conference track :)00:32
*** jdennis has joined #rdo00:32
number80one hour ago, we were trying to split koji client important bits in a separate module rather than importing CLi as a module00:32
*** sarath has quit IRC00:34
*** apetrich has quit IRC00:40
*** apetrich has joined #rdo00:42
*** dprince has quit IRC00:44
*** limao has joined #rdo00:45
*** aortega has quit IRC00:46
*** limao has quit IRC00:47
*** limao has joined #rdo00:47
*** richm has quit IRC00:50
*** sdake has quit IRC00:57
*** sdake has joined #rdo00:59
*** limao has quit IRC01:00
*** limao has joined #rdo01:00
*** paragan has quit IRC01:17
*** apevec has quit IRC01:22
*** akshai has joined #rdo01:23
*** gfidente has quit IRC01:24
*** akshai has quit IRC01:35
*** saneax is now known as saneax_AFK01:37
*** sdake has quit IRC01:44
*** ohochman has quit IRC01:55
*** aortega has joined #rdo01:56
*** fzdarsky_ has joined #rdo01:58
*** anand_ts has joined #rdo01:59
anand_tsHi Guys, I just ran in to an error while packstack install . "Error: /Stage[main]/Keystone::Db::Sync/Exec[keystone-manage db_sync]: Failed to call refresh: keystone-manage  db_sync returned 1 instead of one of [0]"02:00
anand_tsetc/hosts is updated correctly. Any help is appreciated here02:00
*** fzdarsky has quit IRC02:02
anand_tsMore from logs "^[[1;31mError: Could not prefetch keystone_domain provider 'openstack': Execution of '/usr/bin/openstack domain list --quiet --format csv' returned 1: An unexpected error prevented the server from fulfilling your request. (HTTP 500) (Request-ID: req-6656a304-9b4b-4610-8748-0a070ccf3849)^[[0m ^[[1;31mError: Not managing Keystone_domain[Default] due to earlier Keystone API failures.^[[0m ^[[1;31mError: /Stage[main]/Keyst02:02
*** chandankumar has joined #rdo02:08
*** chandankumar has quit IRC02:08
*** gildub has quit IRC02:17
*** coolsvap has joined #rdo02:19
*** manous has quit IRC02:23
*** aortega has quit IRC02:31
*** crossbuilder has quit IRC02:36
*** crossbuilder_ has joined #rdo02:36
*** ohochman has joined #rdo02:38
*** manous has joined #rdo02:39
*** jhershbe has joined #rdo02:41
*** manous has quit IRC02:43
*** sasha2 has quit IRC02:47
*** paragan has joined #rdo02:55
*** sasha2 has joined #rdo02:55
*** manous has joined #rdo02:57
*** manous has quit IRC03:02
*** manous has joined #rdo03:14
*** gildub has joined #rdo03:15
*** manous has quit IRC03:19
*** ohochman has quit IRC03:21
*** manous has joined #rdo03:30
*** manous has quit IRC03:35
*** vimal has joined #rdo03:44
*** manous has joined #rdo03:46
*** ayoung has quit IRC03:48
*** Son_Goku has quit IRC03:49
*** Son_Goku has joined #rdo03:50
*** sdake has joined #rdo03:50
*** manous has quit IRC03:51
*** Son_Goku has quit IRC03:54
*** sdake_ has joined #rdo03:54
*** Son_Goku has joined #rdo03:54
*** sdake has quit IRC03:58
*** manous has joined #rdo04:03
*** anilvenkata has joined #rdo04:04
*** sdake_ has quit IRC04:05
*** manous has quit IRC04:08
*** links has joined #rdo04:10
ggilliesis anyone able to confirm if you need to have epel installed on either the undercloud or overcloud when using triple/rdo mitaka stable?04:12
*** manous has joined #rdo04:21
*** manous has quit IRC04:26
*** Alex_Stef has joined #rdo04:30
*** manous has joined #rdo04:38
*** vimal has quit IRC04:41
*** rdas has joined #rdo04:41
*** furlongm has quit IRC04:43
*** manous has quit IRC04:43
*** furlongm_ has joined #rdo04:43
*** gildub has quit IRC04:45
*** sasha2 has quit IRC04:46
*** hynekm has quit IRC04:47
*** sasha2 has joined #rdo04:49
*** lkuchlan has joined #rdo04:56
*** vimal has joined #rdo05:01
*** julim has quit IRC05:02
*** jaosorior has joined #rdo05:03
*** lkuchlan has quit IRC05:04
*** lkuchlan has joined #rdo05:04
*** furlongm_ has quit IRC05:05
*** furlongm_ has joined #rdo05:06
*** nehar has joined #rdo05:07
*** ihrachys has joined #rdo05:08
*** eliska has joined #rdo05:08
*** saneax_AFK is now known as saneax05:09
*** mbound has joined #rdo05:16
*** ihrachys has quit IRC05:17
*** mbound has quit IRC05:21
*** manous has joined #rdo05:21
*** Poornima has joined #rdo05:22
*** manous has quit IRC05:26
*** lon has quit IRC05:26
*** mburned has quit IRC05:27
*** rcernin has joined #rdo05:28
*** mburned has joined #rdo05:30
*** hynekm has joined #rdo05:30
*** yfried has joined #rdo05:37
*** tshefi has quit IRC05:37
*** manous has joined #rdo05:38
*** satya4ever has joined #rdo05:43
*** manous has quit IRC05:43
*** chandankumar has joined #rdo05:43
*** fragatina has joined #rdo05:45
*** ihrachys has joined #rdo05:46
*** MRrobotPR has joined #rdo05:46
*** MRrobotPR has quit IRC05:48
*** fragatina has quit IRC05:51
*** fragatina has joined #rdo05:52
mrungeggillies, I thought you should not enable it05:54
mrungethere have been a few incompatibilities with epel packages, iirc05:55
*** manous has joined #rdo05:55
*** julim has joined #rdo05:57
*** ihrachys has quit IRC05:58
*** gildub has joined #rdo05:59
*** manous has quit IRC06:00
*** ekuris_ has joined #rdo06:02
*** aufi has joined #rdo06:03
*** sdake has joined #rdo06:07
*** sdake has quit IRC06:07
*** sdake has joined #rdo06:07
ggilliesmrunge: that's what I thought but it looks like we still enable epel when building the images06:08
*** oshvartz has joined #rdo06:10
mrungeggillies, image building may be different; unfortunately I don't have insight there06:10
*** leanderthal|afk is now known as leanderthal06:12
*** manous has joined #rdo06:13
*** manous has quit IRC06:17
*** edannon has joined #rdo06:20
*** florianf has joined #rdo06:23
*** sdake has quit IRC06:23
*** mburned has quit IRC06:24
*** anand_ts has quit IRC06:26
*** flepied has joined #rdo06:27
*** Alex_Stef has quit IRC06:28
*** lkuchlan has quit IRC06:29
*** manous has joined #rdo06:30
*** lkuchlan has joined #rdo06:30
*** mburned has joined #rdo06:30
*** hewbrocca-afk is now known as hewbrocca06:31
*** manous has quit IRC06:34
*** milan has joined #rdo06:36
*** pgadiya has joined #rdo06:36
*** belmoreira has joined #rdo06:42
*** panda|Zz is now known as panda06:44
number80ggillies, mrunge: partly because, upstream image templates still relies on packages like dkms that we don't provide06:45
number80list is quite short but still annoying06:45
*** tesseract- has joined #rdo06:45
mrungethanks number80, good to know06:46
chandankumarnumber80, mrunge \o/06:47
*** ade_b has joined #rdo06:47
*** manous has joined #rdo06:47
number80ggillies: best person to ask about image building would be trown06:47
chandankumarnumber80, mock-rdo-config is not working for newton, the fedora-review is failing due to missing packages06:48
chandankumarnumber80, can we get a mock-config for rdo-trunk directly?06:48
*** paramite has joined #rdo06:48
number80chandankumar: I saw that and it's normal, we don't have rebuilt newton stable06:48
chandankumarnumber80, so will i directly dlrn.cfg for my blog?06:48
chandankumar*directly use06:49
*** gildub has quit IRC06:49
number80chandankumar: yes06:49
ggilliesnumber80: so the answer is yes? We need epel on the overcloud images?06:49
chandankumarnumber80, send a pr for rdo blog then06:49
number80ggillies: for now, yes, but it's likely due to cruft as dkms is mostly used within ubuntu images06:50
number80that's not something we -RH- support06:50
ggilliesright :|06:50
ggilliesso the images built in CBS don't have EPEL currently06:51
ggilliesI'm assuming they must be built differently?06:51
number80Yeah, but that's something that must be tweaked in our CI jobs, so trown should give you more insight on that06:51
*** manous has quit IRC06:52
*** Alex_Stef has joined #rdo06:55
*** toanju has joined #rdo06:59
*** dgurtner has joined #rdo06:59
*** anand_ts has joined #rdo07:02
anand_tsHi Guys, I am getting error while running packstack on centos 7 """Error: /Stage[main]/Keystone::Db::Sync/Exec[keystone-manage db_sync]: Failed to call refresh: keystone-manage  db_sync returned 1 instead of one of [0]"""07:03
anand_tsfrom install.log i can't get much information.07:03
anand_tsfrom keystone logs Error: Not managing Keystone_service[cinder] due to earlier Keystone API failures07:04
anand_tsAny help is much appreciated07:04
*** manous has joined #rdo07:04
*** jbadiapa has joined #rdo07:05
anand_tsMy server is behind a proxy server but I have put the entry in yum.conf.07:07
anand_tsI think I am hitting a bug here which is very old and fixed in previous releases of openstack07:08
*** manous has quit IRC07:09
*** ihrachys has joined #rdo07:14
anand_ts ERROR keystone OperationalError: (pymysql.err.OperationalError) (1045, u"Access denied for user 'keystone_admin'@'anats_mitaka' (using password: YES)")07:14
*** iranzo has joined #rdo07:14
anand_tsError from keystone logs.07:14
*** jaosorior is now known as jaosorior_brb07:15
*** jpena|off is now known as jpena07:16
*** garrett has joined #rdo07:16
*** dgurtner has quit IRC07:16
*** zoli_gone-proxy is now known as zoliXXL07:19
*** ohamada has joined #rdo07:19
number80fbo: small nitpick on your job review, but good for merging after that07:21
*** pcaruana has joined #rdo07:21
*** ushkalim has joined #rdo07:21
number80(I'm just an awk nerd)07:21
*** manous has joined #rdo07:22
*** dgurtner has joined #rdo07:23
*** dgurtner has joined #rdo07:23
*** links has quit IRC07:23
*** mcornea has joined #rdo07:24
*** hewbrocca is now known as hewbrocca-afk07:25
*** manous has quit IRC07:26
*** anshul has joined #rdo07:27
*** tshefi has joined #rdo07:29
zodbotnumber80: Karma for coolsvap changed to 1 (for the f24 release cycle):
coolsvapnumber80: :)07:30
*** jpich has joined #rdo07:35
*** links has joined #rdo07:36
rdogerritJavier Peña created openstack/nova-distgit: Set proper version for microversion-parse
*** furlongm_ has quit IRC07:38
number80jpena: funny that it didn't break CI earlier ^07:40
jpenanumber80: master is building 0.1.4 (pre-versioning, I guess)07:41
*** furlongm_ has joined #rdo07:42
number80makes sense, then there's case to check lowerbounds in package checks during reviews07:42
number80I mean, if there was no upstream 0.1.4 release at GA time, it could have been bad07:43
number80not critically bad, but still a small impediment to fix before release07:43
flepiedjpena: what's the bootstrap process?07:45
flepiedjpena: I'm using dlrn --stop --order --head-only07:46
jpenaflepied: on a first pass, run dlrn with --order07:46
jpenanow, we're fixing issues as we find them07:46
jpenaif we used --stop, it could take some more time07:47
flepiedjpena: yes but it's to be sure to not have to restart from the beginning07:47
*** enikher has joined #rdo07:49
*** hewbrocca-afk is now known as hewbrocca07:52
*** anand_ts has quit IRC07:53
*** Goneri has joined #rdo07:54
rdogerritMerged openstack/heat-distgit: fix building without -doc sub rpm
rdogerritMerged openstack/glance-distgit: Conditionalize -doc building
*** Goneri has quit IRC07:56
*** Goneri has joined #rdo07:56
*** devvesa has joined #rdo07:56
*** gfidente has joined #rdo07:57
*** gfidente has quit IRC07:57
*** gfidente has joined #rdo07:57
*** shardy has joined #rdo07:59
*** jistr is now known as jistr|training07:59
*** coolsvap is now known as coolsvap_08:08
*** zoliXXL is now known as zoli|brb08:08
*** nstrug__ is now known as nstrug08:10
*** limao_ has joined #rdo08:10
*** jaosorior_brb has quit IRC08:12
*** jaosorior_brb has joined #rdo08:13
*** limao has quit IRC08:13
*** dgurtner has quit IRC08:13
*** jlibosva has joined #rdo08:14
*** apevec has joined #rdo08:16
jpenamorning apevec :)08:17
apevechojla jpena08:17
rdogerritFabien Boucher proposed config: Fix the package listing plus install the freshly built packages
apevechm, nova failed No Package found for python-microversion-parse >= 0.1.408:18
apevecbut I see python-microversion-parse SUCCESS in report08:18
apevecah crap08:19
*** oshvartz has quit IRC08:19
apevecyeah., another pin side-effect08:19
jpenain this case it was a good side-effect, actually. There's no 0.1.4 microversion-parse08:20
apevecjpena, I guess you have seen in irc log, we'll need to start forking newton-rdo08:20
*** snecklifter has joined #rdo08:20
jpenaI saw the oslo.messaging issue and fix in rdoinfo08:20
apevecah where is 0.1.4 from then08:20
jpenapre-release versioning?08:20
jpenamaster is building 0.1.408:21
apevecyeah, but we shouldn't use that in Requires08:21
flepiedjpena: apevec: should we make tags=newton-uc by default in projects.ini to ease the use of dlrn?08:21
*** oshvartz has joined #rdo08:21
chandankumarjpena, getting this on each dlrn build locally in root.log08:21
chandankumarsome packages not found08:21
apevecflepied, yes, once we have first CI pass08:21
flepiedapevec: ok08:22
jpenachandankumar: are you using --dev? Can you paste your projects.ini file?08:22
*** jaosorior_brb is now known as jaosorior08:23
apevechmm, tags=08:24
apevecI forgot, what is default then, no filtering?08:24
jpenachandankumar: ok then, the projects.ini file is the default one, so we should be good. Just make sure you add --dev to the command line, or dlrn will try to use the local repo for dependencies08:24
chandankumarjpena, actually i am writting my blog for fedora-review where i am generating dlrn.cfg to pull any repo using dlrn08:25
chandankumarjpena, apevec number80
*** danpawlik has joined #rdo08:27
*** kbyrne has quit IRC08:29
chandankumarapevec, can i use this directly instead of ?08:29
chandankumarto generate dlrn.cfg08:30
chandankumarin project.ini file08:30
apevecnot really, we want to get rid of rdo-trunk-master on buildlogs08:30
apeveconly -tested should be left08:30
apevecand -tested is what should be used in dlrn.cfg08:30
apeveci.e. current-passed-ci08:31
*** dgurtner has joined #rdo08:31
*** fzdarsky_ is now known as fzdarsky08:32
*** ganesh has joined #rdo08:32
*** ganesh is now known as Guest7075408:33
*** Guest70754 is now known as gkadam08:33
*** Amita has joined #rdo08:34
rdogerritMerged openstack/nova-distgit: Set proper version for microversion-parse
*** pnavarro has joined #rdo08:37
*** zoli|brb is now known as zoli08:42
*** zoli is now known as zoliXXL08:42
*** lkuchlan has quit IRC08:42
*** priteau has joined #rdo08:48
*** jistr|training is now known as jistr08:49
*** ihrachys has quit IRC08:54
*** Ablu has quit IRC08:56
*** Ablu has joined #rdo08:57
*** Amita has quit IRC08:58
jpenaapevec: we need to pin mistralclient-distgit to a commit before
jpenawe could use rpm-mitaka08:59
*** dgurtner has quit IRC08:59
apevecah damn, how I didn't see that09:01
jpenatroubleshooting after midnight ;)09:01
apevecindeed, don't do that at home!09:01
apevecjpena, let's push newton-rdo for consistency, we can merge rpm-master once they have new release09:02
*** Ablu has quit IRC09:02
apevecto keep linear history09:02
jpenaapevec: ack, should I do it?09:02
*** Ablu has joined #rdo09:02
apevecyes please09:02
*** shardy is now known as shardy_mtg09:02
apevec+ rdoinfo change too09:03
rdogerritJavier Peña created rdoinfo: Pin newton-uc distro branch for mistralclient to newton-rdo
jpenaapevec: ^^09:06
*** lkuchlan has joined #rdo09:08
*** TSCHAK has quit IRC09:11
*** Amita has joined #rdo09:12
*** lkuchlan has quit IRC09:13
flepiedjpena: apevec: how could I have some stats about dlrn master? like the number of packages built per day, number of failures per day, ...09:13
rdogerritChristian Schwede created openstack/tripleo-common-distgit: Add missing scripts from tripleo-common/scripts
jpenaflepied: probably the easiest way is to query the database directly09:14
flepiedjpena: could you give me a copy?09:15
jpenaflepied: sure, give me a sec09:16
*** Amita has quit IRC09:16
*** mdnadeem has joined #rdo09:17
*** akrivoka has joined #rdo09:20
*** kbyrne has joined #rdo09:21
*** mflobo has joined #rdo09:21
jpenaflepied: email sent09:22
flepiedthx jpena09:22
*** dtantsur|afk is now known as dtantsur09:24
rdogerritMerged rdoinfo: Pin newton-uc distro branch for mistralclient to newton-rdo
apevecjpena, merged09:26
apevecah bot was faster :)09:27
jpenaapevec: ack, I've updated rdoinfo and will retry09:27
apevecjpena, have you rechecked nova ?09:27
apevecnova was in part209:27
jpenaapevec: yes, it worked fine09:27
apevecjpena, once all is built, you'll rsync to trunk.rdo and then we can ask Wes to setup temp CI pipeline against it09:29
jpenawe have a temp url for that,
*** Amita has joined #rdo09:30
rdogerritChristian Schwede proposed openstack/tripleo-common-distgit: Add missing scripts from tripleo-common/scripts
*** nyechiel has joined #rdo09:34
jpenadanpawlik: I've had a look at the watcherclient spec, added some comments in github09:36
*** paragan has quit IRC09:36
*** furlongm_ has quit IRC09:37
*** furlongm_ has joined #rdo09:37
*** yfried has quit IRC09:38
danpawlikjpena: thanks09:39
danpawlikjpena: I will check ASAP :)09:39
*** iberezovskiy has joined #rdo09:42
*** shardy_mtg is now known as shardy09:43
*** yfried has joined #rdo09:43
*** kambiz has joined #rdo09:44
*** zoliXXL is now known as zoli|lunch09:47
apevecany ideas what failed in ?09:55
jpenaapevec: 60-minute timeout, I think09:56
apevecyeah, that was my conclusion as well, but wanted 2nd opionon09:56
apevecrecheck it09:56
apevecthat's the only blocker left, need that in before running CI pipeline09:56
gfidenteapevec, there is a submission for puppet-nova failing gate on centos7 due to an old version of zaqar installed09:57
gfidenteapetrich, in centos7 puppet ci is at commit 1d0c95e while we need to bump it up to 8ef5fc9 to make it work09:57
apevecgfidente, link? what's the repo setup and what is old09:57
gfidenteapevec, ^^ can you help?09:57
apevecgfidente, so p-o-i job not tripleo ci?09:58
jpenalog file ends abruptly with no error indication, first log entry at 01:08:27.350637 and last one at 02:07:11.13604509:58
gfidenteapevec, not tripleo no09:58
*** Jaison has joined #rdo09:58
gfidentepuppet-nova change is, it needs in zaqar09:58
*** links has quit IRC09:59
apevecgfidente, p-o-i rdo trunk pin is blocked on the same issue as 27. in
apevecwhich should be fixed by
apevecwhich is where check failed on timeout I just discussed w/ jpena :)10:00
*** dgurtner has joined #rdo10:00
gfidenteapevec, so you're saying that if we get merged the pin will be updated to eventually include too?10:01
apevecpin bump is
*** nmagnezi has joined #rdo10:01
*** iranzo has quit IRC10:04
*** Goneri has quit IRC10:08
*** fragatina has quit IRC10:09
*** _degorenko|afk is now known as degorenko10:10
jpenawe got consistent on newton-uc, yay!10:11
*** jhershbe has quit IRC10:12
*** garrett has quit IRC10:14
*** sanjayu|AFK has quit IRC10:14
*** sanjayu|AFK has joined #rdo10:15
*** serverascode has quit IRC10:15
*** serverascode has joined #rdo10:16
*** tdasilva has quit IRC10:16
*** sasha2 has quit IRC10:18
*** garrett has joined #rdo10:20
*** jhershbe has joined #rdo10:24
*** Goneri has joined #rdo10:24
*** sasha2 has joined #rdo10:24
*** tosky has joined #rdo10:25
jpenaapevec: newton-uc is now visible at, ready for CI10:26
jpenaone warning: has a baseurl that points to centos7/ instead of centos7-new/10:27
jpenaI can manually edit it if needed10:27
apevecyes, edit it10:27
apevecat least some CI jobs d/l that repo10:27
apevecand some do construct repourl10:27
jpenadone (just for current/consistent)10:28
*** limao_ has quit IRC10:29
*** rodrigods has quit IRC10:30
*** rodrigods has joined #rdo10:30
apevecadarazs, myoung - could we run internal CI pipeline against ?10:31
apevecthat's newton with pinned versions10:31
apevec(oslo, clients etc)10:31
adarazsapevec: what is "internal CI pipeline"?10:32
apevecthe one myoung runs on RHEL10:32
adarazsapevec: oh, okay. myoung logs in a few hours later I think.10:32
adarazsapevec: and I'm trying to get the rdo ci promote working today by replacing the HA job :)10:32
apevecI mean, duplicate it for a temp test run, not change it for real yet!10:33
apevecadarazs, yes! that would be greate10:33
flepiedso 143 builds per day since the beginning of the newton cycle (more than 10K pkg built)10:33
apevecadarazs, how easy is to duplicate current master pippeline on ci.centos? Is it all jjb?10:34
apeveconce you have HA job replaced, I'd run it against centos7-new too, for double checking10:34
apevecbefore we switch centos7-master to u-c pins10:35
adarazsapevec: yes.10:35
adarazsapevec: it's more or less this:
apevecjpena, centos7-master is still hours behind? 10:25 build for 05:11 commit10:36
*** social has quit IRC10:36
apevecjpena, improving perfomance will be next prio after u-c pins are done10:37
jpenaapevec: it seems so. It's had some ups and downs (it was 4h behind yesterdau)10:37
apevecwe need to somehow parallelize it10:37
* jpena looks for some non-NFS storage10:37
*** dgurtner has quit IRC10:37
*** tdasilva has joined #rdo10:38
*** enikher has quit IRC10:38
*** tosky has quit IRC10:39
apevecwe were throwing ideas last week to trigger on post-commit from upstream and build in zuul then somehow (details TBD) collect builds centrally10:39
*** tosky has joined #rdo10:39
*** tosky has quit IRC10:39
apevecit would require decomposing current dlrn architecture10:39
*** Amita has quit IRC10:40
*** tosky has joined #rdo10:40
apevecdistributed builders but still central db+repo10:40
*** social has joined #rdo10:40
*** social has quit IRC10:40
*** aortega has joined #rdo10:41
*** tosky has quit IRC10:41
*** social has joined #rdo10:41
*** tosky has joined #rdo10:43
rookshardy do we have early builds of the composable roles working in RDO? (ie i can play with this now on baremetal?)10:43
apevecadarazs, so changing url in is all we need?10:44
apevecjpena, ^10:44
shardyrook: sorry, I'm not sure - assuming there are working RDO trunk or newton milestone builds then yes, but I've not tested them myself10:45
shardyrook: I'd check with trown|outtypewww when he comes online10:45
rookah ha, will do. thank shardy10:45
apevecshardy, he's out for 4 weeks10:45
shardyapevec: aha, so really really outtypewww then ;)10:45
apevecrook, you can look at git hash in RPM Release10:46
apevecand compare what's upstream10:46
jpenaapevec: good, but if we change that it could promote the repo, wouldn't it?10:46
apevecshardy, do you have THT hash where is this usable?10:46
shardyapevec: anything after the n-2 milestone should have the composable services architecture10:46
apevecjpena, we would not include that last step in the temp pipeline copy10:46
adarazsapevec: do I understand correctly that this is the "preparation to cut a newton branch from master" kind of thing?10:47
apevecshardy, rook - then current-passed-ci is good10:47
apevecit's from last week10:47
adarazsapevec: and it's already diverging from master, so we need to start promoting/gating for newton and master?10:47
shardyapevec: ack, thanks!10:47
apevecadarazs, it's
apevecit could be seen as prep for RC, but this is now10:48
*** milan has quit IRC10:48
apevecadarazs, we'll replace master pipeline with a variant where some projects are pinned to the versions from upstream upper-constraints10:48
apevecit won't be separate10:49
*** Jaison has quit IRC10:49
*** sasha2 has quit IRC10:49
*** tosky has quit IRC10:49
apevecmaster-of-all we'll keep running in the background as a pre-warning system, but will not be part of the promotion pipeline10:49
jpenaapevec: so your plan for the test is to create a copy of promote-master.yml with the required changes, right?10:50
rookohh that is right apevec -- I just saw him last week, duh10:50
apevecjpena, right10:50
apevecjpena, but only temp, so we can verify newton-uc10:51
adarazsapevec: okay, I'm starting to understand... isn't it a good time actually to create a newton promote pipeline then that will test this? and keep the master as that "master-of-all"?10:51
apevecif that works, we switch centos7-master and use current pipeline10:51
apevecadarazs, maybe10:51
apevecflepied, ^ what do you say?10:51
rookapevec sweet, so those packages - just use to get these packages?10:51
apevecwe could as well do this rename10:51
apevecactually, it is called centos-newton internally, we'll just replace symlinks on DLRN side10:52
flepiedapevec: extracted some stats: 143 builds per day since the beginning of the newton cycle (more than 10K pkg built)10:52
adarazsapevec: btw there are more things to do than just the JJB, but not much more.10:52
apevecbut we can as well rename job pipeline10:53
adarazsapevec: we need to add a different release for it.10:53
apevecflepied, only 10k ! :)10:53
*** links has joined #rdo10:53
*** tosky has joined #rdo10:54
rookapevec so, centos-newton? or will the RPM in the link i sent suffice?10:54
*** paragan has joined #rdo10:54
*** paragan has joined #rdo10:54
apevecrook, best to follow testday instructions10:54
rookapevec: roger.10:55
chandankumarjpena, number80 apevec
rookah ha, thanks apevec i was going to use milestone1 (maybe there isn't much difference).10:56
apevecyeah, it's same symlinks10:56
*** sasha2 has joined #rdo11:00
*** gszasz has joined #rdo11:04
*** kbyrne has quit IRC11:11
*** kbyrne has joined #rdo11:12
*** alexismonville has joined #rdo11:14
*** mflobo has quit IRC11:18
*** alexismonville has quit IRC11:19
apevecadarazs, jpena - " add a different release for it" where does that go and who can propose a patch to rdo-infra/ci-config ?11:21
*** lucasagomes is now known as lucas-hungry11:21
adarazsapevec: anybody can propose a patch, and I thought I got submit rights, but looks like trown|outtypewww didn't manage to add it actually before he left :)11:21
apevecit also wouldn't be pullurl cron triggered, only manually11:21
apevecadarazs, is that in gerrithub still?11:22
adarazsapevec: yes.11:22
adarazsapevec: also I guess we should build new images for this. so it might be a bit more involved. by adding a release I meant a yml file like this with the proper URLs:
apevecadarazs == your github id?11:23
adarazsapevec: yup11:23
apevecadarazs, hehe, done, added you as owner11:24
adarazsthanks :)11:24
adarazsyes, I see, thanks. :)11:24
*** Amita has joined #rdo11:24
apevecah, folders on artifacts, not sure how is that created11:24
*** rhallisey has joined #rdo11:25
apevecdmsimard will be able to help when he's online11:25
*** Amita has quit IRC11:25
*** amiii has joined #rdo11:25
adarazsapevec: I know how that works, should not be a problem. rsync should be able to create it when it uploads.11:25
apevecalso, that url was only for pullurl, jobs needs some parameter which repo to take11:26
apevecso centos7-master is encoded in each job's definition11:26
apevecthey just take hash from /tmp/delorean_master_hash11:27
adarazsapevec: yeah, and the build image bakes it into the images I think.11:28
*** mflobo has joined #rdo11:28
adarazsapevec: this is the image building script:
adarazsthat runs for promote11:29
adarazsmaybe they don't bake it into the image, just use that delorean_master_hash, right.11:29
*** dgurtner has joined #rdo11:29
* adarazs is a bit shaky around the image building part, only looked into it a bit.11:29
apevecthat's for oooq, weirdo jobs construct repourl11:30
hewbroccaadarazs: you're working on turning the RDO promote job green again, right?11:30
apevecok, so we'll need myoung and dmsimard to be online to sort this out for temp centos7-new11:30
adarazshewbrocca: yes.11:30
apevechewbrocca, it will be just a little bit cheating :)11:31
hewbroccaI'll take it however you get there11:31
apevechewbrocca, but we also have one unmerged blocker in puppet ci11:31
adarazswouldn't be cheating if we had stronger boxes in centosci.11:31
hewbroccaapevec: yeah I see that is behind as well11:31
apevecyeah, it's "little bit"11:31
adarazsor if we could cherry pick the strogn ones.11:31
*** aortega has quit IRC11:31
hewbroccathat is difficult for other reasons11:32
apevecadarazs, neither are an option atm, so your workaround is fine11:32
jpenaapevec: the puppet-tempest patch passed integration tests on centos, it's waiting for the ubuntu ones11:32
apevecwe'll recheck it until it passes :)11:32
apevecthen p-o-i bump11:33
* myoung waves at apevec and sips the days first coffee...11:33
*** kaminohana has joined #rdo11:33
hrwsomeone at flock?11:33
*** steveg_afk has quit IRC11:33
apevecmyoung, moin - enjoy coffee then take for an internal rhel-ride :)11:34
apevechrw, number80 is11:34
hrwapevec: thanks11:35
*** mflobo has quit IRC11:36
number80hrw: I'm at ralph bean's talk on Factory 2.011:37
* number80 got nicknamed "time cop" for making funny time reminder slides11:38
hrwnumber80: I am in Chełmoński looking at vagrant stuff11:38
number80ack, maybe we'll meet at the cruise tonight11:38
hrwI have green tshirt with 'there is no arm below armv7' text on it11:39
JehaneI'm hitting something very similar to this bug bug 1330289 in openstack-packstack "Failure to install Controller/Network&&Compute Cluster on RDO Mitaka with keystone API V3" [High,Post] - Assigned to jpena11:39
hrwnumber80: I have r/pi talk at 16:3011:40
Jehane(it's failing at the keystone module but the error message is similar)11:40
Jehaneany clue how to solve it ?11:40
number80hrw: I have a call at 4pm, but i finish early enough, I'll try to get in11:40
number80Jehane: still ldap issue?11:40
Jehanenumber80: yes, I want to re-run packstack to switch to v3 API11:41
Jehaneso I can use multi-domain authentication11:41
number80Jehane: faster would be just reconfiguring keystone service11:41
Jehanebut I should have done it manually11:41
*** alexismonville has joined #rdo11:42
JehaneI was afraid of having to pass in several different place to change it11:42
JehaneI have a meeting in 20 min, will try manually after (and I will pray that I haven't break everything)11:43
jpenaJehane: I'm not sure if installing with Keystone v2, then re-running Packstack to use v3 would work11:43
*** ohochman has joined #rdo11:43
number80Jehane: listen to jpena, he's packstack kingpin11:43
*** Poornima has quit IRC11:43
Jehanejpena: ok, thanks11:44
*** livelace has joined #rdo11:44
*** ppowell has joined #rdo11:44
*** sshnaidm is now known as sshnaidm|afk11:44
zodbotJehane: Karma for jpena changed to 4 (for the f24 release cycle):
*** coolsvap_ is now known as coolsvap11:45
*** dprince has joined #rdo11:45
*** limao has joined #rdo11:47
*** steveg_afk has joined #rdo11:47
*** danielbruno has joined #rdo11:47
*** danielbruno has quit IRC11:47
*** danielbruno has joined #rdo11:47
*** morazi has joined #rdo11:50
*** fultonj has joined #rdo11:51
*** rbowen has joined #rdo11:54
*** ChanServ sets mode: +o rbowen11:54
*** spr2 has joined #rdo11:58
*** limao has quit IRC11:59
*** yfried has quit IRC11:59
Jehaneok, meeting has changed, it's tomorrow11:59
*** limao_ has joined #rdo12:00
Jehanejpena: do you know if there is a clear tutorial somewhere on the internet on how switch to v3 ?12:00
*** zoli|lunch is now known as zoli12:01
*** zoli is now known as zoliXXL12:01
jpenaJehane: maybe
jpenaalso to update the keystonerc_admin and keystonerc_user files12:02
*** mlammon has joined #rdo12:04
*** milan has joined #rdo12:05
Jehaneok, thanks12:05
*** mosulica has joined #rdo12:05
*** spr has quit IRC12:05
*** lon has joined #rdo12:06
*** weshay has joined #rdo12:07
*** ccamacho is now known as ccamacho|lunch12:11
*** spr has joined #rdo12:12
*** toanju has quit IRC12:16
*** toanju has joined #rdo12:16
*** limao_ has quit IRC12:17
*** limao has joined #rdo12:17
*** marrusl has joined #rdo12:18
*** jpena is now known as jpena|lunch12:18
*** oshvartz has quit IRC12:19
*** aortega has joined #rdo12:19
*** dgurtner has quit IRC12:22
*** adarazs is now known as adarazs_brb12:23
*** dgurtner has joined #rdo12:27
*** dgurtner has joined #rdo12:27
*** jayg|g0n3 is now known as jayg12:28
*** vimal has quit IRC12:31
*** limao has quit IRC12:31
*** ade_b has quit IRC12:33
*** adarazs_brb is now known as adarazs12:34
*** sshnaidm|afk is now known as sshnaidm12:34
JehaneI think I will open a bug to swith packstack default to v3 for keystone :/12:36
*** dustins has joined #rdo12:36
*** sdake has joined #rdo12:37
*** dgurtner_ has joined #rdo12:39
*** dgurtner has quit IRC12:39
*** rlandy has joined #rdo12:41
*** lucas-hungry is now known as lucasagomes12:43
*** julim has quit IRC12:48
*** tosky has quit IRC12:53
*** tosky has joined #rdo12:55
*** puzzled has joined #rdo12:55
*** iranzo has joined #rdo12:59
*** iranzo has quit IRC12:59
*** iranzo has joined #rdo12:59
*** rhallisey has quit IRC13:00
*** _elmiko is now known as elmiko13:04
mnaseri wish running v3 was as easy as it seems to be :(13:04
*** akrivoka has quit IRC13:05
*** rhallisey has joined #rdo13:05
*** kgiusti has joined #rdo13:06
*** oshvartz has joined #rdo13:12
number80Jehane: that may be a good idea13:12
*** jcoufal has joined #rdo13:12
*** jmelvin has joined #rdo13:12
*** paragan has quit IRC13:16
*** jpena|lunch is now known as jpena13:16
dmsimardhello #rdo13:16
dmsimardadarazs: need me ?13:16
dmsimardapevec: pong13:16
*** ccamacho|lunch is now known as ccamacho13:17
*** ihrachys has joined #rdo13:18
*** akrivoka has joined #rdo13:19
adarazsdmsimard: hi. not working on the problems we mentioned you yet, but maybe.13:19
*** gkadam has quit IRC13:19
adarazsdmsimard: I saw yesterday you were looking for me.13:20
adarazsis that still relevant?13:20
dmsimardjpena: centos7-new ? That's the newton-uc worker ?13:20
dmsimardadarazs: more or less, I was looking for general recommendations re: amount of workers/processes/threads13:20
jpenadmsimard: that's a temp name while we check everything is ok13:20
jpenaonce CI passes, we'll switch symlinks and make it centos7-master13:20
jpenaand current master will be centos7-master-head13:21
adarazsdmsimard: there's not really a good recommendation. it really depends on the machine you're using.13:21
dmsimardjpena: ok, I was really afraid that name was going to stick13:21
dmsimardjpena: I know naming things is hard but ... :)13:21
*** jdob has quit IRC13:22
*** jdob has joined #rdo13:23
*** hynekm has quit IRC13:23
*** pgadiya has quit IRC13:25
*** fragatina has joined #rdo13:28
*** ihrachys has quit IRC13:29
*** MRrobotPR has joined #rdo13:31
*** rdas has quit IRC13:31
*** belmoreira has quit IRC13:31
*** paragan has joined #rdo13:32
*** amiii has quit IRC13:32
apevecdmsimard, it's specifically ugly temp name so it doesn't stick :)13:35
*** jdob has quit IRC13:36
*** jdob has joined #rdo13:37
*** satya4ever has quit IRC13:37
*** richm has joined #rdo13:43
*** anilvenkata has quit IRC13:43
dmsimardapevec: if you need to test those repos we can use the weirdo generic jobs13:45
apevecbut we need full pipeline i.e. oooq too13:46
*** dgurtner_ has quit IRC13:47
apevecEmilienM, yay, in the gate!13:49
EmilienMapevec: yeah, we'll promote in the next hours13:49
*** hynekm has joined #rdo13:49
*** nehar has quit IRC13:50
apevecgfidente, social ^ so we could also have clean rebase later today13:50
*** fragatina has quit IRC13:51
*** eharney has joined #rdo13:54
*** akshai has joined #rdo13:54
*** Son_Goku has quit IRC13:55
gfidenteapevec, EmilienM thanks13:55
*** fragatina has joined #rdo14:00
*** akshai has quit IRC14:00
dmsimardweshay: the ci.centos team fixed the rsyncd ACL so that the new cloud slaves can also upload logs and we got a pass:
dmsimardweshay: may I add the new slaves to the rdo label ?14:01
dmsimardI am confident there will be no problem for the weirdo jobs.14:02
*** akshai has joined #rdo14:02
*** eliska is now known as eliska-mtg14:03
*** dgurtner has joined #rdo14:03
*** dgurtner has joined #rdo14:03
*** dgurtner has quit IRC14:03
*** dgurtner has joined #rdo14:03
*** dgurtner has joined #rdo14:03
*** eharney_ has joined #rdo14:04
dmsimardweshay: there are three new performant slaves each with 8 threads each, I'll reduce the "primary" slave to 8 threads as well14:04
*** eharney has quit IRC14:04
*** ppowell has quit IRC14:05
*** dgurtner has quit IRC14:05
weshaydmsimard, question.. are we certain all three have the same content, keys etc..14:06
dmsimardweshay: they were setup with ansible so yes14:06
weshaydmsimard, cool! I'm nervous.. but cool14:06
dmsimardweshay: hey it's for the better .. the primary slave keeps crashing so it can't be worse14:07
weshaydmsimard, k.. thanks man14:07
dmsimardI'm running a last batch of jobs to make sure and I'll put them in the pool after14:07
*** lkuchlan has joined #rdo14:07
*** ayoung has joined #rdo14:12
*** lkuchlan has quit IRC14:13
*** rdobot has joined #rdo14:19
*** ChanServ sets mode: +v rdobot14:19
*** jeckersb is now known as jeckersb_gone14:20
*** mlammon1 has joined #rdo14:21
*** ppowell has joined #rdo14:21
*** hynekm has quit IRC14:23
*** links has quit IRC14:23
*** mlammon has quit IRC14:24
*** jaosorior has quit IRC14:25
*** ekuris_ has quit IRC14:27
*** galiral has joined #rdo14:27
galiralhi guys14:27
galiralcan I ask something about packstack in here?14:28
*** gszasz has quit IRC14:28
jpenagaliral: sure, go ahead14:29
galirali'm trying to add a node to my lab (only 1 controller installed) using packstack14:30
*** Amita has joined #rdo14:30
galiralbut I get ArgumentError: Could not find declared class ::remote::db14:30
galiralwhen adding a storage node14:30
galiral(cinder + glance + swift)14:30
*** eliska-mtg has quit IRC14:30
galirali'm using a modified answer file14:30
galiralfrom the first run for the controller14:30
*** gszasz has joined #rdo14:31
*** ihrachys has joined #rdo14:31
jpenagaliral: are you running RDO mitaka?14:31
galiralsorry for the caps14:32
galirali'm on centos 7.214:32
jpenalet me do a quick check. Having a separate storage node used to work somehow, but it was officially deprecated. So it could be just a minor change, or something bigger14:33
galirali actually got the error even just adding a nova node (it complpained about not having a storage and automatically enabling it)14:34
*** rbowen changes topic to "RDO: An OpenStack Distribution for CentOS/Fedora/RHEL | Community meeting is held here every Wednesday at 15:00UTC | Please read our IRC community guidelines: | Logged at""14:36
jpenagaliral: in that case you were not adding a compute node, were you? Only adding compute nodes works fine, you can't have a second Nova API server14:36
rbowenThe RDO IRC meeting is in about 25 minutes. Please update the agenda at if you have anything that we need to discuss.14:36
dmsimardThat damn cirros image14:36
galiralsorry, yes a compute node14:36
galirali have the apis on the controller14:37
galiralalready installed14:37
jpenathat's strange14:37
jpenagaliral: can you paste your answer file?14:37
*** myoung is now known as myoung|mtg14:38
*** ohochman has left #rdo14:38
*** ohochman has joined #rdo14:39
*** ohochman has quit IRC14:39
*** ohochman has joined #rdo14:39
jpenaok, so your first run was on .200, I guess14:40
jpenathen you wanted to add .202 as a compute node?14:41
galirali want 201 to be the storage14:41
galiral202 compute14:41
galirali didn't manage to install neither the compute nor the storage14:41
jpenaif you only want to add .202 (let's fix things one at a time), we should have:14:41
jpena(to keep storage host at .200)14:41
jpenaso it doesn't try to touch your already working controller14:42
galiralwhat about the new service records14:42
jpenathen, you can run packstack --answer-file=xyz.txt on the same host you used the first time to run packstack (I assume .200)14:42
galiralin the sql?14:42
jpenaif you're just adding a compute node, the only new records will be used for the neutron and nova agents, and that's already done from the controller side14:43
*** chandankumar has quit IRC14:44
galiralok trying it now14:45
*** ppowell has quit IRC14:46
*** fragatina has quit IRC14:49
*** myoung|mtg is now known as myoung14:49
*** fragatina has joined #rdo14:49
galiralok it finished its run14:50
galirallets see14:50
*** mlammon1 has quit IRC14:50
*** eharney_ is now known as eharney14:51
*** mlammon has joined #rdo14:51
*** mlammon has quit IRC14:54
jristhey all - RDO meeting still despite Flock?14:54
rbowenWell, we have a meeting chair, and one agenda item, so I thing yes.14:55
rbowenthink, that is.14:55
rbowenI expect it'll be short.14:55
rbowenBut not all of us are at Flock.14:55
jristfair :)14:55
*** mlammon has joined #rdo14:56
jruzickanumber80, sorry for the drop, my internet is funky since the thunderstorm.14:56
*** Ryjedo has joined #rdo14:56
*** pradk_ has joined #rdo14:59
jruzicka#startmeeting RDO meeting - 2016-08-0315:00
*** ppowell has joined #rdo15:00
zodbotMeeting started Wed Aug  3 15:00:41 2016 UTC.  The chair is jruzicka. Information about MeetBot at
zodbotUseful Commands: #action #agreed #halp #info #idea #link #topic.15:00
zodbotThe meeting name has been set to 'rdo_meeting_-_2016-08-03'15:00
openstackMeeting started Wed Aug  3 15:00:41 2016 UTC and is due to finish in 60 minutes.  The chair is jruzicka. Information about MeetBot at
openstackUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.15:00
openstackThe meeting name has been set to 'rdo_meeting___2016_08_03'15:00
galiralhey jrist15:00
* jrist waves15:00
galiralusually on the controller15:00
galiralwhat service are installed15:01
*** mosulica has quit IRC15:01
jruzickadouble meetbot for double victory15:01
*** rbowen sets mode: -v openstack15:01
toskythe agenda item is me :)15:01
*** kaminohana has quit IRC15:01
jruzicka#chair tosky number80 imcsk8 social jpena rbowen15:02
zodbotCurrent chairs: imcsk8 jpena jruzicka number80 rbowen social tosky15:02
openstackCurrent chairs: imcsk8 jpena jruzicka number80 rbowen social tosky15:02
*** gszasz has quit IRC15:02
*** fragatina has quit IRC15:02
jruzicka#topic jar and exceptions for openstack-sahara-tests ( bug 1318765 in Package Review "Review Request: openstack-sahara-tests - Sahara Scenario Test Framework" [Unspecified,Assigned] - Assigned to apevec15:02
*** nmagnezi has quit IRC15:02
galirallet me get it right, it is referring to the nova apis or the compute node?15:02
jruzickagaliral, please wait until the meeting is over which is soon :)15:02
jruzicka#chair coolsvap dmsimard15:03
zodbotCurrent chairs: coolsvap dmsimard imcsk8 jpena jruzicka number80 rbowen social tosky15:03
openstackCurrent chairs: coolsvap dmsimard imcsk8 jpena jruzicka number80 rbowen social tosky15:03
jruzickahmm, bot doesn't comment on the topic... well go on anyway, tosky ;)15:03
toskyso, the topic was partially discussed in the previous meetings but I could not attend15:03
toskybasically: there are some jar (yeah, I know) which used to live in sahara (and some of them are still there) and now in sahara-tests (used for testing)15:04
toskyif I'm not mistaken we shipped it in the past, I know all the issues about shipping binaries and so on, so my point here is:15:04
*** anilvenkata has joined #rdo15:05
tosky- I would like to ask an exception to have the binaries shipped as they are (they are mostly examples) in sahara-tests15:05
tosky- I would work upstream for fixing this (the sources are mostly in sahara-extras) but most likely there is no time for N, more likely for O15:05
toskyand I would need guidance on that15:05
toskyhow to organize things in the way that they could be accepted15:06
*** paramite has quit IRC15:06
toskyso I'm asking for an exception so that openstack-sahara-tests could go still potentially in mitaka15:06
* social does nog like bundled jars15:06
jruzickanoone does ;)15:06
*** ade_b has joined #rdo15:06
number80I'm ok with it on the principle as long as it IS DOCUMENTED IN THE SPEC15:07
jpenaalso, if the source location is known, we could add at least a pointer to them somewhere in the spec15:07
number80apevec was doing the review, so I'd like to get his opinion before we vote15:07
number80jpena: good point15:07
toskyI don't like them either, but it's not so simple to untangle, also because... java15:07
number80true, one of the things I'd do is create a trello card to follow that with an actual deadline15:08
*** toanju has quit IRC15:08
toskywhat I'd need, as I said, are best example on how to deal with such cases (I guess we have other mixed packages in java)15:08
toskyor packages with java parts15:09
*** akshai has quit IRC15:09
number80tosky: well, it will depend on a lot of factors, if we get MEAD in CBS, it may get simpler15:09
number80(MEAD is not universal solution but it's one of them)15:09
number80I suggest since apevec is likely on a call, to do informal vote15:10
toskyso, on the upstream side, just tell me what I should provide so that, for any possible solution implemented in packaging (I will ask zigo too), things don't become too complicated15:10
*** dprince has quit IRC15:10
* apevec out of call, reads back15:11
number80proposal: grant sahara-tests bundling exceptions and track progress on unbundling jars15:11
*** ade_b has quit IRC15:11
number80at least, reviewer should have a say before final decision :)15:11
*** ade_b has joined #rdo15:11
apevecyes, that's what I wanted to do, collect all source then ask for exception, but didn't get to it yet15:12
apevecon upstream side, it would help to provide at least READM next to each binary jar15:12
apevecto document how was it built15:12
*** mlammon has quit IRC15:12
number80that should be an action, do you take it tosky?15:12
*** mlammon has joined #rdo15:12
toskynumber80: add a README? Yes, does it need to be really in the same directory or could it be with the global documentation?15:13
toskyapevec: ^^15:13
number80tosky: preferably same dir15:14
*** gszasz has joined #rdo15:14
apevectosky, either way, if upstream prefers to keep developer docs in one place, that's also fine15:14
number80#action tosky ask upstream to document how sahara-tests jars are built15:15
number80so let's formally vote the proposal ^15:15
toskyit's more "tosky send a review to document how sahara-tests jars are built"15:15
zodbotRemoving item from minutes: ACTION by number80 at 15:15:23 : tosky ask upstream to document how sahara-tests jars are built15:15
openstackRemoving item from minutes: <ircmeeting.items.Action object at 0x7f5bbfdcee50>15:15
number80#action tosky send a review to document how sahara-tests jars are built15:16
*** julim has joined #rdo15:16
number80tosky: I'm fine with you doing the work :)15:16
*** rpioso has joined #rdo15:16
apeveclet's info  the exact proposal15:17
apevec#info  proposal: grant sahara-tests bundling exceptions and track progress on unbundling jars15:17
apevec+1 from me15:17
number80+1 (for minutes)15:18
jruzickaonce there is an exception, there will be no pressure to fix but OK15:18
number80jruzicka: next action would be creating a tracking card in trello :)15:18
*** akshai has joined #rdo15:18
apevecjruzicka, exception comes with docs15:18
jruzickawith great docs comes great exception? :)15:18
number80no, that's the exception of that quote :)15:19
number80#agreed grant sahara-tests bundling exceptions and track progress on unbundling jars15:19
*** priteau has quit IRC15:19
number80#action number80 create follow-up card in trello15:19
jruzicka#topic open floor15:20
apeveccheck agenda :)15:20
jruzickaoh, new agenda :D15:20
jruzickanever too late :-p15:20
*** READ10 has joined #rdo15:20
apevecmeh, I can't undo15:20
zodbotRemoving item from minutes: <MeetBot.items.Topic object at 0x3cdb2310>15:20
openstackRemoving item from minutes: <ircmeeting.items.Topic object at 0x7f5bbfc09210>15:20
jruzicka#topic temp CI pipeline for
jruzicka#chair apevec15:21
zodbotCurrent chairs: apevec coolsvap dmsimard imcsk8 jpena jruzicka number80 rbowen social tosky15:21
openstackCurrent chairs: apevec coolsvap dmsimard imcsk8 jpena jruzicka number80 rbowen social tosky15:21
toskyjruzicka: the pressure is you coming to my cubicle15:21
apevecthat's me, I've been asking on IRC about that but no definitive answer yet15:21
jruzickatosky, I'm not that motivated to unbundle the jars :-p15:21
number80jruzicka: that should be a motivation to encourage tosky doing it ;)15:22
apevecflepied, jpena, weshay, ^ can we have one temp full CI pipeline run against that temp repo?15:22
apevecwe need that pass before switching centos7-master15:22
apevecdmsimard suggested generic weirdo jobs, which is fine but we also need oooq15:22
apevecat least minimal15:22
jpenaI'm not very familiar with creating CI pipelines, expected dmsimard to help 0:)15:22
apevecweshay, other option is to run internally on RHEL15:22
dmsimardI can look15:23
apevecjpena, that's why I'm asking, who can take this, it's critical before we do the switch15:23
apevecI mean, we could also just switch15:23
*** hewbrocca is now known as hewbrocca-afk15:23
apevecand related what would be good timing for the switch?15:23
apevecflepied, ^15:23
flepiedapevec: I'm in favor of switching15:23
*** chandankumar has joined #rdo15:24
jpenawhat would happen to the CIs that use hashed repos? They won't be available under the new url15:24
apevecwe should get pass on old centos7-master today15:24
*** shaunm has quit IRC15:24
*** tshefi has quit IRC15:24
apevecwe're already 8d behind15:24
apevecjpena, yes, that's the price of the progress15:24
*** ade_b has quit IRC15:24
apevecjpena, we have it archived on buildlogs15:24
flepiedjpena: who is using a hash outside of Puppet ?15:25
apevecjpena, but good point, we'd break puppet-ci and tripleo-ci15:25
weshayapevec, ah.. probably best to create a temp pipeline for it15:25
apevecweshay, that's what I proposed :)15:25
apevecbut not sure who can do it15:25
number80flepied: maybe kolla does15:25
apevecjpena, flepied - so we'll need to sync tripleo and p-o-i pins15:25
weshayapevec, I'll do it15:25
flepiedapevec: agreed15:26
weshayapevec, can you create a card in trello w/ the details15:26
jpenaapevec: ok, let's hope they're not the same as any existing hash in newton-uc15:26
apevecjpena, hmm, good point15:26
apevecdmsimard, ^ re. kola, do you know if they use passed-ci symlink  or exact hash like puppet ci?15:27
jpenalet me check15:27
apevecweshay, ok, I'll fork from
dmsimardapevec: I haven't checked in a while, let me look15:27
apevecjpena, poi has pin bump proposed15:27
dmsimardKolla uses the current-passed-ci cdn15:27
apevecok, then kolla is good15:28
apevecjpena, bump in poi is
jpenawe're lucky, neither the current or proposed hashes in are in use in newton-uc15:28
apevecpoi bump is blocked on
*** edannon has quit IRC15:29
apevecfailed in gate :(15:29
apevecok, actions15:29
*** ihrachys has quit IRC15:29
apevec#action weshay to create temp CI pipeline for
number80dmsimard: thanks15:30
*** ihrachys has joined #rdo15:30
apevec#action jpena to copy puppet-openstack-integration and tripleo-ci odl hashed repos to centos7-newton15:30
apevecanything else?15:30
apeveclet's sync tomorrow and then decide when to switch15:30
jpenaapevec: that's after the temp CI pipeline gives us the ok, right?15:31
apevechopefully we'll have another promotion with current centos-master by then15:31
apevecjpena, right15:31
apeveconce that passes, we switch centos7-master to centos-newton15:31
jpenabtw, for kolla we need to sync as well, they download
jpenawhich points to a hashed repo15:31
apevecah ok, then we need old current-passed too15:32
apevecwhich we need anyway until we get promotion with the new repo15:32
jpenawhere's the tripleo CI repo configured?15:32
jpenaI'd assume it uses current-tripleo, but just in case15:33
weshayjpena, apevec this may take more than a day to get a temp pipeline15:33
sdakeheard a beep on kolla - anything you need from us apevec?15:34
apevecsdake, not for now, we'll try to NOT break you :)15:34
apevecjpena, it is that symlink15:34
sdakeapevec cool - heads up - we will soon be mirroring rdo repos inside openstack infrastructure15:34
sdakei hae a patch in the works15:34
sdakeif you break something (links/etc) we can easily repair that15:35
*** Amita has quit IRC15:35
sdakebut if there are link changes would be nice to know ahead of time so i can get the mirroring done correctly15:35
apevecsdake, can you give us link to the review?15:36
*** mflobo has joined #rdo15:36
apevecwe might want to rethink delorean.repo b/c it links back to trunk.rdo15:37
sdakeapevec sitting on my ssd atm, will publish later today - i'll send a link to rdo mailer when done15:37
*** leanderthal is now known as leanderthal|afk15:37
apevecI'm good with the topic, anything else?15:37
sdakeya something that would break kolla is removing trunk builds all-together15:37
sdakecould you expand on what you have in mind related to rethinking delorean.repo?15:38
apevecI'd prefer to point external users to the mirror we have on buildlogs.centos.org15:38
jpenawe have our passed-ci stuff in the centos CDN, however the delorean.repo file points back to (non-CDN)15:38
apevec already redirects to
apevecbut delorean.repo inside still points to specific hashed repo on trunk.rdo15:39
apevecdefeating the purpose of the buildlogs CDN :)15:39
*** itlinux has joined #rdo15:39
apevecwhat jpena  said :)15:39
apevecalright, but that's off-topic15:39
apevecjust related15:39
dmsimardapevec: it's a highly available delorean.repo !15:39
apevecyeah, "great success"15:40
apevec(that's my 11yr old talk)15:40
number80next topic?15:40
jruzickaastounding success indeed15:40
jruzicka#topic Ideas to improve DLRN instance performance15:40
jruzickajpena, ^15:40
*** iranzo has quit IRC15:40
jpenaapevec mentioned today that there were some talks last week about how to improve the current worker's performance15:40
apevecif someone has doubts this isn't needed check
jpenaI started doing some brainstorming today, and most of the time I ended up redesigning koji15:41
apevecbuild time 15:33 for commit 11:0115:41
apevecstill 4h  behind after yesterday's 3h pause...15:41
apevecjpena, that's good, I heard number80 is doing that at flock in a hallways :)15:42
jpenaI had another idea, which is to run parallel builds using some python multiprocessing support. It might work, but I'll have to test it15:42
*** MRrobotPR has quit IRC15:42
number80only client-side :)15:42
apevecthat's where you start :)15:43
apevecjpena, so you think distributed builds on post-commit won't fly?15:43
apevecwith 3rd party setup against review.o.o15:44
sdakeapevec understood re cdn - that was a hue help to kolla's cis - however only have the repos are in the cdn15:44
jpenaapevec: I came up with a separation between api thread (receiving input from post-commit) and DB thread (downloading rpms from api builders, creating repos, etc)15:44
sdakethe deps repos are not15:44
*** zoliXXL is now known as zoli|gone15:45
dmsimardapevec: jpena wasn't in the meeting re: post commit. Did anyone fill him in ?15:45
jpenaand then we have to think how to authenticate, sync for multiple commits for the same project...15:45
jpenaadd retrigger logic15:45
sdakeapevec if you see harm in a oo mirror of rdo let me know15:45
sdakeand i'll consider rethinking my patch15:46
apevecsdake, I don't but wanted to double-check what exactly would be mirrored15:46
sdakedelorean and delorean-deps15:46
apevecdmsimard, I didn't keep notes15:46
sdakeas well as the stable branches - down the road15:46
number80dmsimard: that's different efforts fixing the same problem, we're //-izing solution completions15:47
sdakeso basically mirroring everything for everyone in OO to use15:47
number80we still need to improve DLRN the time being15:47
sdakeshould be goodness for RDO but I could be wrong15:47
apevecjpena, yeah, we'd have to serialize per project15:47
sdakei'd like to make a consumer of rdo in addition to kolla15:48
number80jpena: plan to //-ize mock builds?15:48
apevecnumber80, yeah, we're single threaded per DLRN instance15:48
jpenanumber80: yes, I think we could do it. We have a beefy server (8 cores), so we could run 3-4 mocks in parallel for the current master, which takes most of the work15:48
number80(it'd just mean hacking mock profiles to use different path for chrooted root)15:48
apevecalso there's 2min delay after each build15:49
jpenanumber80: exactly that's what I was thinking15:49
number80good plan overall15:49
number80IMHO, authentication and retriggering is lower prio15:49
number80but good15:49
apevecok, let's try multithreading on a single machine first15:49
apevecb/c machine itself is not very loaded, afaict15:49
number80with a beefy server, I think we can make it reasonably working15:50
apevecjpena, what would be the action, do you want to post proposal on rdo-list?15:50
*** zoli|gone is now known as zoli_gone-proxy15:50
number80we don't have C extensions of stuff that are CPU-intensive15:50
jpenaapevec: if there are no other ideas, action would be for me to try a patch15:51
apevecnumber80, most intensive is sphinx build15:51
*** ihrachys has quit IRC15:51
jruzicka#action jpena to try parallel mock builds15:51
*** dyasny_ has quit IRC15:51
apevecyep, one patch says more than thousands of specs15:51
jpenaI even thought of making build plugin-aware, so we could offload to koji. But that's another story :)15:52
apevecand then you'd DoS Koji instance :)15:52
*** jlibosva has quit IRC15:52
jpenamwahaha :D15:52
jruzickahaha, that's fun.15:53
apeveclet's keep CBS Koji for stable branch builds15:53
jruzickawe should wrap up15:53
number80but jpena++15:53
jruzicka#topic chair for next meeting15:54
*** aufi has quit IRC15:54
number80I can take it15:54
jruzicka#info number80 to chair next meeting15:55
jruzicka#topic open floor15:55
chandankumarI have written a blog on how to use fedora-review for rdo packages
chandankumarfeel free to try and have your comments there15:56
rbowenVote for OpenStack Summit presentations open for a few more days:
dmsimardrbowen: oh thanks for that, reminds me I need to advertise my presentations a bit15:56
rbowenWell, there aren't direct links to presentations any more.15:56
dmsimardYeah, I know that :)15:56
rbowenDue to the megathread on the community mailing list a couple of months back.15:56
* number80 thinks that is just tradition that serves no purpose15:56
dmsimardI'm okay with that, there's definitely less spam in my various feeds15:57
toskyyeah, there is a selection anyway later15:57
rbowenYes, that seemed to be the consensus of the thread.15:57
apevecpartisan voting? never!15:57
* number80 haven't even looked15:57
rbowenBut, I thought I'd mention it anyway. :-)15:57
*** danpawlik has quit IRC15:57
*** galiral has quit IRC15:57
zodbotMeeting ended Wed Aug  3 15:58:01 2016 UTC. Information about MeetBot at .15:58
zodbotMinutes (text):
openstackMeeting ended Wed Aug  3 15:58:01 2016 UTC.  Information about MeetBot at . (v 0.1.4)15:58
openstackMinutes (text):
toskyand that was supposed to be a short meeting15:58
jruzickaIt wasn't so short after all :)15:58
number80thanks jruzicka  :)15:58
toskynever say that at the beginning15:58
jruzickahaha, yeah15:58
number80gotta go, wifi is disconnecting me so it wants me to go outside15:58
*** dmsimard sets mode: +v openstack15:58
*** akshai has quit IRC15:59
jruzickadmsimard, completely offtopic, I wanted to talk with a database myself in errbot instead of the built in storage and that wasn't a nice experience at all :)15:59
dmsimardjruzicka: I don't follow16:00
dmsimardjruzicka: you used an "errbot" database ?16:00
dmsimardlike that plugin config framework ?16:00
jruzickadmsimard, sorry, too compressed... I wanted to store karma ina sqlite/postgresql DB from my errbot karmabot16:01
jruzickaSQLite objects created in a thread can only be used in that same thread.16:01
dmsimardjruzicka: eh, what if you just open a session and close that session right away ?16:02
jruzickaand similar funs, still haven't found a way to do a extremely simple thing using errbot16:02
dmsimardi.e, message received, message contains ++, open session, increment karma, close session16:02
jruzickadmsimard, yeah, I wasn't ready for such a hack... maybe neccessary16:02
dmsimardjruzicka: I wouldn't really have expected otherwise tbh :p16:03
jruzickadmsimard, oh yeah, you'd just do it like that?16:03
EmilienMapevec: hey, I'm trying to upgrade tripleo in CI and I use depends-on tripleoclient patch. it should build the package in tripleo CI but delorean fails,
jruzickawell thanks :D16:03
EmilienMapevec: have you seen that before?16:03
EmilienMapevec: another log
dmsimardjruzicka: /me shrugs16:03
*** dyasny_ has joined #rdo16:04
jruzickadmsimard, I was like "there must be a nice way to do it I'm just blind" but second opinion helped ;)16:04
apevecEmilienM, not familiar with tripleo-ci setup, weshay or sshnaidm could  help, but that's obviously missing repo16:04
dmsimardjruzicka: well, there might be, I haven't had to store data from a plugin before16:05
apevecsshnaidm, adarazs - was DLRN pin removed in upstream tripleoci script?16:05
jruzickadmsimard, but you wouldn't consider it disgusting so that's that ;)16:05
dmsimardjruzicka: but it sounds pretty easy to just do a sqlalchemy model and use sessions16:05
EmilienMapevec: ok thanks16:05
jruzickadmsimard, yup, that was the plan.16:06
weshayEmilienM, /me looks.. adarazs does the gate role work w/ depends-on yet?16:06
sshnaidmapevec, EmilienM not yet, waiting for merge:
apevecEmilienM, there were few issues resolved in
*** jistr is now known as jistr|biab16:06
EmilienMok good to know thanks16:06
EmilienMsshnaidm: +2 and recheck16:06
EmilienMI'll land it if green later16:07
EmilienMbut I'm not sure it's all related16:07
weshayadarazs, disregard16:07
*** Ryjedo has quit IRC16:08
*** Ryjedo has joined #rdo16:10
*** dyasny_ has quit IRC16:11
*** ppowell has quit IRC16:11
dmsimardweshay: does work for you ? Interface looks down to me.16:12
dmsimardnevermind, just came back16:12
*** saneax is now known as saneax_AFK16:12
weshaydmsimard, oooq ha has been removed from the master promote and replaced w/ min + pacemaker16:13
apevecweshay, adarazs - thank you!16:14
apevecI mean, all slow cpu in ci.centos thank you ;)16:14
*** Ryjedo has quit IRC16:15
apevecEmilienM, are we that unlucky? failed unrelatedly in the  gate :(16:15
EmilienMapevec: yeah16:15
EmilienMapevec: welcome to my life16:15
apevecerrr wront paste16:15
*** dtantsur is now known as dtantsur|afk16:15
apevecthis puppet-tempest fix16:15
apevecyeah, receck allmighty16:16
dmsimardweshay: ok16:16
dmsimardweshay: fyi new slaves in prod:
*** Goneri has quit IRC16:18
*** chandankumar has left #rdo16:18
*** chandankumar has joined #rdo16:19
chandankumarapevec, number80 jruzicka jpena Good night :-)16:19
*** chandankumar has quit IRC16:20
jruzickachandankumar, gn, sorry for the drop ;)16:20
*** anshul has quit IRC16:20
dmsimardweshay: the pacemaker job failed16:20
dmsimard16:20:04 ERROR! the file_name '/home/centos/workspace/tripleo-quickstart-periodic-liberty-delorean-roles-deploy-minimal_pacemaker/config/general_config/minimal_pacemaker.yml' does not exist, or is not readable16:20
weshayhrm.. did that merge?16:21
* dmsimard shrugs16:21
*** dyasny_ has joined #rdo16:21
weshaydmsimard, has workflow +1 but hasn't merged16:21
*** hrw has quit IRC16:22
weshaypanda, can you help us out here ^16:22
*** jistr|biab is now known as jistr16:22
* weshay checks zuuuuuuul16:22
imcsk8jpena: regarding if i do this in packstack/puppet/modules/packstack/manifests/provision/bridge.pp i get an error because br-ex is not being created :S16:23
*** hrw has joined #rdo16:23
*** devvesa has quit IRC16:23
*** Ryjedo has joined #rdo16:23
jpenaimcsk8: then let's just skip if CONFIG_NEUTRON_L3_EXT_BRIDGE == ''16:25
imcsk8jpena: i was thinking on that, but i wasn't shure if we needed to open stuff, the neutron puppet module is not clear on how to handle provider networks16:26
*** ccamacho is now known as ccamacho|out16:26
*** fultonj has quit IRC16:26
*** nstrug has quit IRC16:27
*** Ryjedo has joined #rdo16:27
*** nstrug has joined #rdo16:28
*** fultonj has joined #rdo16:28
*** ohamada has quit IRC16:29
jpenaimcsk8: let's test it and see. I think there should be no harm16:29
jpenaI gotta go now, dentist appointment16:29
imcsk8jpena: ok16:29
*** akshai has joined #rdo16:30
*** jpena is now known as jpena|off16:30
pandaweshay: looking16:31
*** oshvartz has quit IRC16:31
*** akshai_ has joined #rdo16:32
weshaypanda, ha.. just merged..16:32
weshaypanda, all you have to do is look at it.. impressive16:32
*** dprince has joined #rdo16:33
*** snecklifter has quit IRC16:33
*** Ryjedo has quit IRC16:34
*** akshai has quit IRC16:35
openstackgerritIvan Chavero proposed openstack/packstack: Fix CONFIG_NEUTRON_L3_EXT_BRIDGE invalid parameter
*** ushkalim has quit IRC16:36
*** rcernin has quit IRC16:38
*** RoozbehShafiee has joined #rdo16:40
dmsimardimcsk8: what do you mean about provider networks ?16:41
dmsimardI use provider networks with flat networking with puppet-neutron with no problem16:41
dmsimardnot with Packstack, though16:41
*** vimal has joined #rdo16:42
*** garrett has quit IRC16:42
dmsimardayoung: you need to listen to this:
dmsimardweshay: is this also the same error ? 16:27:35 ERROR! the file_name '/home/centos/workspace/tripleo-quickstart-gate-mitaka-delorean-quick-ooo-usbkey/config/general_config/centosci-logs.yml' does not exist, or is not readable16:43
*** nstrug_ has joined #rdo16:43
dmsimardit errored out on a couple jobs there16:44
dmsimardnot just that one16:44
weshaydmsimard, it just merged16:45
dmsimardthe centosci-logs one too ?16:45
weshayoh.. /me checks16:45
*** lucasagomes is now known as lucas-dinner16:45
weshaydmsimard, crap  that config file was removed on purpose and it looks like someone forgot to update the jjb16:46
*** devvesa has joined #rdo16:46
weshaydmsimard, I'll dive in .. sigh16:46
*** devvesa has quit IRC16:47
dmsimardcareful not to hit your head on the dive16:47
*** nstrug has quit IRC16:47
weshaydmsimard, hrm.. that config is there.. that seems to be another example of pip install -r being wonky16:49
dmsimardWhere is that file supposed to be ?16:49
*** furlongm_ has quit IRC16:49
weshaydmsimard, it's in the art-collect-logs/config dir16:49
*** furlongm_ has joined #rdo16:50
*** Ryjedo has joined #rdo16:50
ayoungdmsimard, listening16:50
dmsimardwell it's in config/centosci-logs.yml, not general_config/config/centosci-logs.yml ?16:50
dmsimardah nm,
weshaydmsimard, actually.. master is fine.. caused by
weshaydmsimard, pabelanger is experimenting w/ using zuul vs. python setuptools16:52
*** oshvartz has joined #rdo16:52
dmsimardweshay: /me nods this is what I do for weirdo16:52
dmsimardIt's not unlike what pabelanger does for windmill, I think16:53
*** Ryjedo has quit IRC16:55
pabelangerweshay: dmsimard: Ya, I am working on testing some things now16:55
*** Ryjedo has joined #rdo16:58
*** nstrug__ has joined #rdo16:58
pabelangerdmsimard: weshay: in fact, I don't think should be part of that repo.  I believe it should exist in tripleo-quickstart16:58
pabelangerthen we can stop installing config files from roles16:58
weshaypabelanger, collecting logs is not part of a users workflow outside of ci16:59
pabelangerfor example, we could move that into vars like:
pabelangerweshay: right, but ci-tools is using it16:59
pabelangerin quickstart16:59
*** pcaruana has quit IRC17:00
weshaypabelanger, one can make the same argument for ci-scripts17:00
pabelangerRight, I would be happy to see that get moved out17:01
pabelangerand into JJB17:01
weshaypretty sure tripleo cores wanted it removed17:01
*** nstrug_ has quit IRC17:01
weshayit's a tool/utility17:01
*** pnavarro has quit IRC17:01
weshaynot specific to jjb either17:01
*** null_ref has joined #rdo17:02
weshayso I'm kind of splitting hairs...17:02
weshaybut.. it's important17:02
*** degorenko_ has joined #rdo17:02
pabelangerci-scipts/ should exist in a repo that uses17:02
weshayalthough collect-logs is not typically used by a user outside of ci17:02
weshayit could be.. very helpful when someone needs to write a bug17:02
*** iberezovskiy_ has joined #rdo17:03
weshayrun collect-logs... dump the tar to the bug17:03
*** jbadiapa has quit IRC17:03
weshaywhich is why I classify collect-logs as just a tool for quickstart, but not part of it17:03
*** tesseract- has quit IRC17:03
pabelangerokay, so looping back to:
weshaypabelanger, kind of getting back to quickstart-extras17:04
*** nyechiel has quit IRC17:04
weshaypabelanger, that's the default vars..17:04
weshayer.. not default17:04
weshaybut ya17:04
pabelangermy view, we should not ship default configuration for projects consuming the role, inside the ansible-role17:04
*** mlammon is now known as mlammon|afk17:05
pabelangerI don't think we would merge pabelangerci-logs.yml ansible-role-tripleo-collect-logs which had my configuration17:05
weshaypabelanger, ok.. makes it much harder to test/develop but what's the alternative?17:05
pabelangerwe'd expect that to be setup in the playbook using the role17:05
pabelangerI think it is possible to either setup sane defaults that work with ci.centos.org17:06
weshaypabelanger, right.. we would not want to merge.. pabel-logs.yml17:06
pabelangerand / or override the settings from the vars file in the playbook17:06
*** openstackgerrit_ has joined #rdo17:07
*** nmagnezi has joined #rdo17:07
pabelangertoday, we have ansible-role-tripleo-collect-logs being installed into tripleo-quickstart, right or wrong, that what we can work with today.17:07
*** iberezovskiy_ has quit IRC17:08
*** degorenko_ has quit IRC17:08
*** openstackgerrit_ has quit IRC17:08
pabelangeradding a vars file into quickstart to setup ansible-role-tripleo-coolect-logs for the ci host would be possible, which only gets run when ci-scripts/ rung17:08
pabelangershouldn't be hard to do an example of that17:08
weshaypabelanger, would you consider a tripleo-quickstart-ci-centos.repo that has all the configs required for ci.centos17:08
weshayand likewise for internal, and upstream17:08
*** ppowell has joined #rdo17:09
weshaypabelanger, so site specific config per site17:09
*** akshai_ has quit IRC17:11
*** Ryjedo has quit IRC17:12
*** Alex_Stef has quit IRC17:13
pabelangerweshay: Ya, maybe we need to do that.17:15
*** jhershbe has quit IRC17:15
weshaycool.. I'm down w/ that17:15
pabelangerI don't have a good answer right now, but I think we can flesh out something in the coming days17:15
*** eharney has quit IRC17:20
*** apevec has quit IRC17:20
*** Ryjedo has joined #rdo17:22
*** dgurtner has joined #rdo17:24
*** dgurtner has joined #rdo17:24
*** nstrug_ has joined #rdo17:24
*** vimal has quit IRC17:26
*** nstrug has joined #rdo17:26
*** Ryjedo has quit IRC17:27
*** nstrug__ has quit IRC17:28
*** nstrug_ has quit IRC17:29
*** apevec has joined #rdo17:32
*** apevec has quit IRC17:32
*** apevec has joined #rdo17:32
*** fzdarsky is now known as fzdarsky|afk17:32
*** abehl has joined #rdo17:33
*** abehl has quit IRC17:34
dmsimardEmilienM: that vpnaas p-o-i failure is that tempest patch I saw you and apevec talk about ?17:39
larsksHey dlrn peeps, 'yum upgrade' on my undercloud, which points at edd36d00776cca0989e815223bbf0b21f42f6048, is failing with a variety of dependency problems:
apeveclarsks, upgrade from where to where?17:40
*** vimal has joined #rdo17:41
dmsimardapevec: hm, still not merged I guess
apevecyeah, failed in gate :(17:42
*** lkuchlan has joined #rdo17:42
larsksI'm running a quickstart with the devmode I'm not exactly sure how to answer your question, but e.g. I have openstack-swift-2.7.1-0.20160613173619.d75e6d3.el7.centos.noarch and an upgrade of that component fails like:
*** dgurtner has quit IRC17:43
apevecok, lemme try that timestamp17:43
apevecI thought we tested upgrade scenario but maybe missed some case17:44
*** shardy has quit IRC17:44
apevecin general, trunk 2 trunk is not guaranteed to be yum upgradeable17:44
apevecstable 2 trunk should work17:44
larsksI am not even intentionally trying to upgrade anything; part of 'openstack undercloud install' seems to run a 'yum upgrade'.17:45
larsksweshay: (just so this is all in one place) the output of 'yum repolist':
apevecah timestamp was purged in trunk.rdo17:47
*** nehar has joined #rdo17:47
larsksapevec: which was purged?17:48
*** vimal has quit IRC17:51
*** dyasny has joined #rdo17:51
*** dyasny_ has quit IRC17:53
*** milan has quit IRC17:53
*** sarath has joined #rdo17:53
*** julim has quit IRC17:53
*** nmagnezi has quit IRC17:54
*** julim has joined #rdo17:56
apevecold build you have 20160613173619.d75e6d317:56
*** RoozbehShafiee has quit IRC17:57
tdasilvahello, where can i find openstack packages for fedora 24?17:58
larsksHuh.  So, that suggests the image is stale? cc: myoung weshay17:58
*** ohochman has left #rdo17:59
apevecit is old, lemme see if can find older build internally17:59
*** nehar has quit IRC18:00
apevecin RDO Trunk we had to aggressively purge as we were near filesystem capacity18:00
weshaylarsks, it's a week old.. you could pull testing if you want18:00
apeveconly keeping symlinked repos i.e. current-passed18:00
apevecand tripleo18:00
*** alexismonville has quit IRC18:01
larsksweshay: I thought this was fetched freshly this morning...18:01
weshaylarsks, what I just pointed at was an unpromoted image18:01
larsksAh, got it.18:01
*** gszasz has quit IRC18:02
apevecso which image has  20160613173619.d75e6d3 swift build?18:02
larsksapevec: whichever one has md5sum == e572b882433becac566f78650de00c9b18:02
larsksI am not even convinced I have using the image I thought I was.18:03
larsksI may have gooded my command line.  Let me retry and see what happens.18:03
larsksThe opposite of gooded.18:04
*** degorenko is now known as _degorenko|afk18:07
rdogerritrdo-trunk created openstack/nova-distgit: openstack-nova: failed to build e740748
weshaydmsimard, oh forgot about this.. re multiple slaves18:09
*** tosky has quit IRC18:09
weshayit's going to mess up the promotion pipeline18:09
rbowenAs a number of people have pointed out, s/mitaka/newton/ in the RDO newsletter. At least it's gratifying that people are reading it.18:09
weshaydmsimard, the property files for the build number are stored on the slave in /tmp18:09
weshayonce we use multiple slaves.. we're going to get wonky results18:10
weshaywe should pin the promote jobs to one slave18:10
*** akrivoka has quit IRC18:10
weshayapevec, ^ fyi18:10
apevecah, why are we passing properties via files in /tmp ?18:10
*** jdob has quit IRC18:11
apeveccouldn't job pass parameters via env or something?18:11
*** jdob has joined #rdo18:11
apevecbut pinning to one slave is fine as a workaround18:11
apevecalthought that creates spof18:12
weshayapevec, it's so one job can pass vars to another18:13
weshayapevec, there are not a lot of ways to do that18:13
dmsimardweshay: hmmmm... you're right18:13
dmsimardweshay: I'll add a label: rdo-promote18:14
*** Ryjedo has joined #rdo18:15
weshayapevec, the issue is that delorean can update at any moment.. so to ensure that all the jobs start w/ the same hash.. it must be stored18:15
*** mlammon|afk is now known as mlammon18:15
weshayapevec, one job can start 3hrs after the other depending on the slave queue18:15
rdobot[sensu] NEW: - check-delorean-newton-current @ |#| Build failure on centos7-master/current: nova:
apevecyeah, I know but I thought this is passed as a variable to the child jobs18:17
apevecobviously, I was mistaken18:18
weshayapevec, no.. you're right18:18
weshaybut done via an ini file on the filesystem18:18
weshaydmsimard, apevec the way we do it downstream.. is .. we store the var on master18:19
weshaybut we don't have access to the master in ci.centos18:19
*** mcornea has quit IRC18:19
dmsimardweshay: I removed the "rdo" label from the original slave and set it to "rdo-promote" and pushed
weshaydmsimard, unless you can convince KB to let us do that18:20
dmsimardweshay: this is a stopgap, though, I'm not happy about this, it defeats the whole point of having different slaves for redundancy and capacity18:20
weshaydmsimard, so the parent job would run from master18:20
dmsimardI'll look for an alternative18:20
weshayand the child jobs run on slaves18:20
*** jdob has quit IRC18:20
*** fzdarsky|afk is now known as fzdarsky18:20
*** jdob has joined #rdo18:20
dmsimardweshay: back when I experimented with pipelines I don't believe I needed a file, I started parameterized builds downstream18:21
dmsimardI'll check18:21
weshaydmsimard, master has two executors.. in ci.centos18:22
apevecquick googling finds
dmsimardapevec: yeah that's what I used18:22
weshaythat's what we're using18:22
dmsimardweshay: you sure ? It's more multi job phases iirc18:22
*** jpich has quit IRC18:22
weshayoh ya.. my mistake18:23
*** Ryjedo has quit IRC18:23
apevecdmsimard, does JJB support that plugin and do we have it in ci.centos?18:24
dmsimardweshay: anyway, pin to a single slave is pushed18:24
apeveclooks like promotion could be refactored using that?18:24
dmsimardapevec: yes to both18:24
dmsimardapevec: except it's probably mutually exclusive with the multijob plugin18:25
dmsimardapevec: and that's why I used the pipeline plugin instead18:25
apevecwhat does multijob bring us?18:25
dmsimardthe phases and the view18:25
dmsimardapevec: the build pipeline basically gives you this view
dmsimardof upstream and downstream jobs triggered through parameterized builds18:26
dmsimardi.e, the promote get hash would be on the far left, then build image, then all of the jobs18:27
*** mflobo has quit IRC18:27
dmsimardpromote get hash would trigger a parameterized build of image build (with hash as parameter or whatever), and so on18:27
weshayrlandy, my virtjob just failed on validate18:27
*** rwsu has quit IRC18:27
weshayrlandy, so.. hopefully I'll have time later to poke at that too18:27
*** oshvartz has quit IRC18:28
rlandyweshay: what was the error?18:28
rlandyheat stack-show pingtest_stack18:28
rlandyweshay: ^^18:28
rlandyanything about external network being reachable?18:28
rlandyI have added some polling and sleeps but sometimes it works and sometimes not18:29
rlandywhich means I don't have the right condition checked18:29
*** jaosorior has joined #rdo18:29
*** jaosorior has quit IRC18:30
weshayrlandy, bah diff issue.. dont ask.. unrelated to validate. /me shakes his head18:30
*** iberezovskiy has quit IRC18:32
rlandyweshay: anyways I modified the commit  - maybe it will help. idk to be honest18:36
rlandybut we do recognize the failure now18:36
*** ayoung has quit IRC18:38
*** ccamacho|out is now known as ccamacho18:39
*** anilvenkata has quit IRC18:41
*** dustins has quit IRC18:41
*** oshvartz has joined #rdo18:42
*** dustins has joined #rdo18:45
*** julim has quit IRC18:46
apeveclarsks, that must be some mixup w/ yum priorities18:49
larsksapevec: I think maybe just an old image due to my fingers goofing the command line, actually.18:50
apevecdo you see anything like "ignore due priorities" earlier?18:50
larsksJust a weird failure mode.18:50
apeveclarsks, right, but latest openstack-swift should still update any old build cleanly, unless prevented by yum prio18:50
larsksAh, I see.  Tragically, that environment is all gone, and the new deploy seems to be successful.18:51
apevecok, then ignore mode on :)18:51
larsksThanks for looking into things!18:51
apeveciirc (delorean-tripleo) is added with higher priority18:51
apevechm, but from you gist it should be recent build, Aug 218:53
*** dyasny has quit IRC18:54
*** cwolferh has quit IRC18:56
*** eharney has joined #rdo18:58
*** paramite has joined #rdo18:59
apevecI couldn't find exact 20160613173619 build but with close enough 2.7.1-0.20160608225429.11c5ef7 upgrade to current-tripleo works:19:01
apevec python-swift                 noarch 2.9.1-0.20160802144313.d819ae0.el7.centos19:01
apevec                                                          rdo-trunk-newton-tripleo 1.0 M19:01
apevec     replacing  openstack-swift.noarch 2.7.1-0.20160608225429.11c5ef7.el7.centos19:01
*** READ10 has quit IRC19:01
*** cwolferh has joined #rdo19:02
*** ppowell has quit IRC19:05
dmsimardweshay: looking at the doc for
dmsimardweshay: the plugin exposes env vars such as LAST_TRIGGERED_JOB_NAME or TRIGGERED_BUILD_NUMBER_<project name>19:05
dmsimardweshay: so we could probably store the properties file as an artifact of the job (like we do for the logs.html file), not necessarily upload it to the artifacts rsync server19:06
dmsimardand reconstruct the URL with the env vars to get access to it19:06
*** READ10 has joined #rdo19:07
weshaynot sure if env vars will survive across jobs19:07
dmsimardthat's not what I'm saying19:07
*** dyasny has joined #rdo19:07
dmsimardhang on19:07
dmsimardweshay: so, see this job:
weshaydmsimard, we could send the property up to the log file..19:08
dmsimardit saves nothing19:08
weshaypull it down19:08
weshayand read it in19:08
*** fultonj has quit IRC19:08
dmsimardweshay: but this job saves a logs.html file:
dmsimardweshay: the link is easily craftable
dmsimardit looks like we have access to the "parent" job name and number19:09
weshayya.. that we should :)19:09
weshayya.. if we can get an artifact up to a log.. we could have each job read it in .. but that would be done in the shell19:11
weshayit would not be a job property.. but would still work19:11
dmsimardso we could (pseudocode) for line in $(curl -s; do key=$(echo $line |cut -f1 -d =); value=$(echo $line |cut -f2 -d =); set -gx $key $value; done19:11
dmsimardoh, don't even need all that19:12
dmsimardwe could just craft the properties.env file in such a way that we can source it19:12
weshaydmsimard, ya19:13
weshayjust needs to be in the env of the shell where the ci-script lives and we're good19:13
*** READ10 has quit IRC19:21
*** maeca1 has joined #rdo19:28
*** spr2 has quit IRC19:28
*** spr1 has joined #rdo19:28
*** lkuchlan has quit IRC19:28
*** livelace has quit IRC19:30
*** tesseract- has joined #rdo19:31
*** aortega has quit IRC19:34
*** jcoufal has quit IRC19:34
*** tesseract- has quit IRC19:37
*** dsneddon has quit IRC19:39
*** dsneddon has joined #rdo19:39
*** rwsu has joined #rdo19:40
EmilienMwe're so unlucky19:40
EmilienMrecheck again19:40
*** ohochman has joined #rdo19:40
*** ihrachys has joined #rdo19:41
*** ayoung has joined #rdo19:45
dmsimardEmilienM: ugh19:47
weshaydmsimard, we should do more reviews19:51
*** Goneri has joined #rdo19:51
weshaythat is for the parent job only19:51
dmsimardweshay: hmm19:51
dmsimardso I guess we need to override the default node for every quickstart promote job templates ?19:53
dmsimardweshay: this one and this one
*** shaunm has joined #rdo19:56
weshaydmsimard, just the default for those jobs.. one place19:56
weshaydmsimard, sec.. talking to morazi19:56
weshaydmsimard, I'll update it19:57
*** dprince has quit IRC19:58
*** maeca2 has joined #rdo20:01
*** permalac has quit IRC20:02
*** permalac has joined #rdo20:03
*** maeca1 has quit IRC20:04
dmsimardweshay: lgtm20:05
*** Goneri has quit IRC20:10
*** coolsvap has quit IRC20:11
rdobot[sensu] NEW: - check-delorean-newton-current @ |#| Build failure on centos7-master/current: nova:
EmilienMdmsimard: when we get our stuff merged, we have a nice promotion coming20:16
EmilienMon trunk, with no blocker on our side.20:16
*** ihrachys has quit IRC20:20
*** dgurtner has joined #rdo20:22
*** dgurtner has joined #rdo20:22
*** puzzled has quit IRC20:23
*** sarath has quit IRC20:23
*** ohochman has quit IRC20:27
*** dustins has quit IRC20:30
*** jayg is now known as jayg|g0n320:38
*** paramite has quit IRC20:38
*** danpawlik has joined #rdo20:42
*** dustins has joined #rdo20:44
*** danpawlik has left #rdo20:45
*** raissa has quit IRC20:54
*** raissa has joined #rdo20:54
*** jmelvin has quit IRC20:56
*** furlongm has joined #rdo20:57
*** furlongm_ has quit IRC20:57
*** READ10 has joined #rdo21:06
*** dustins has quit IRC21:07
*** fzdarsky has quit IRC21:10
*** v1k0d3n has quit IRC21:16
*** kgiusti has left #rdo21:17
*** Ryjedo has joined #rdo21:24
*** rpioso has quit IRC21:26
*** rhallisey has quit IRC21:32
*** mlammon has quit IRC21:39
*** rwsu has quit IRC21:40
*** dbecker has quit IRC21:42
*** sdake has quit IRC21:43
*** limao has joined #rdo21:49
*** limao has quit IRC21:50
*** limao_ has joined #rdo21:50
*** morazi has quit IRC21:56
*** rlandy has quit IRC21:59
*** eharney has quit IRC22:02
*** rwsu has joined #rdo22:03
*** florianf has quit IRC22:03
*** eharney has joined #rdo22:03
*** spr1 has quit IRC22:13
*** rwsu has quit IRC22:13
rdobot[sensu] NEW: - check-delorean-newton-current @ |#| Build failure on centos7-master/current: nova:
*** fultonj has joined #rdo22:18
*** Ryjedo has quit IRC22:23
*** rwsu has joined #rdo22:24
*** Ryjedo has joined #rdo22:24
*** Ryjedo has quit IRC22:25
*** Ryjedo has joined #rdo22:25
*** Ryjedo has quit IRC22:25
*** Ryjedo has joined #rdo22:26
*** Ryjedo has joined #rdo22:27
*** Ryjedo has quit IRC22:27
*** Ryjedo has joined #rdo22:28
*** elmiko is now known as _elmiko22:28
*** Ryjedo has joined #rdo22:29
*** Son_Goku has joined #rdo22:32
*** julim has joined #rdo22:37
*** rwsu has quit IRC22:40
*** Ryjedo_ has joined #rdo22:40
*** Ryjedo has quit IRC22:40
*** Ryjedo_ is now known as Ryjedo22:40
*** rwsu has joined #rdo22:45
*** limao has joined #rdo22:48
*** limao_ has quit IRC22:51
*** limao has quit IRC22:51
*** limao has joined #rdo22:51
*** kaminohana has joined #rdo22:55
ggilliesrbowen: that rdo newsletter you sent out with the links to mitaka-222:59
ggilliesthose looks like repo files for newton master to me?22:59
*** kaminohana has quit IRC23:00
*** dgurtner has quit IRC23:00
*** kaminohana has joined #rdo23:00
*** sdake has joined #rdo23:03
*** limao_ has joined #rdo23:12
*** stanchan has joined #rdo23:15
*** limao has quit IRC23:16
*** aortega has joined #rdo23:18
*** limao_ has quit IRC23:26
apevecggillies, back from the future :)23:30
*** Ryjedo_ has joined #rdo23:33
*** Ryjedo has quit IRC23:35
*** apevec has quit IRC23:36
*** gfidente has quit IRC23:37
*** Ryjedo_ has quit IRC23:38
*** gfidente has joined #rdo23:38
*** gfidente has quit IRC23:38
*** gfidente has joined #rdo23:38
*** jrist has quit IRC23:49
*** sdake has quit IRC23:51
*** maeca2 has left #rdo23:55
*** sdake has joined #rdo23:55

Generated by 2.14.0 by Marius Gedminas - find it at!