Monday, 2019-10-14

*** mixor has joined #openstack-kolla00:02
*** mixor has quit IRC01:58
*** calbers has quit IRC02:06
*** calbers has joined #openstack-kolla02:09
*** BjoernT has joined #openstack-kolla02:11
openstackgerritcaoyuan proposed openstack/kolla-cli master: tox: Move the whitelist_externals to testenv section  https://review.opendev.org/68830603:06
openstackgerritMerged openstack/kolla-cli master: tox: Move the whitelist_externals to testenv section  https://review.opendev.org/68830603:43
*** k_mouza has joined #openstack-kolla03:43
*** k_mouza has quit IRC03:48
*** dave-mccowan has quit IRC04:18
*** BjoernT has quit IRC04:24
openstackgerritjacky06 proposed openstack/kolla-cli master: Add --hosts support for check action  https://review.opendev.org/68832504:43
openstackgerritcaoyuan proposed openstack/kolla-cli master: Add --hosts support for check action  https://review.opendev.org/68832505:03
openstackgerritcaoyuan proposed openstack/kolla-cli master: Add --hosts flag into deploy unit test  https://review.opendev.org/68832705:29
openstackgerritEddie Yen proposed openstack/kolla master: Add disk dev name check function  https://review.opendev.org/68832805:32
openstackgerritMerged openstack/kolla-cli master: Add --hosts support for check action  https://review.opendev.org/68832505:35
openstackgerritDincer Celik proposed openstack/kolla-ansible stable/stein: Remove /etc/hosts entries pointing hostname to localhost and prevent cloud-init to manage /etc/hosts  https://review.opendev.org/68832905:35
*** jbadiapa has joined #openstack-kolla05:36
openstackgerritEddie Yen proposed openstack/kolla master: Add disk dev name check function  https://review.opendev.org/68832805:39
openstackgerritcaoyuan proposed openstack/kolla-cli master: Add --hosts flag into deploy unit test  https://review.opendev.org/68832705:48
*** cah_link has joined #openstack-kolla06:06
openstackgerritcaoyuan proposed openstack/kolla-cli master: Add --hosts flag into deploy unit test  https://review.opendev.org/68832706:06
*** dpawlik has joined #openstack-kolla06:23
*** pcaruana has joined #openstack-kolla06:43
mnasiadkamorning07:11
*** gfidente has joined #openstack-kolla07:37
*** rpittau|afk is now known as rpittau07:40
openstackgerritMichal Nasiadka proposed openstack/kolla master: DNM: Train RDO RC testing  https://review.opendev.org/68761507:41
*** ivve has joined #openstack-kolla07:46
yoctozeptomorning07:47
mnasiadkaseems we should be good with RDO, the only unbuildable thing is tripleoclient due to the fact tripleo did not do an RC yet07:53
*** gfidente has quit IRC07:59
*** dpawlik has quit IRC08:03
*** dougsz has joined #openstack-kolla08:06
*** iniazi_ has joined #openstack-kolla08:09
*** iniazi has quit IRC08:12
*** k_mouza has joined #openstack-kolla08:12
*** kevinz has joined #openstack-kolla08:14
hrwgooooood08:24
hrwmorning08:24
*** rgogunskiy has joined #openstack-kolla08:24
yoctozeptomnasiadka: nice :-)08:32
mnasiadkaalthough I see UCA is not in the same state :D08:32
mnasiadkahmm, seems Ceph broke at some point - maybe after changing the keys a bit08:34
yoctozeptomnasiadka: broken where08:42
openstackgerritRadosław Piliszek proposed openstack/kolla-ansible master: Set RabbitMQ cluster_partition_handling to pause_minority  https://review.opendev.org/67256208:42
*** kevinz has quit IRC08:44
openstackgerritRadosław Piliszek proposed openstack/kolla-ansible master: Set RabbitMQ cluster_partition_handling to pause_minority  https://review.opendev.org/67256208:44
hrwmnasiadka: will we manage to move kolla-ansible to use py3 in Ussuri?08:44
mnasiadkayoctozepto: http://zuul.openstack.org/build/2e89e20098ef45fb876209ae3d3f700a/log/primary/logs/ansible/deploy#5792508:45
mnasiadkahrw: first we need to identify what needs to be done :-)08:45
hrwD:08:49
*** gfidente has joined #openstack-kolla08:51
hrwfsck. managed to remove my _do_all_to_deploy.sh script08:54
yoctozeptohrw: happens08:54
hrwand it expired from paste.debian ;(08:54
yoctozeptomnasiadka: ah yeah, looking bad08:55
yoctozeptohrw: :<08:55
openstackgerritcaoyuan proposed openstack/kolla-cli master: Add Generating Documentation  https://review.opendev.org/68835508:59
mnasiadkayoctozepto: well, I haven't touched that in any change, so that is even more interesting08:59
yoctozeptomnasiadka: ubuntu b0rken?09:01
mnasiadkayoctozepto: seems so, like some path missing...09:05
mnasiadkabut fetch_ceph_keys.py is in /usr/bin/09:06
mnasiadkagah09:06
hrwmeh...09:07
hrw(venv3) 09:06 (0s) hrw@j13-qrep-04:kolla-ansible$ kolla-genpwd -p $PWD/hrw/passwords.yml09:08
hrwFileNotFoundError: [Errno 2] No such file or directory: '/home/hrw/devel/kolla/kolla-ansible/hrw/passwords.yml'09:08
hrw(venv3) 09:06 (0s) hrw@j13-qrep-04:kolla-ansible$ touch hrw/passwords.yml09:08
hrw(venv3) 09:08 (0s) hrw@j13-qrep-04:kolla-ansible$ kolla-genpwd -p $PWD/hrw/passwords.yml09:08
hrwAttributeError: 'NoneType' object has no attribute 'items'09:08
hrwso to generate passwords I need to generate passwords?09:08
yoctozeptohrw: it needs the template used09:10
yoctozeptoit's kind of counterintuitive09:10
yoctozeptobut yeah09:10
hrwthanks09:10
stingrayzamorning folks o/09:12
*** dpawlik has joined #openstack-kolla09:13
*** dpawlik has quit IRC09:18
*** skramaja has joined #openstack-kolla09:20
openstackgerritWill Szumski proposed openstack/kayobe master: format pxe_append_params as string and not a list  https://review.opendev.org/68810209:21
openstackgerritWill Szumski proposed openstack/kayobe master: Fix issue with image download when checksum url is none  https://review.opendev.org/68835909:21
openstackgerritWill Szumski proposed openstack/kayobe master: format pxe_append_params as string and not a list  https://review.opendev.org/68810209:22
*** vmixor has joined #openstack-kolla09:23
openstackgerritWill Szumski proposed openstack/kayobe master: Fix issue with image download when checksum url is none  https://review.opendev.org/68835909:24
hrw~hail 96GB ram for allowing /var/lib/docker to be in tmpfs09:25
openstackgerritWill Szumski proposed openstack/kayobe-config-dev master: DNM: Switch to CentOS based IPA images  https://review.opendev.org/68783009:25
yoctozeptomnasiadka: yup, http://zuul.openstack.org/builds?job_name=kolla-ansible-ubuntu-source-ceph&branch=master09:29
yoctozeptovery b0rken09:29
*** dpawlik has joined #openstack-kolla09:30
yoctozeptoonly https://review.opendev.org/687544 to blame mnasiadka09:31
yoctozeptothough needs investigating why09:31
*** brtknr_ has quit IRC09:31
mnasiadkayoctozepto: well, and it caused no such file or directory when running a python script? :D09:31
*** brtknr has joined #openstack-kolla09:32
yoctozeptomnasiadka: hint: upgrades not broken http://zuul.openstack.org/builds?job_name=kolla-ansible-ubuntu-source-ceph&branch=master&job_name=kolla-ansible-ubuntu-source-upgrade-ceph#09:32
yoctozeptowell, no idea09:32
yoctozeptocould be something the script wants09:32
mnasiadkahttps://www.irccloud.com/pastebin/lKmGRdBp/09:32
mnasiadkabingo09:32
yoctozeptobut only fresh deploys are affected09:32
yoctozeptomnasiadka: yeah but that is a lol09:33
hrwhow could you run unversioned python...09:33
yoctozeptoas how could your change break that then09:33
mnasiadkait's not my change - I guess we stopped installing python2 in ubuntu images?09:33
mgoddardmorning all09:33
hrwmnasiadka: s/python2/unversioned-python/ even09:34
yoctozeptobut when mnasiadka09:34
yoctozeptoyou mean ubuntu stopped shipping it for us by default?09:34
yoctozeptomorning mgoddard09:34
yoctozeptoboth bosses agree to your trip to Białystok? :-)09:35
hrwspeaking of it...09:35
yoctozeptotoday/tomorrow I should know if we are going to do it this month09:35
hrwhow many days we want for it?09:35
yoctozeptoyeah, that's important too, as it makes coordination different09:36
hrwand can we do it on 6th Nov? so I can go for Red Hat Forum 2019 Warsaw on 5th and next day drive to Białystok09:36
yoctozeptohrw: thanks for reminding09:37
yoctozeptoI should register immediately09:38
hrwyoctozepto: RHF?09:38
yoctozeptohrw: yup09:41
yoctozeptohrw: 6th Nov is actually bad as I have a doctor's appointment then, at least during the morning and then depends on how I feel :-)09:44
hrwyoctozepto: 6th as 'let arrive in Białystok' and then 7-8th meetings, 9th go home09:44
yoctozeptosounds like a plan09:46
yoctozeptomnasiadka, mgoddard: ^ if you could confirm that09:46
yoctozeptoI would be able to work on detail09:46
yoctozeptodetails*09:46
mnasiadkaLet's await mgoddard's attention :)09:48
mgoddardjust speaking to boss 109:48
mnasiadkahrw: so, we have python2.7 in CentOS and python3 in Ubuntu - what should I use so it works on both? :D09:48
mnasiadkamgoddard: well, you can include me in the financial plan ;)09:48
hrwmnasiadka: proper ansible tasks to be sure that you have unversioned python installed?09:49
mnasiadkahrw: ugh, I’ll probably rework that crap and get rid of this python cli script09:55
hrw;D09:55
openstackgerritViktor Michalek proposed openstack/kolla-ansible master: Neutron: Added option to force legacy iptables in neutron_openvswitch_agent  https://review.opendev.org/68596709:57
*** k_mouza has quit IRC10:07
hrwINFO:kolla.common.utils.aodh-api:Trying to push the image10:12
hrwERROR:kolla.common.utils.aodh-api:Get http://10.101.16.1:5000/v2/: dial tcp 10.101.16.1:5000: connect: no route to host10:12
hrwINFO:kolla.common.utils.aodh-api:Pushed successfully10:12
hrw;D10:12
hrwhave to take a look at code...10:13
mnasiadkalol10:19
hrwanother piece of kolla code where technical debt has to be paid10:20
openstackgerritjacky06 proposed openstack/kolla-ansible master: tox: Keeping going with docs  https://review.opendev.org/68837110:21
mgoddardhrw: was it a retry?10:22
hrwmgoddard: first. my builds do not retry10:22
openstackgerritjacky06 proposed openstack/kolla master: tox: Keeping going with docs  https://review.opendev.org/68837310:24
*** k_mouza has joined #openstack-kolla10:27
openstackgerritWill Szumski proposed openstack/kolla-ansible master: Add extras directory to prometheus config  https://review.opendev.org/68837410:28
openstackgerritjacky06 proposed openstack/kolla master: tox: Keeping going with docs  https://review.opendev.org/68837310:33
hrwand I see why it goes that way10:34
hrwin case of error set status to error. and then set to built anyway10:37
openstackgerritEddie Yen proposed openstack/kolla master: Add disk dev name check function  https://review.opendev.org/68832810:43
hrwINFO:kolla.common.utils.base:Trying to push the image10:48
hrwERROR:kolla.common.utils.base:Get https://some.example.com:8432/v2/: dial tcp: lookup some.example.com: no such host10:48
hrwINFO:kolla.common.utils:===========================10:48
hrwINFO:kolla.common.utils:Images that failed to build10:48
hrwINFO:kolla.common.utils:===========================10:48
hrwERROR:kolla.common.utils:base Failed with status: push_error10:48
hrwnow it looks better10:48
yoctozeptohrw: ah, mgoddard "fixed" that and I accepted :D10:52
yoctozeptosee https://review.opendev.org/68320010:52
yoctozeptowell, you did too10:53
yoctozeptothe technical is "to be paid" as always10:53
yoctozeptodebt*10:53
yoctozepto;D10:53
hrwcheck http://paste.debian.net/1106572/10:53
yoctozeptoexceptions ftw10:54
hrwif there is an error during push then raise up and consider it done10:54
yoctozeptowhat about retries though, sir10:54
hrwretries should be forbidden ;D10:55
yoctozeptoxD10:55
hrwtested - it just retries10:59
hrwINFO:kolla.common.utils:Attempt number: 4 to run task: PushTask(base)11:00
hrwINFO:kolla.common.utils.base:Trying to push the image11:00
hrwERROR:kolla.common.utils.base:Get https://some.example.com:8432/v2/: dial tcp: lookup some.example.com: no such host11:00
hrw1/2/3/411:00
openstackgerritEddie Yen proposed openstack/kolla master: Add disk dev name check function  https://review.opendev.org/68832811:01
hrwbtw - kolla/image/build.py will probably need split as it has 1481 lines already11:01
openstackgerritMerged openstack/kolla-ansible stable/stein: Remove /etc/hosts entries pointing hostname to localhost and prevent cloud-init to manage /etc/hosts  https://review.opendev.org/68832911:04
yoctozeptoit does11:08
*** dswebb has quit IRC11:10
*** kplant has joined #openstack-kolla11:16
mgoddardhrw: change looks sensible. use logger.exception rather than logger.error though11:18
hrwmgoddard: good catch11:18
*** pcaruana has quit IRC11:34
*** dave-mccowan has joined #openstack-kolla11:39
*** pcaruana has joined #openstack-kolla11:41
hrwmgoddard: no. logger.error is proper. logger.exception shows traceback which is not relevant here11:44
*** k_mouza has quit IRC11:46
openstackgerritMarcin Juszkiewicz proposed openstack/kolla master: handle push error properly  https://review.opendev.org/68839011:46
*** k_mouza has joined #openstack-kolla11:47
*** hjensas has quit IRC11:54
openstackgerritMarcin Juszkiewicz proposed openstack/kolla master: handle push error properly  https://review.opendev.org/68839012:02
hrwyoctozepto: can you extend what you mean about commit message?12:02
yoctozeptohrw: the change id is above your message12:04
yoctozeptoand they got treated as tags in there12:04
hrwmoved12:04
yoctozeptoand about the revert - it should be in the commit message for posterity12:05
yoctozeptoand is it possible to have this tested12:05
yoctozeptoremoving unit tests is a bad practice ;p12:05
*** FlorianFa has joined #openstack-kolla12:05
hrwhave to check it ;d12:06
openstackgerritMichal Nasiadka proposed openstack/kolla-ansible master: Allow fetch ceph keyrings to work with unversioned python binary  https://review.opendev.org/68839212:07
openstackgerritWill Szumski proposed openstack/kolla-ansible master: Add extras directory to prometheus config  https://review.opendev.org/68837412:12
*** jroll has quit IRC12:14
*** jroll has joined #openstack-kolla12:15
yoctozeptomgoddard, mnasiadka: quick for review: https://review.opendev.org/67256212:20
yoctozepto"quickie" ;-)12:20
mgoddardyoctozepto: did you look at the issue raised by noxoid?12:21
yoctozeptomgoddard: I could not reproduce12:21
yoctozeptoand no google results12:21
yoctozeptonothing I can do12:21
yoctozeptobut we surely need this done as default12:22
yoctozeptoat lest in all new deployments12:22
*** schwicht has joined #openstack-kolla12:24
openstackgerritWill Szumski proposed openstack/kayobe master: Fix issue with image download when checksum url is none  https://review.opendev.org/68835912:25
*** devep has joined #openstack-kolla12:25
mgoddardyoctozepto: did you speak to noxoid about it?12:26
yoctozeptomgoddard: hmm, in all my wisdom I have forgot about this simple thing12:27
mgoddardyoctozepto: communication :)12:27
yoctozeptonoxoid: please let me know if there are any details regarding the partitioning control in rabbitmq - just ping me :-)12:28
yoctozeptomgoddard: yeah, I usually remember12:28
yoctozeptotoo much ipv612:28
yoctozeptoand having said that, I think you should review that sooner than later ;-)12:28
*** schwicht has quit IRC12:30
openstackgerritMerged openstack/kolla-ansible master: Do not install ntp package on host when not used  https://review.opendev.org/68827012:32
*** schwicht has joined #openstack-kolla12:33
mgoddardyoctozepto: I'll scratch yours if you scratch mine ;)12:34
mgoddard(cells)12:34
yoctozeptomgoddard: yeah, I have it open in a tab12:34
yoctozeptobut mine is simpler12:35
mgoddardI'm currently writing docs12:35
yoctozeptotrust me12:35
mgoddardno doubt, but I have already given you one review :p12:35
yoctozeptooh and docs are being written12:35
yoctozeptotrue12:35
mgoddardcomplexity is why I'd like reviews sooner than later12:35
yoctozeptook, so we will be trading w+1s now12:35
yoctozepto:D12:35
mgoddardespecially as I'm out thursday and friday this week12:35
yoctozeptooh my, then it's top prio :O12:36
*** sm806 has quit IRC12:36
*** sm806 has joined #openstack-kolla12:37
mgoddarddougsz should be around though12:37
dougsz👍12:39
*** schwicht has quit IRC12:46
hrwbtw - https://review.opendev.org/#/c/399426/ is interesting12:49
hrwspeaking of out... I will skip Wed meeting12:50
openstackgerritMarcin Juszkiewicz proposed openstack/kolla master: handle push error properly  https://review.opendev.org/68839012:53
hrwjust added bug12:54
*** dswebb has joined #openstack-kolla12:54
yoctozeptohrw: why is that interesting?12:55
hrwyoctozepto: bug shown sensible usecase12:56
hrwthe 'I am in internal network and the only registry I have requires login but I forgot to login before run kolla'12:56
mgoddardhrw:  you missed this comment? "Please don't revert my patch, or you'll break retries again. Please test a retry that eventually succeeds."12:57
hrwmgoddard: one step at time, ok?12:57
yoctozeptoxD12:57
mgoddardhrw: two steps please12:57
yoctozeptodancers12:58
hrwmgoddard: your test assumes that retries are in PushTask when they are outside of it12:58
*** Guest31368 has joined #openstack-kolla12:58
yoctozeptothen rewrite :D12:58
hrwyoctozepto: I did. by dropping it12:59
hrw;d12:59
yoctozeptoyeah, indeed12:59
hrwnow looking at how to test it right12:59
yoctozeptogood boy12:59
yoctozepto;D12:59
*** hjensas has joined #openstack-kolla13:00
*** Guest31368 is now known as redrobot_13:00
hrwbuild.PushTask->run() is expected to raise exception once there is any error13:00
*** redrobot_ is now known as redrobot13:01
hrwbut in tox it is run in a way that exception is handled other way13:02
yoctozeptohow so13:02
hrwPushTask->run() is called by tox, calls push_image() which errors out, raise generic exception, gets back to PushTask->run() but not as PushError13:03
yoctozeptohmmm13:04
yoctozeptoI still don't get it how tox is doing that13:05
yoctozeptotox is just running some stuff in a virtual environment13:05
yoctozeptoyou must mean something else13:05
hrwself.dc is mocked13:06
*** Guest69423 has quit IRC13:06
hrwso self.dc.push() call goes to mock_client->push() which probably just raise Exception and consider it done13:06
yoctozeptoah, you mean the mock13:07
mgoddardhrw: I don't think my test assumes retries are in PushTask - it calls task.run() twice13:07
hrwmgoddard: but we go back to run() in a wrong way.13:07
mgoddardhrw: why?13:07
openstackgerritMerged openstack/kayobe master: format pxe_append_params as string and not a list  https://review.opendev.org/68810213:08
hrwmgoddard: look at build.PushTask.push_image() function13:08
mgoddardhrw: in the case I was testing, the Docker client raises an exception13:08
mgoddardyour case is different - it returns an error message, so we raise a new exception13:09
hrwmgoddard: 'for response in self.dc.push()' call goes to mock, mock raises GENERIC exception, build.PushTask.run() gets into wrong part of Exception handling13:09
mgoddardhrw: that was what happened in the bug that was raised13:10
mgoddardhrw: we saw "Unknown error when pushing"13:10
hrwmgoddard: and once PushTask->run() fails then we have PushTask->run() added into queue if there are any retries left13:10
mgoddardif you want to test a different path, add another new unit test13:10
yoctozeptounit test, unit test13:10
mgoddardhrw: here's the bug: https://bugs.launchpad.net/tripleo/+bug/184469713:11
openstackLaunchpad bug 1844697 in tripleo "[kolla] periodic: container build job can fail during push" [Undecided,New]13:11
hrwok13:12
*** schwicht has joined #openstack-kolla13:12
openstackgerritMerged openstack/kolla-ansible master: setup.cfg: Cleanup  https://review.opendev.org/68829313:14
*** schwicht has quit IRC13:14
*** schwicht has joined #openstack-kolla13:14
*** openstackstatus has quit IRC13:15
hrwimho whole 'image.status = STATUS_BUILT' inside of push is wrong. it should be done before as we do not push unbuilt images13:17
mgoddardhrw: built == success13:20
hrwmgoddard: built == built, exists as docker image in local docker storage. we can now push13:20
hrwif need13:21
mgoddardwell yes, but in terms of our state machine, if we're pushing images then they end up in 'built' on success13:21
mgoddardyou'd need to it to change that13:22
mgoddard*to modify13:22
hrwok13:23
openstackgerritWill Szumski proposed openstack/kayobe stable/stein: format pxe_append_params as string and not a list  https://review.opendev.org/68840313:24
openstackgerritWill Szumski proposed openstack/kayobe stable/rocky: format pxe_append_params as string and not a list  https://review.opendev.org/68840413:25
openstackgerritWill Szumski proposed openstack/kayobe stable/queens: format pxe_append_params as string and not a list  https://review.opendev.org/68840513:25
*** BjoernT has joined #openstack-kolla13:26
openstackgerritMarcin Juszkiewicz proposed openstack/kolla master: handle push error properly  https://review.opendev.org/68839013:27
hrwthis version should satisfy13:28
openstackgerritMichal Nasiadka proposed openstack/kolla-ansible master: Allow fetch ceph keyrings to work with unversioned python binary  https://review.opendev.org/68839213:30
openstackgerritViktor Michalek proposed openstack/kolla-ansible master: Neutron: Added option to force legacy iptables in neutron_openvswitch_agent  https://review.opendev.org/68596713:32
hrwtox passed13:44
*** dougsz has quit IRC13:47
*** BjoernT_ has joined #openstack-kolla13:47
*** BjoernT has quit IRC13:49
*** dougsz has joined #openstack-kolla14:04
*** munimeha1 has joined #openstack-kolla14:06
*** dpawlik has quit IRC14:07
noxoidyoctozepto, re rabbitmq partitioning. im afraid i can only offer anecdotal evidence. my cluster may have been silently partitioned when i made the change to pause_minority. i have since stopped relying solely on cluster_health and added "rabbitmqctl node_health_check" to a list of things to check14:11
noxoidif you cant replicate then i wouldnt worry about it. although adding a node_health_check before converting may be useful as cluster_health reported no partitions but node health checks would fail on stuck processes14:13
*** Blinkiz has joined #openstack-kolla14:18
*** priteau has joined #openstack-kolla14:20
hrwpep8 is funny14:21
hrwyou fix W503 and then it became W504 ;D14:21
*** ivve has quit IRC15:05
openstackgerritJames Kirsch proposed openstack/kolla-ansible master: Custom CA certificates deployment  https://review.opendev.org/68602415:23
openstackgerritJoseph M proposed openstack/kolla stable/stein: Bump stein octavia to 4.1.0 (CVE-2019-17134)  https://review.opendev.org/68842615:31
noxoid^ fixes CVE-2019-1713415:31
noxoidat least on the kolla end15:31
mgoddardnice, thanks noxoid15:33
noxoiddont thank me yet, i didnt propose changes for the other affected stable branches :P15:33
noxoidi guess i can go ahead and do that15:33
mgoddardplease :)15:34
openstackgerritJoseph M proposed openstack/kolla stable/rocky: Bump rocky octavia to 3.2.0 (CVE-2019-17134)  https://review.opendev.org/68842915:38
yoctozeptonoxoid: thanks!15:40
mnasiadkayoctozepto: https://review.opendev.org/#/c/675659 - go have a look and let's finish this cells soap opera :)15:41
openstackgerritJoseph M proposed openstack/kolla stable/queens: Bump queens octavia to 2.1.2 (CVE-2019-17134)  https://review.opendev.org/68843015:42
yoctozeptomnasiadka: unfortunately it will take more than a "look" for me :-)15:42
yoctozeptoI'm too thorough15:42
mnasiadkayoctozepto: oh come on, Mark and Doug have been polishing it for too long :)15:42
yoctozeptobut I'm into closing this today/tomorrow15:43
yoctozeptoI am sure of their dedication though it's about doing the review15:43
yoctozeptonot honorary w+1 :-)15:43
openstackgerritMichal Nasiadka proposed openstack/kolla-ansible master: Allow fetch ceph keyrings to work with unversioned python binary  https://review.opendev.org/68839215:44
mgoddardyoctozepto: comments welcome15:44
mnasiadkayoctozepto: just joking :)15:44
mnasiadkaso - do we have any features that we need to squeeze in Train?15:44
mnasiadka(apart the cells one)?15:44
mnasiadkaugh, I need to raise two more issues in ceph-ansible...15:45
mgoddardmnasiadka: IP615:45
openstackgerritMark Goddard proposed openstack/kolla-ansible master: Docs: add nova cells  https://review.opendev.org/68843215:45
mnasiadkamgoddard: I was looking for something less complicated than cells :)15:46
mgoddardyoctozepto, mnasiadka: ^ some docs which may help in the grokking of code15:46
*** hamzaachi has joined #openstack-kolla15:51
*** hamzaachi has quit IRC15:52
yoctozeptomnasiadka: ipv6 is simpler but tedious15:53
noxoidk patches have been proposed for all stable versions where a fixed tarball is available on tarballs.openstack.org15:54
noxoidolder stable versions have had their release notes updated but there are no tarballs avail15:54
openstackgerritViktor Michalek proposed openstack/kolla-ansible master: Neutron: Added option to force legacy iptables in neutron_openvswitch_agent  https://review.opendev.org/68596716:02
*** rgogunskiy has quit IRC16:02
*** openstackstatus has joined #openstack-kolla16:03
*** ChanServ sets mode: +v openstackstatus16:03
hrwmnasiadka: can you just use ansible_python_interpreter var?16:06
mnasiadkahrw: and then ubuntu containers on centos stop working?:)16:06
*** rpittau is now known as rpittau|afk16:06
hrwmnasiadka: jsut asking16:07
hrwnoxoid: +2 on all. thanks16:07
noxoidno problem16:08
mnasiadkahrw: we could gather facts in the container image, but probably current approach is faster and simpler16:09
*** hamzaachi has joined #openstack-kolla16:10
hrwmnasiadka: /usr/bin/python{{ distro_python_version }}16:12
mnasiadkahrw: that var is only in kolla? Or we have that in k-a also?16:13
hrwboth16:13
hrwmnasiadka: https://review.opendev.org/#/c/675581/16:13
* hrw off16:17
*** igordc has joined #openstack-kolla16:21
openstackgerritMerged openstack/kayobe stable/queens: format pxe_append_params as string and not a list  https://review.opendev.org/68840516:22
*** dave-mccowan has quit IRC16:23
*** hamzaachi has quit IRC16:24
yoctozeptohrw, mnasiadka: so distro_python_version or not? :S16:25
*** hamzaachi has joined #openstack-kolla16:25
mnasiadkaGeez, static var16:26
mnasiadkaWhatever, I’ll use it.16:26
yoctozeptoWHATEVA :D16:28
*** BjoernT_ is now known as BjoernT16:34
*** k_mouza_ has joined #openstack-kolla16:44
yoctozeptomnasiadka: anyways, did you get why this command suddenly started failing?16:45
yoctozeptowas it just not run before16:46
yoctozeptoor what16:46
*** k_mouza has quit IRC16:47
*** devep has quit IRC16:49
*** k_mouza_ has quit IRC16:49
mnasiadkayoctozepto: well, that's the effect of non-voting CIs :)16:51
yoctozeptomnasiadka: that's what I get16:52
yoctozeptobut how was it done by your change16:52
yoctozeptois yet beyond me16:52
mnasiadkait was rather not my change, let me analyse when it started failing16:52
*** hamzaachi has quit IRC16:54
openstackgerritMark Goddard proposed openstack/kolla-ansible master: Docs: add nova cells  https://review.opendev.org/68843216:55
mnasiadkaok, it started failing on 10th in the evening16:55
openstackgerritMerged openstack/kayobe stable/rocky: format pxe_append_params as string and not a list  https://review.opendev.org/68840416:56
mnasiadkahumm, no changes that could cause this16:59
*** munimeha1 has quit IRC17:02
openstackgerritPierre Riteau proposed openstack/kayobe master: Try running `docker info` again if it fails  https://review.opendev.org/68847417:05
mnasiadkawhatever, not going to track bionic changefs17:05
mnasiadkachanges17:05
*** dougsz has quit IRC17:05
*** hjensas has quit IRC17:08
openstackgerritMichal Nasiadka proposed openstack/kolla-ansible master: Allow fetch ceph keyrings to work with unversioned python binary  https://review.opendev.org/68839217:09
*** priteau has quit IRC17:11
yoctozeptomnasiadka: and your change had this passing?17:16
mnasiadkayoctozepto: of course17:16
yoctozeptomnasiadka: though so17:16
yoctozepto(saw so)17:16
mnasiadkaanyway - normally I would rework this, but I don't like to meddle with a dead (deprecated) horse :)17:16
yoctozeptomnasiadka: dead indeed17:17
mnasiadkaat least ceph-ansible is nearly there, needs some more testing and some docs how to set it up17:18
*** hamzaachi has joined #openstack-kolla17:18
mnasiadkabut ceph-ansible guys broke key fetching in 4.017:18
yoctozeptooh my17:18
*** hjensas has joined #openstack-kolla17:20
*** BjoernT has quit IRC17:20
*** signed8bit has joined #openstack-kolla17:22
kplantmnasiadka: yeah....17:22
kplanti noticed that myself17:22
kplantit kind-of works with non-containerized17:22
kplantbut then, for some reason, the flow for containerized was totally different17:22
mnasiadkakplant: well, they fixed it for mds, but now seems rgw has the same problem17:23
mnasiadkaso I need to raise another issue17:23
kplanti was in the middle of submitting a package to make it symetrical17:23
kplantbut then i fell down :>17:23
mnasiadkakplant: I wouldn't bother, this is a downstream managed open source project17:23
yoctozeptouplifting kolla does better with its CI :-)17:23
*** k_mouza has joined #openstack-kolla17:23
kplantstable-3.2 was great in terms of keys17:24
kplanteven dumped the fsid to a nice file for you17:24
kplantnow, as far as i can tell, if you don't statically set it - you have no idea wqhat it is17:24
kplantjust hope it's the only dir in /etc/ceph/fetch/17:24
*** gmann is now known as gmann_afk17:25
mnasiadkakplant: they removed the fetch dir in 4.017:27
mnasiadka(just like this)17:27
mnasiadka:)17:27
mnasiadkano deprecation, no nothing :)17:27
kplantnah, it's still there17:27
kplanti use stable-4.0 myself17:27
kplant# Directory to fetch cluster fsid, keys etc...17:27
kplantfetch_directory: fetch/17:27
kplantit does output17:28
mnasiadkakplant: yeah, but they removed it from most of the plays, and in containerised mode - it does not put anything in there :)17:28
kplantyeah that's what i was mentioning earlier17:28
kplantfor containerized it's different which makes no sense17:28
mnasiadkakplant: and they plan to remove the rest - so I wouldn't stick to it :)17:28
kplantwell that's dumb17:28
kplanthow am i supposed to programmatically get my fsid and keys if they're not outputted?17:29
*** hjensas has quit IRC17:34
*** hjensas has joined #openstack-kolla17:37
mnasiadkakplant: and you think they care? RH just cares for their own customers using this :)17:38
mnasiadkaok, over and out :)17:43
kplantlol - you mean ibm!17:47
*** hamzaachi has quit IRC17:47
openstackgerritMerged openstack/kayobe stable/stein: format pxe_append_params as string and not a list  https://review.opendev.org/68840317:51
*** verdurin has quit IRC17:51
*** k_mouza has quit IRC17:51
*** verdurin has joined #openstack-kolla17:51
*** ivve has joined #openstack-kolla17:53
*** k_mouza has joined #openstack-kolla18:17
openstackgerritMerged openstack/kolla master: Custom CA certificates installation mechanism  https://review.opendev.org/68612118:24
*** cah_link1 has joined #openstack-kolla18:25
*** cah_link has quit IRC18:27
*** cah_link1 is now known as cah_link18:27
*** schwicht has quit IRC18:38
*** schwicht has joined #openstack-kolla18:38
openstackgerritMerged openstack/kolla stable/rocky: Bump rocky octavia to 3.2.0 (CVE-2019-17134)  https://review.opendev.org/68842918:41
openstackgerritMerged openstack/kolla stable/stein: Bump stein octavia to 4.1.0 (CVE-2019-17134)  https://review.opendev.org/68842618:41
yoctozeptomgoddard: are you doing the ascii graphics by hand or using some tool? cause they look nice18:47
yoctozeptoalso in kayobe case18:47
*** factor has quit IRC18:50
*** k_mouza has quit IRC19:11
*** Jeffrey4l has quit IRC19:23
*** Jeffrey4l has joined #openstack-kolla19:24
*** kplant has quit IRC19:38
*** dpawlik has joined #openstack-kolla19:56
*** gfidente has quit IRC19:57
*** gmann_afk is now known as gmann20:01
openstackgerritMerged openstack/kolla master: setup.cfg: Cleanup  https://review.opendev.org/68827420:17
*** BjoernT has joined #openstack-kolla20:24
*** pcaruana has quit IRC20:26
*** benj has joined #openstack-kolla20:28
*** benj is now known as Guest3991020:28
*** dpawlik has quit IRC20:31
*** BjoernT_ has joined #openstack-kolla20:32
*** BjoernT has quit IRC20:32
*** dpawlik has joined #openstack-kolla20:42
*** dpawlik has quit IRC20:46
*** ldumont has joined #openstack-kolla21:04
*** LaurentDumont has quit IRC21:05
*** k_mouza has joined #openstack-kolla21:12
openstackgerritMerged openstack/kolla stable/queens: Bump queens octavia to 2.1.2 (CVE-2019-17134)  https://review.opendev.org/68843021:12
*** k_mouza has quit IRC21:16
*** goldyfruit has joined #openstack-kolla21:20
*** vmixor has quit IRC21:22
*** k_mouza has joined #openstack-kolla21:23
*** schwicht has quit IRC21:37
*** schwicht has joined #openstack-kolla21:40
*** k_mouza has quit IRC21:41
*** cah_link has quit IRC21:49
*** jistr has quit IRC21:52
*** vesper11- has joined #openstack-kolla21:53
*** vesper11 has quit IRC21:53
*** jistr has joined #openstack-kolla22:01
*** jistr has quit IRC22:02
*** jistr has joined #openstack-kolla22:03
*** schwicht has quit IRC22:32
*** ivve has quit IRC22:49
*** schwicht has joined #openstack-kolla22:53
*** schwicht has quit IRC23:29
*** vmixor has joined #openstack-kolla23:34
*** k_mouza has joined #openstack-kolla23:41
*** k_mouza has quit IRC23:46
*** schwicht has joined #openstack-kolla23:54

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!