*** threestrands has joined #openstack-ansible | 00:01 | |
*** pedja has joined #openstack-ansible | 00:02 | |
*** pedja has quit IRC | 00:04 | |
*** pedja has joined #openstack-ansible | 00:04 | |
*** pedja has quit IRC | 00:05 | |
*** pedja has joined #openstack-ansible | 00:06 | |
*** pedja has quit IRC | 00:07 | |
*** pedja has joined #openstack-ansible | 00:07 | |
*** pedja has quit IRC | 00:08 | |
*** TxGirlGeek has quit IRC | 00:11 | |
*** chyka has quit IRC | 00:11 | |
*** davidy has joined #openstack-ansible | 00:14 | |
davidy | hi folks :) | 00:14 |
---|---|---|
*** davidy is now known as funkypenguin | 00:15 | |
funkypenguin | right, that's better | 00:15 |
funkypenguin | I'm having an issue with OSA, building a repo container. OSA seems to want to clone every openstack git repo in parallel, which (on my limited connectivity here in NZ) times out before the playbook completes | 00:16 |
funkypenguin | is there a way to avoid this? | 00:16 |
*** SerenaFeng has quit IRC | 00:19 | |
*** yifei has joined #openstack-ansible | 00:20 | |
*** SerenaFeng has joined #openstack-ansible | 00:21 | |
*** gouthamr has quit IRC | 00:25 | |
SamYaple | funkypenguin: you could clone then all yourself then rerun the playbooks | 00:28 |
funkypenguin | mmm. turns out it may be some sort of network fault. even cloning manually is killer-slow | 00:36 |
funkypenguin | exhibit A: | 00:37 |
funkypenguin | remote: Compressing objects: 100% (77332/77332), done. | 00:37 |
funkypenguin | Receiving objects: 0% (1349/170492), 372.00 KiB | 38 KiB/s | 00:37 |
*** SerenaFeng has quit IRC | 00:43 | |
*** weezS has quit IRC | 00:43 | |
*** savvas has joined #openstack-ansible | 00:44 | |
*** vnogin has joined #openstack-ansible | 00:45 | |
*** evrardjp[m] has joined #openstack-ansible | 00:45 | |
*** vnogin has quit IRC | 00:49 | |
*** cmart has quit IRC | 00:49 | |
*** csmart has joined #openstack-ansible | 00:54 | |
*** germs has quit IRC | 00:54 | |
*** cmart has joined #openstack-ansible | 01:07 | |
*** esberglu has joined #openstack-ansible | 01:16 | |
*** poopcat has quit IRC | 01:17 | |
*** poopcat has joined #openstack-ansible | 01:17 | |
*** savvas has quit IRC | 01:17 | |
*** esberglu has quit IRC | 01:21 | |
*** ajmaidak has joined #openstack-ansible | 01:23 | |
yifei | Hi guys | 01:24 |
yifei | cannot Add GPG key for COPR LXC repo | 01:24 |
yifei | seems that "copr-be.cloud.fedoraproject.org" is down | 01:25 |
*** poopcat has quit IRC | 01:27 | |
*** savvas has joined #openstack-ansible | 01:32 | |
SamYaple | funkypenguin: try the github mirror, or the git.openstack.org one | 01:34 |
SamYaple | what ever you have, try the other | 01:34 |
*** germs has joined #openstack-ansible | 01:38 | |
*** poopcat has joined #openstack-ansible | 01:44 | |
*** vnogin has joined #openstack-ansible | 01:45 | |
*** vnogin has quit IRC | 01:50 | |
*** mcarden has quit IRC | 02:02 | |
*** woodard has quit IRC | 02:02 | |
*** woodard has joined #openstack-ansible | 02:02 | |
*** mcarden has joined #openstack-ansible | 02:02 | |
*** savvas has quit IRC | 02:06 | |
*** hw_wutianwei has joined #openstack-ansible | 02:07 | |
cloudnull | yifei: sadly COPR (fedoraproject) has been up and down a lot lately . | 02:30 |
yifei | cloudnull: that's awful...but thanks for your information | 02:32 |
yifei | cloudnull: excuse me, is there a variable in OSA that can be enabled to collect all the logs into one directory? | 02:33 |
*** madorn has quit IRC | 02:35 | |
*** madorn has joined #openstack-ansible | 02:37 | |
*** savvas has joined #openstack-ansible | 02:42 | |
*** poopcat1 has joined #openstack-ansible | 02:44 | |
*** poopcat has quit IRC | 02:44 | |
*** vnogin has joined #openstack-ansible | 02:46 | |
*** vnogin has quit IRC | 02:51 | |
*** poopcat1 has quit IRC | 02:57 | |
*** poopcat has joined #openstack-ansible | 02:57 | |
*** savvas has quit IRC | 03:16 | |
*** savvas has joined #openstack-ansible | 03:22 | |
*** savvas__ has quit IRC | 03:25 | |
*** dave-mccowan has quit IRC | 03:26 | |
*** poopcat has quit IRC | 03:38 | |
*** poopcat has joined #openstack-ansible | 03:38 | |
*** savvas_ has joined #openstack-ansible | 03:45 | |
*** vnogin has joined #openstack-ansible | 03:47 | |
*** vnogin has quit IRC | 03:51 | |
*** udesale has joined #openstack-ansible | 03:54 | |
*** savvas__ has joined #openstack-ansible | 03:57 | |
*** savvas has quit IRC | 03:59 | |
*** savvas has joined #openstack-ansible | 04:08 | |
*** funkypen_ has joined #openstack-ansible | 04:09 | |
*** funkypenguin has quit IRC | 04:12 | |
*** savvas__ has quit IRC | 04:12 | |
*** chyka has joined #openstack-ansible | 04:13 | |
*** funkypen_ has quit IRC | 04:14 | |
*** openstackgerrit has joined #openstack-ansible | 04:16 | |
openstackgerrit | Taseer Ahmed proposed openstack/openstack-ansible-os_congress master: Introduce os_congress role in gerrit https://review.openstack.org/522491 | 04:16 |
*** savvas_ has quit IRC | 04:18 | |
*** chyka has quit IRC | 04:19 | |
*** savvas_ has joined #openstack-ansible | 04:45 | |
*** vnogin has joined #openstack-ansible | 04:48 | |
*** funkypenguin has joined #openstack-ansible | 04:52 | |
*** vnogin has quit IRC | 04:53 | |
*** adreznec has joined #openstack-ansible | 04:54 | |
*** adreznec has quit IRC | 04:55 | |
*** funkypenguin has quit IRC | 04:57 | |
*** adreznec has joined #openstack-ansible | 04:57 | |
*** bhujay has joined #openstack-ansible | 05:02 | |
*** nshetty has joined #openstack-ansible | 05:03 | |
*** poopcat has quit IRC | 05:06 | |
*** bhujay has quit IRC | 05:18 | |
*** bhujay has joined #openstack-ansible | 05:18 | |
*** savvas_ has quit IRC | 05:19 | |
*** threestrands has quit IRC | 05:24 | |
*** savvas_ has joined #openstack-ansible | 05:36 | |
*** germs has quit IRC | 05:38 | |
*** threestrands has joined #openstack-ansible | 05:39 | |
bhujay | one silly question , i cant connect to gerrit using git review -s , it says could not connect to gerrit . any suggestion ? | 05:43 |
bhujay | bhujay@ansiblenew:~/gnocchi$ git config --global gitreview.username bhujay | 05:43 |
bhujay | bhujay@ansiblenew:~/gnocchi$ git review -s | 05:43 |
bhujay | Could not connect to gerrit. | 05:43 |
*** funkypenguin has joined #openstack-ansible | 05:48 | |
*** jwitko has quit IRC | 05:50 | |
*** funkypenguin has quit IRC | 05:50 | |
*** funkypenguin has joined #openstack-ansible | 05:52 | |
*** armaan has joined #openstack-ansible | 06:01 | |
*** armaan has quit IRC | 06:02 | |
*** funkypen_ has joined #openstack-ansible | 06:03 | |
gokhan | good morning folks. I installed OSA 16.0.4 with ceph following https://docs.openstack.org/project-deploy-guide/openstack-ansible/latest/app-config-prod-ceph.html doc. I have problems when adding images. then I see that ceph cluster health is warn. | 06:04 |
gokhan | http://paste.openstack.org/show/628409/ | 06:04 |
gokhan | I can not understand where I did wrong thing. I added http://paste.openstack.org/show/628408/ to user variables | 06:04 |
*** funkypenguin has quit IRC | 06:05 | |
gokhan | and have 1 osd host with 3 disks | 06:05 |
gokhan | I tried adding image but get CommunicationError: Error finding address for http://192.168.236.5:9292/v2/images/f9e8bbf4-dc7e-4b77-af62-1a508d853ac9/file: [Errno 32] Broken pipe | 06:05 |
*** savvas_ has quit IRC | 06:09 | |
*** indistylo has joined #openstack-ansible | 06:22 | |
*** threestrands has quit IRC | 06:28 | |
SamYaple | gokhan: you needto get ceph healthly first | 06:34 |
SamYaple | if you only have 1 osd host, you need to make sure your crush rule is replicating based on osd rather than the default (host) | 06:35 |
SamYaple | or fortesting, you can set the number of replicas to 1 | 06:35 |
*** bhujay has quit IRC | 06:37 | |
gokhan | SamYaple, yes I have only 1 osd host. ok I will try it | 06:41 |
*** savvas_ has joined #openstack-ansible | 06:50 | |
gokhan | SamYaple, I added ceph confs osd pool default size = 2 | 06:50 |
gokhan | osd crush chooseleaf type = 0 | 06:50 |
gokhan | and yhen How can I restart ceph services clearly on osa installation | 06:51 |
gokhan | sorry I am new to ceph :( | 06:51 |
*** ivveh has quit IRC | 06:54 | |
*** indistylo has quit IRC | 06:56 | |
*** pmannidi has quit IRC | 06:56 | |
*** indistylo has joined #openstack-ansible | 06:56 | |
-openstackstatus- NOTICE: Due to some unforseen Zuul issues the gate is under very high load and extremely unstable at the moment. This is likely to persist until PST morning | 07:01 | |
*** ChanServ changes topic to "Due to some unforseen Zuul issues the gate is under very high load and extremely unstable at the moment. This is likely to persist until PST morning" | 07:01 | |
*** electrocucaracha has quit IRC | 07:06 | |
*** alanmeadows has quit IRC | 07:06 | |
*** serverascode has quit IRC | 07:07 | |
*** alanmeadows has joined #openstack-ansible | 07:07 | |
*** electrocucaracha has joined #openstack-ansible | 07:07 | |
*** mugsie has quit IRC | 07:07 | |
*** bgmccollum has quit IRC | 07:07 | |
*** persia has quit IRC | 07:07 | |
*** serverascode has joined #openstack-ansible | 07:08 | |
*** aetaric_ has quit IRC | 07:08 | |
*** kencjohnston has quit IRC | 07:08 | |
*** odyssey4me has quit IRC | 07:08 | |
*** mugsie has joined #openstack-ansible | 07:08 | |
*** mugsie has quit IRC | 07:08 | |
*** mugsie has joined #openstack-ansible | 07:08 | |
*** persia has joined #openstack-ansible | 07:08 | |
*** aetaric has joined #openstack-ansible | 07:11 | |
cloudnull | yifei: all of the logs are shipped to the logging host. which will be in /openstack/$rsyslog_container_name/log-storage/ path | 07:11 |
yifei | cloudnull: got that. Thanks | 07:12 |
*** bgmccollum has joined #openstack-ansible | 07:12 | |
*** odyssey4me has joined #openstack-ansible | 07:12 | |
*** kencjohnston has joined #openstack-ansible | 07:13 | |
*** csmart has quit IRC | 07:14 | |
*** evrardjp[m] has quit IRC | 07:14 | |
*** john51 has quit IRC | 07:17 | |
*** armaan has joined #openstack-ansible | 07:17 | |
*** john51 has joined #openstack-ansible | 07:17 | |
*** ajmaidak has quit IRC | 07:17 | |
*** jamesdenton has quit IRC | 07:18 | |
*** ajmaidak has joined #openstack-ansible | 07:20 | |
*** armaan has quit IRC | 07:21 | |
*** savvas_ has quit IRC | 07:24 | |
evrardjp | cloudnull: do you have an idea on how we can improve repo venv build speed? | 07:32 |
evrardjp | I have an idea to improve the git cloning, but I have nothing for the venv building itself | 07:33 |
*** funkypen_ has quit IRC | 07:35 | |
*** nshetty is now known as nshetty|lunch | 07:38 | |
*** pmannidi has joined #openstack-ansible | 07:46 | |
*** savvas_ has joined #openstack-ansible | 07:47 | |
*** pmannidi has quit IRC | 07:51 | |
evrardjp | ok found something | 07:51 |
*** vnogin has joined #openstack-ansible | 07:54 | |
*** shardy has joined #openstack-ansible | 07:58 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set the default pip index dynamically https://review.openstack.org/525576 | 07:58 |
evrardjp | cloudnull: are you there? | 08:02 |
*** epalper has joined #openstack-ansible | 08:11 | |
*** hamza21 has joined #openstack-ansible | 08:17 | |
*** mrch has joined #openstack-ansible | 08:19 | |
*** savvas_ has quit IRC | 08:20 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set the default pip index dynamically https://review.openstack.org/525576 | 08:24 |
*** pmannidi has joined #openstack-ansible | 08:28 | |
*** shardy is now known as shardy_afk | 08:31 | |
*** adreznec has quit IRC | 08:32 | |
*** adreznec has joined #openstack-ansible | 08:32 | |
*** funkypenguin has joined #openstack-ansible | 08:33 | |
*** pmannidi has quit IRC | 08:40 | |
*** savvas_ has joined #openstack-ansible | 08:41 | |
*** armaan has joined #openstack-ansible | 08:45 | |
*** bhujay has joined #openstack-ansible | 08:47 | |
*** shardy_afk is now known as shardy | 08:48 | |
*** armaan has quit IRC | 08:53 | |
*** armaan has joined #openstack-ansible | 08:55 | |
*** zkynet has joined #openstack-ansible | 08:56 | |
*** armaan has quit IRC | 08:57 | |
openstackgerrit | Periyasamy Palanisamy proposed openstack/openstack-ansible master: correct ha proxy configuration for opendaylight https://review.openstack.org/526055 | 09:02 |
*** funkypenguin has quit IRC | 09:05 | |
*** nshetty|lunch is now known as nshetty | 09:08 | |
*** funkypenguin has joined #openstack-ansible | 09:11 | |
*** funkypenguin has quit IRC | 09:13 | |
*** jbadiapa has joined #openstack-ansible | 09:14 | |
*** savvas_ has quit IRC | 09:14 | |
*** evrardjp[m] has joined #openstack-ansible | 09:17 | |
*** cshen has quit IRC | 09:17 | |
*** DanyC has joined #openstack-ansible | 09:18 | |
*** zkynet has quit IRC | 09:19 | |
*** savvas_ has joined #openstack-ansible | 09:20 | |
*** pbandark has joined #openstack-ansible | 09:20 | |
*** cshen has joined #openstack-ansible | 09:21 | |
*** jbadiapa has quit IRC | 09:21 | |
*** savvas has quit IRC | 09:21 | |
*** savvas has joined #openstack-ansible | 09:25 | |
*** ccha has quit IRC | 09:26 | |
*** csmart has joined #openstack-ansible | 09:30 | |
*** yifei has quit IRC | 09:33 | |
*** funkypenguin has joined #openstack-ansible | 09:33 | |
*** funkypenguin has quit IRC | 09:37 | |
*** vnogin has quit IRC | 09:41 | |
*** vnogin has joined #openstack-ansible | 09:50 | |
*** zkynet has joined #openstack-ansible | 09:54 | |
*** hw_wutianwei has quit IRC | 09:57 | |
*** savvas has quit IRC | 09:59 | |
*** savvas has joined #openstack-ansible | 10:04 | |
*** vnogin has quit IRC | 10:09 | |
hwoarang | good morning | 10:10 |
hwoarang | evrardjp: i am curious about your rootwrap.d updates. Basically i was wondering why do we have our own copies of these files and not use what upstream has? | 10:11 |
evrardjp | this is the standard release bump | 10:11 |
evrardjp | this is a copy from upstream basically | 10:12 |
evrardjp | odyssey4me: is planning a refactor of that so that projects carry those files and those files would be pip installable on the target, so we can configure them in place | 10:12 |
hwoarang | yeah but what i am saying is that if they are just copies, why do we need them? can we not use the upstream file directly? | 10:12 |
evrardjp | you mean fetch from upstream at the beginning of the process, and distribute? | 10:13 |
hwoarang | these files are in the pip package we install for each openstack service right? | 10:14 |
evrardjp | not so sure | 10:14 |
evrardjp | or not the way we want for some | 10:15 |
hwoarang | ah ok | 10:15 |
evrardjp | that's what Jesse is fixing | 10:16 |
*** pedja has joined #openstack-ansible | 10:17 | |
*** vnogin has joined #openstack-ansible | 10:25 | |
*** zkynet has quit IRC | 10:30 | |
*** electrofelix has joined #openstack-ansible | 10:30 | |
*** zkynet has joined #openstack-ansible | 10:35 | |
*** savvas has quit IRC | 10:37 | |
*** zkynet has quit IRC | 10:47 | |
*** zkynet has joined #openstack-ansible | 10:54 | |
openstackgerrit | bhujay kumar proposed openstack/openstack-ansible-os_ceilometer master: Run ceilometer-upgrade task post ceiloemeter installation to create the resources in gnocchi' https://review.openstack.org/526655 | 10:58 |
bhujay | evarardjp: I was stuck up for one and half month to sort out issue with vm not getting ip and finally solved it yesterday. Back on mu autoscalling goal . Had to reinstall telemetry and discovered this issue with gnocchi . Hope the proposal is ok with you guys | 11:01 |
evrardjp | bhujay: vm not getting ip? not OSA related I hope? | 11:04 |
evrardjp | bhujay: if you're willing to help on those you may want to have a look at this: | 11:05 |
evrardjp | https://etherpad.openstack.org/p/osa-telemetry-stack-fixing | 11:05 |
bhujay | evrardjp: It was due to OSA layer3 routed config where I missed the address_prefix part and the vxlan traffic was going to vlan . So OSA is perfect but it was my mistake and lot of learning :) | 11:07 |
evrardjp | :) | 11:07 |
evrardjp | I am glad you found it out then | 11:07 |
*** zkynet has quit IRC | 11:14 | |
*** openstackgerrit has quit IRC | 11:17 | |
*** askb_ has quit IRC | 11:18 | |
*** DanyC_ has joined #openstack-ansible | 11:24 | |
*** chyka has joined #openstack-ansible | 11:26 | |
bhujay | sorry vxlan traffic was going to br-mgmt | 11:26 |
*** savvas has joined #openstack-ansible | 11:26 | |
*** DanyC has quit IRC | 11:27 | |
*** udesale has quit IRC | 11:29 | |
*** chyka has quit IRC | 11:30 | |
*** openstackgerrit has joined #openstack-ansible | 11:34 | |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-os_barbican master: Add support for the openSUSE Leap distributions https://review.openstack.org/522786 | 11:34 |
*** stuartgr has joined #openstack-ansible | 11:36 | |
*** stuartgr has left #openstack-ansible | 11:36 | |
*** stuartgr has joined #openstack-ansible | 11:37 | |
*** hamza21 has quit IRC | 11:41 | |
*** DanyC has joined #openstack-ansible | 11:44 | |
*** dave-mccowan has joined #openstack-ansible | 11:45 | |
*** DanyC_ has quit IRC | 11:47 | |
*** vnogin has quit IRC | 11:52 | |
*** savvas has quit IRC | 11:59 | |
hwoarang | evrardjp: about the usual swift error in AIO are you sure it's just a spurious error and not a real issue? | 12:06 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest master: Run tempest with its defaults https://review.openstack.org/526372 | 12:07 |
evrardjp | which one? | 12:07 |
evrardjp | I didn't have that on my machine IIRC | 12:07 |
evrardjp | I am testing right now, will tell you | 12:07 |
evrardjp | did you see the results on two runs of tempest there ^ | 12:08 |
evrardjp | maybe I was lucky, but that's definitely a whole less | 12:08 |
evrardjp | lot* | 12:08 |
evrardjp | I've also found a repo optimization | 12:08 |
evrardjp | in the repo we checkout the version, then we are git cloning this version to create the wheel to a different place. Instead we could probably directly use the folder as a file location, which would bypass git checks | 12:09 |
evrardjp | I am only earning 1 minute with that though | 12:10 |
evrardjp | salvation will come from baremetal | 12:11 |
hwoarang | hmm i will try locally too | 12:11 |
evrardjp | hwoarang: could you give me a link of the issue? | 12:11 |
hwoarang | evrardjp: http://logs.openstack.org/76/525576/9/check/openstack-ansible-deploy-aio-ubuntu-xenial/7fce27d/job-output.txt.gz#_2017-12-08_11_49_52_329884 | 12:12 |
evrardjp | maybe there is a sha update that should happen... | 12:12 |
evrardjp | we have tempest lagging behind, because the gates are broken | 12:12 |
hwoarang | that's the usual AIO+pip urls patchset | 12:12 |
hwoarang | yeah possibly | 12:12 |
evrardjp | isn't this a glance issue? | 12:13 |
evrardjp | Connection aborted.', error(32, 'Broken pipe') http://172.29.236.100:9292/v2/images/927592e1-acbe-4148-ac74-ffc8c40d038e | 12:14 |
*** vnogin has joined #openstack-ansible | 12:14 | |
evrardjp | I've seen it more than once lately | 12:14 |
hwoarang | ah yes sorry i was looking at the previous one | 12:15 |
hwoarang | ok | 12:15 |
evrardjp | I can't say for certain until I have seen the machine | 12:16 |
evrardjp | found an optimization on 525576 | 12:16 |
evrardjp | but I don't want to delay the gates | 12:16 |
evrardjp | so if it passes, good, else I will add a change | 12:16 |
evrardjp | slowly improving :p | 12:16 |
evrardjp | oh wait I thought I did a recheck | 12:18 |
evrardjp | ok let me fix that | 12:18 |
*** vnogin has quit IRC | 12:19 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set the default pip index dynamically https://review.openstack.org/525576 | 12:20 |
evrardjp | even the gates gods are against us: https://review.openstack.org/#/c/526050/1 | 12:24 |
evrardjp | to be honest I haven't done any monthly offering to infra. | 12:24 |
*** vnogin has joined #openstack-ansible | 12:26 | |
*** DanyC has quit IRC | 12:26 | |
*** vnogin has quit IRC | 12:27 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Update all SHAs https://review.openstack.org/522850 | 12:29 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Update variable scopes https://review.openstack.org/523468 | 12:30 |
mhayden | happy friday | 12:36 |
evrardjp | mhayden: unhappy friday to you too | 12:46 |
evrardjp | how is ^3con? | 12:46 |
evrardjp | talking about q80s a lot more than openstack I guess | 12:47 |
evrardjp | sorry: "k8s" | 12:47 |
*** vnogin has joined #openstack-ansible | 12:48 | |
mhayden | trolol | 12:48 |
mhayden | it was pretty good | 12:48 |
evrardjp | it's over? | 12:48 |
*** savvas has joined #openstack-ansible | 12:49 | |
evrardjp | I still thinks qbn80s was better than k8s. | 12:49 |
mhayden | it's still running today, but we got snow last night (turning to ice) and after driving to austin three days in a row, i'm not interested in doing it again ;) | 12:49 |
evrardjp | haha ok | 12:49 |
*** zkynet has joined #openstack-ansible | 12:49 | |
evrardjp | saw a lot of ppl there? Still openstack ppl there? | 12:49 |
openstackgerrit | Merged openstack/openstack-ansible master: Update overview-osa for new supported Operating Systems https://review.openstack.org/526460 | 12:51 |
mhayden | quite a few openstack people there actually | 12:57 |
mhayden | has some folks asking about how to work with ansible-hardening, and some folks wanted to merge that work with some of the mindpoint ansible roles | 12:58 |
*** pbandark has quit IRC | 13:01 | |
*** vnogin has quit IRC | 13:02 | |
evrardjp | hwoarang: broken on my local install | 13:02 |
evrardjp | mindpoint ansible roles? | 13:02 |
*** savvas_ has quit IRC | 13:07 | |
*** savvas_ has joined #openstack-ansible | 13:08 | |
evrardjp | logan-: hwoarang ahah FYI I found out that ansible git module doesn't do shallow git clones on any reference, so it doesn't respect the uploadpack.allowReachableSHA1InWant from the git server | 13:09 |
evrardjp | that could explain why we got larger git clones. | 13:09 |
evrardjp | so more issues :p | 13:10 |
evrardjp | github and gerrit support it, so I think we should use it | 13:10 |
evrardjp | oh no my bad, they don't | 13:13 |
evrardjp | darn | 13:13 |
*** epalper has quit IRC | 13:14 | |
evrardjp | good to know | 13:21 |
*** ppetit has joined #openstack-ansible | 13:21 | |
*** ppetit has quit IRC | 13:22 | |
*** savvas has quit IRC | 13:22 | |
*** ppetit has joined #openstack-ansible | 13:22 | |
evrardjp | hello ppetit :) | 13:23 |
ppetit | Hello evrardjp!!! | 13:23 |
evrardjp | how are you? | 13:23 |
ppetit | I am having a little issue with deploying OSA in AIO mode. Hoped you could give me sme handā¦. | 13:24 |
ppetit | When running openstack-ansible setup-infrastructure.yml I am getting an error no sure where how to fix it. | 13:25 |
ppetit | On repo_build | 13:25 |
evrardjp | ppetit: which version? | 13:26 |
evrardjp | Because I broke everything this week :p | 13:26 |
evrardjp | on master | 13:26 |
evrardjp | large changes before m2 | 13:26 |
evrardjp | the rest should be fine, if you're using head of the branch | 13:27 |
ppetit | 17.0.0.0b1 | 13:27 |
evrardjp | we got a problem with pyasn | 13:27 |
evrardjp | mmm | 13:27 |
evrardjp | yeah a milestone of master | 13:27 |
*** DanyC has joined #openstack-ansible | 13:27 | |
evrardjp | you need master absolutely? | 13:27 |
ppetit | No matching distribution found for pyasn1-modules>=0.1.5 (from python-ldap)", "stderr_lines": [" Could not find a version that satisfies the requirement pyasn1-modules>=0.1.5 (from python-ldap) (from versions: 0.1.4) | 13:27 |
evrardjp | yeah | 13:27 |
evrardjp | that's outside our will | 13:27 |
evrardjp | there was a bump of requirements outside what's bundled by upper constraints | 13:28 |
evrardjp | that happened this week too. | 13:28 |
evrardjp | Fun week. | 13:28 |
ppetit | I I donāt need master absolutely. | 13:28 |
*** hamza21 has joined #openstack-ansible | 13:28 | |
evrardjp | today I am tagging stable/pike with the fix | 13:28 |
evrardjp | but head of stable/pike already has the fix included | 13:29 |
*** dasTor has left #openstack-ansible | 13:29 | |
evrardjp | https://github.com/openstack/openstack-ansible/commit/6836c7f26e0b18d58c90bfc514b973a3d1f5fa69 + https://github.com/openstack/openstack-ansible-os_keystone/commit/ffc9c9b5e681748ff3e54e43f22c921e83342a51 | 13:29 |
evrardjp | if you wanna know the cause | 13:29 |
evrardjp | ppetit: you'll have to re-install though, we don't support rollback to a previous version | 13:30 |
evrardjp | you can simply destroy your containers and that should do the trick | 13:30 |
ppetit | good to know whu ;-) | 13:30 |
ppetit | why | 13:30 |
evrardjp | cd /opt/openstack-ansible/playbooks; openstack-ansible lxc-containers-destroy.yml | 13:30 |
evrardjp | so you don't have any containers remaining | 13:31 |
evrardjp | (or you can reroll your nodes, even better!) | 13:31 |
evrardjp | then you start the procedure again | 13:31 |
evrardjp | scripts/bootstrap-ansible.sh, etc... | 13:31 |
ppetit | okay Iāll give it a try | 13:31 |
ppetit | thanks! | 13:32 |
evrardjp | stable/ is supposed to be stable, but we can't controll what a dependency of python-ldap was doing :p | 13:32 |
*** DanyC has quit IRC | 13:32 | |
evrardjp | we have a technical solution in place for that: you can freeze your pip requirement. | 13:32 |
evrardjp | but in this case, it was already frozen by upper constraints, except the project bumped it, which kinda broke everything. | 13:33 |
*** savvas has joined #openstack-ansible | 13:35 | |
mgariepy | good morning everyone | 13:39 |
*** markvoelker has quit IRC | 13:40 | |
evrardjp | morning mgariepy | 13:42 |
*** markvoelker has joined #openstack-ansible | 13:43 | |
*** woodard has quit IRC | 13:45 | |
mgariepy | evrardjp, anything that need review ? | 13:45 |
*** woodard has joined #openstack-ansible | 13:45 | |
evrardjp | I've updated the review of yesterday | 13:46 |
evrardjp | still haven't merged therefore. And on top of it, gate issues, as you can see in the chan title | 13:46 |
evrardjp | the review is: https://review.openstack.org/#/c/525576/ | 13:47 |
evrardjp | sadly it doesn't work on my machine, tempest fails. | 13:47 |
*** zkynet has quit IRC | 13:48 | |
*** zkynet has joined #openstack-ansible | 13:51 | |
*** ArchiFleKs has quit IRC | 13:52 | |
mgariepy | evrardjp, is this patch fix this ? http://logs.openstack.org/20/523920/2/gate/openstack-ansible-deploy-aio-ubuntu-xenial/cfc35c0/job-output.txt.gz#_2017-12-05_12_00_00_753002 | 13:55 |
evrardjp | yes | 13:55 |
mgariepy | ok | 13:55 |
*** savvas__ has joined #openstack-ansible | 13:56 | |
*** savvas_ has quit IRC | 13:56 | |
*** ChanServ changes topic to "Launchpad: https://launchpad.net/openstack-ansible || Weekly Meetings: https://wiki.openstack.org/wiki/Meetings/openstack-ansible || Review Dashboard: https://goo.gl/tTmdgs" | 13:57 | |
-openstackstatus- NOTICE: The issues have been fixed, Zuul is operating fine again but has a large backlog. You can recheck jobs that failed. | 13:57 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/pike: Update all SHAs for 16.0.6 https://review.openstack.org/526682 | 13:59 |
*** savvas_ has joined #openstack-ansible | 14:00 | |
*** zkynet has quit IRC | 14:00 | |
*** zkynet has joined #openstack-ansible | 14:02 | |
*** savvas__ has quit IRC | 14:03 | |
*** savvas__ has joined #openstack-ansible | 14:03 | |
*** savvas_ has quit IRC | 14:07 | |
*** savvas has quit IRC | 14:08 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible stable/newton: Update all SHAs for 14.2.14 https://review.openstack.org/526684 | 14:08 |
*** DanyC has joined #openstack-ansible | 14:10 | |
evrardjp | Ocata and Queens missing in the release, I am waiting for https://review.openstack.org/#/c/526050/ for ocata and https://review.openstack.org/#/c/525576/ for queens. Will do a release with that when they merge. | 14:12 |
evrardjp | sadly the second would need attention, because it seems broken | 14:12 |
evrardjp | tempest | 14:13 |
evrardjp | (full transparency here!) | 14:13 |
*** savvas has joined #openstack-ansible | 14:18 | |
*** savvas__ has quit IRC | 14:21 | |
evrardjp | mhayden: are you available? | 14:21 |
*** nshetty has quit IRC | 14:22 | |
cloudnull | mornings | 14:23 |
*** vnogin has joined #openstack-ansible | 14:23 | |
*** pbandark has joined #openstack-ansible | 14:24 | |
*** esberglu has joined #openstack-ansible | 14:24 | |
evrardjp | morning cloudnull | 14:24 |
cloudnull | did you figure something out regarding venv build speed ? | 14:25 |
evrardjp | yeah | 14:25 |
evrardjp | we were using internal_lb_vip_address to point to the repo | 14:25 |
cloudnull | it was done via bash script and executed in parallel, is that not the case any longer? | 14:25 |
evrardjp | yes it is | 14:25 |
cloudnull | oh ok | 14:26 |
evrardjp | it's just the pip part was suboptimal | 14:26 |
cloudnull | ok | 14:26 |
evrardjp | the pip used the pip.conf which tried on internal_lb_vip_address and PyPI | 14:26 |
evrardjp | because the repo build process happens before haproxy that didn't work so well | 14:26 |
evrardjp | -> time loss | 14:26 |
evrardjp | there is another improvement we could have | 14:26 |
evrardjp | is to not git clone again when building wheels, because we generate a constraint file that contains git+file:/// | 14:27 |
cloudnull | I thought we were using that? | 14:27 |
evrardjp | if we omit git+ (and assume everything was properly checked out, which it should), we could shave a few seconds | 14:28 |
evrardjp | cloudnull: what do you mean? | 14:28 |
evrardjp | we are using git+file:/// in the constraints yes. But we could even move further to use file:/// instead, if we assume the checkouts are correct | 14:28 |
evrardjp | I will ask jesse what he thinks of that | 14:29 |
evrardjp | he was the one refactoring those things recently | 14:29 |
cloudnull | ok | 14:29 |
evrardjp | my concern right now is that https://review.openstack.org/#/c/525576/ doesn't pass tempest testing | 14:29 |
evrardjp | I tried with a fully updated shas, running all the openstack projects from master. | 14:30 |
evrardjp | That broke miserably on nova. | 14:30 |
evrardjp | api_db change | 14:30 |
cloudnull | :( | 14:31 |
evrardjp | well it's not nova's fault, it's one dependency. Anyway... Trying with other SHA. | 14:31 |
evrardjp | not so fresh | 14:31 |
evrardjp | :) | 14:31 |
evrardjp | let see what it gives. | 14:31 |
* cloudnull waits eagerly | 14:32 | |
*** cshen has quit IRC | 14:35 | |
*** cloudtoko has joined #openstack-ansible | 14:36 | |
mhayden | evrardjp: in a mtg for the next 24 mins | 14:36 |
evrardjp | ok | 14:37 |
evrardjp | thanks | 14:37 |
evrardjp | I would need your tempest experience | 14:37 |
evrardjp | :D | 14:37 |
* hwoarang is back | 14:37 | |
evrardjp | hwoarang: welcome back :) | 14:37 |
* mhayden has tempest experience? | 14:39 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-os_tempest master: Remove jinja2 delimiters from when https://review.openstack.org/524672 | 14:40 |
evrardjp | evidence just right there! | 14:40 |
evrardjp | ^ | 14:40 |
* mhayden has the rebase technology | 14:40 | |
* mhayden channels his inner cloudnull | 14:40 | |
cloudnull | hahahaha | 14:40 |
evrardjp | mhayden: what would be the impact of this channeling? | 14:45 |
evrardjp | less this more that? | 14:46 |
mhayden | not sure | 14:46 |
evrardjp | haha | 14:46 |
mhayden | i'll use a lot more "the's" in my sentences | 14:46 |
mhayden | so let's talk about the tempest | 14:46 |
evrardjp | haha | 14:46 |
evrardjp | the gate issues | 14:46 |
evrardjp | the failures! | 14:46 |
*** indistylo has quit IRC | 14:46 | |
mhayden | we don't have failures -- we have the best gate jobs | 14:47 |
mhayden | all the best gate jobs | 14:47 |
mhayden | the world will be in awe of our gate jobs | 14:47 |
mhayden | they will wish they had our gate jobs | 14:47 |
*** savvas_ has joined #openstack-ansible | 14:49 | |
hwoarang | true | 14:52 |
hwoarang | you should be proud of these jobs | 14:52 |
*** DanyC has quit IRC | 14:53 | |
* mhayden thinks that evrardjp's comments about our gate jobs are fake news | 14:53 | |
mgariepy | lol | 14:53 |
* mhayden had a good amount of coffee today | 14:55 | |
evrardjp | I am declaring mhayden as capital of coffee then. | 14:57 |
*** savvas__ has joined #openstack-ansible | 15:10 | |
*** savvas has quit IRC | 15:10 | |
mhayden | i'm considering putting that COPR package into another repo somewhere :/ | 15:12 |
mhayden | a reverse proxy in openstack land won't help much | 15:12 |
*** bhujay has quit IRC | 15:12 | |
mhayden | we can either 1) compile from source 2) build RPM from source or 3) put the rpms into the role (sounds terrible) | 15:12 |
mgariepy | can we do like: install from COPR if fail build rpm from source ? | 15:13 |
mhayden | possibly | 15:13 |
mgariepy | how much time would that add ? | 15:13 |
mhayden | i could just mirror the copr rpms elsewhere | 15:13 |
mhayden | perhaps in github pages | 15:13 |
mhayden | the rpm build itself is 1-2 minutes | 15:14 |
mgariepy | k | 15:14 |
mhayden | packagecloud might work, but they charge for bandwidth | 15:14 |
mhayden | we might not hit the limits though | 15:14 |
mgariepy | put your CC # on s3 :P | 15:14 |
mhayden | trolololol | 15:14 |
mhayden | well, open source at pkgcloud is 25GB storage + 250GB bandwidth | 15:15 |
mgariepy | so we are good ? | 15:15 |
mgariepy | use copr if fail, fallback to pkgcloud ? | 15:15 |
mhayden | possibly | 15:16 |
*** savvas has joined #openstack-ansible | 15:16 | |
mgariepy | can infra mirror the repos ? | 15:16 |
mhayden | well, i asked about it, but it's a lot of work for a few pkgs | 15:16 |
mgariepy | really? isn't there already a lot of repos mirrored ? the process should be quite well established by now ... | 15:17 |
mhayden | i asked the maintainer of the pkg to put a 2.0 LXC package into EPEL | 15:18 |
mhayden | which is hosted much more reliably, and mirrored | 15:18 |
mgariepy | noooo | 15:19 |
mgariepy | not epel. | 15:19 |
mgariepy | :P | 15:19 |
spotz | heheh | 15:19 |
mhayden | haha | 15:19 |
mhayden | i doubt he will do that since the epel package appears to be locked into 1.0 territory | 15:19 |
*** savvas__ has quit IRC | 15:20 | |
mgariepy | hmm where are the qemu/rdo/whatever else STIG hosted ? | 15:20 |
mhayden | those are hosed on centos' mirror infra | 15:20 |
mhayden | which is good | 15:20 |
*** savvas_ has quit IRC | 15:22 | |
mgariepy | can that fit in there : https://wiki.centos.org/SpecialInterestGroup/Virtualization | 15:24 |
mhayden | possibly | 15:25 |
* mhayden can ask | 15:25 | |
*** n3v3rm0r3r has quit IRC | 15:26 | |
evrardjp | mhayden: why not carrying it around | 15:27 |
mhayden | carry around the SRPM? | 15:27 |
mhayden | we can totally do that, but we will need to install some extra deps to build the rpm | 15:27 |
evrardjp | nah the compiled thing | 15:27 |
mhayden | 1.3MB to carry around | 15:27 |
evrardjp | or ask the project openstack packaging | 15:27 |
mhayden | for the SRPM | 15:27 |
evrardjp | rpm-packaging | 15:28 |
evrardjp | this way they build it, and we fetch it as an artifact from upstream | 15:28 |
evrardjp | like we do for many other things like cirros | 15:28 |
mhayden | true | 15:28 |
*** vnogin has quit IRC | 15:28 | |
mhayden | to be honest, the root of the problem here is that COPR is fairly unreliable since it isn't mirrored | 15:28 |
evrardjp | I AM BECOMING INSANE | 15:28 |
evrardjp | tempest just passed. | 15:29 |
evrardjp | on my machine | 15:29 |
mhayden | i'm wondering if i can just make a github pages repo and put the repo there | 15:29 |
* mhayden hasn't tried that | 15:29 | |
evrardjp | that sounds worse than publishing on openstack | 15:31 |
evrardjp | but I have a better idea | 15:31 |
evrardjp | close to that | 15:31 |
evrardjp | you have a github repo | 15:31 |
evrardjp | that has a zuul job to generate a package | 15:31 |
evrardjp | publish it as a release artifact | 15:31 |
mhayden | true | 15:32 |
mgariepy | wouldn't be better if it was direcly under the lxc project? | 15:32 |
mhayden | if the LXC bits were in the repo? | 15:33 |
mhayden | err rpm bits? | 15:33 |
evrardjp | worth sharing with them but I doubt they have the pipelines | 15:34 |
mgariepy | nah, if lxc repo was generating the articaft | 15:34 |
mhayden | that would be handy | 15:34 |
mgariepy | nah, if lxc repo was generating the articact | 15:34 |
evrardjp | mhayden: mirror it on icanhazip :p | 15:35 |
*** vnogin has joined #openstack-ansible | 15:35 | |
mhayden | oh geez | 15:35 |
evrardjp | lol | 15:35 |
evrardjp | let's use that bandwidth for a reason! | 15:36 |
evrardjp | you can still have the ips of ppl with it, it's good enough | 15:36 |
evrardjp | :p | 15:36 |
evrardjp | oh you want to return the IPs to them? | 15:36 |
evrardjp | :p | 15:36 |
evrardjp | How funny I am, I am marvelled sometimes. | 15:36 |
evrardjp | hwoarang: are you there? | 15:37 |
evrardjp | so I tried with all the projects master that miserably failed | 15:37 |
hwoarang | evrardjp: yes | 15:38 |
evrardjp | but with https://review.openstack.org/#/c/522850/ it passes | 15:38 |
evrardjp | as you can see, zuul passes there too! | 15:38 |
mgariepy | wand me to +w ? | 15:38 |
hwoarang | :/ | 15:39 |
evrardjp | well I don't know yet | 15:39 |
evrardjp | I guess we'll have to merge those two patches together? | 15:39 |
evrardjp | let's +w them both | 15:40 |
evrardjp | wow I didn't notice the time for 522850. 1h30 for ceph that's quite a record we haven't beaten for a while. | 15:41 |
evrardjp | guys what should I do? Mega patch, or removing voting temporarily? | 15:42 |
evrardjp | I'd be enclined to remove testing on a patch, then immediately restore it. | 15:42 |
evrardjp | but it seems dangerous | 15:42 |
evrardjp | mega patch? | 15:43 |
*** savvas_ has joined #openstack-ansible | 15:43 | |
evrardjp | ok let me do the mega patch approach | 15:44 |
mgariepy | nice consensus. | 15:44 |
mgariepy | haha ;p | 15:44 |
evrardjp | yeah nobody was saying anything | 15:44 |
evrardjp | I am still stressed about those deadlines! | 15:44 |
*** savvas__ has joined #openstack-ansible | 15:45 | |
mgariepy | well megapatch seems better than removing testing | 15:45 |
*** savvas has quit IRC | 15:45 | |
*** shardy has quit IRC | 15:47 | |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible master: Set the default pip index dynamically https://review.openstack.org/525576 | 15:47 |
evrardjp | MEGA PATCH | 15:48 |
evrardjp | (you have to say it like if it was a decepticon talking) | 15:48 |
*** mrch has quit IRC | 15:49 | |
evrardjp | rollout! https://slack-imgs.com/?c=1&url=http%3A%2F%2Friffsy.com%2Fimage%2FtTvt_tinygif.gif | 15:49 |
*** cloudtoko has quit IRC | 15:50 | |
*** cloudtoko has joined #openstack-ansible | 15:50 | |
*** savvas has joined #openstack-ansible | 15:53 | |
hwoarang | lets do it | 15:53 |
openstackgerrit | Jean-Philippe Evrard proposed openstack/openstack-ansible-os_tempest master: Run tempest with its defaults https://review.openstack.org/526372 | 15:53 |
*** savvas_ has quit IRC | 15:56 | |
*** zkynet has quit IRC | 15:58 | |
openstackgerrit | Merged openstack/openstack-ansible-os_magnum master: Clarify cinder integration documentation https://review.openstack.org/526403 | 16:01 |
*** savvas__ has quit IRC | 16:04 | |
evrardjp | spotz: you want to hear a funny story? Your patch is still in the "post" queue. :/ | 16:05 |
*** vnogin has quit IRC | 16:09 | |
spotz | evrardjp: I wonder if it hit when they were having the zuul issues | 16:10 |
*** DanyC has joined #openstack-ansible | 16:10 | |
evrardjp | or maybe the post jobs ARE the issue, and we triggered it :evil: | 16:10 |
evrardjp | I don't know, but we'll need another doc change if this would be broken/cancelled | 16:10 |
*** DanyC has quit IRC | 16:11 | |
*** DanyC has joined #openstack-ansible | 16:14 | |
evrardjp | ok not related to OSA directly, but is there someone here specialized in LIO fabric modules, especially the tcm_loop or vhost? | 16:14 |
*** Brew has joined #openstack-ansible | 16:15 | |
*** Brew has quit IRC | 16:15 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Remove LXC yum repo from containers https://review.openstack.org/526730 | 16:20 |
mhayden | evrardjp: should help gates a bit ^^ | 16:21 |
evrardjp | yeah | 16:21 |
evrardjp | \o/ | 16:21 |
evrardjp | you could differently to achieve the same goal mhayden | 16:21 |
mhayden | ORLY? | 16:22 |
evrardjp | you could have listed what you wanted to remove as a variable on all_containers | 16:22 |
Bico_Fino | Hello all! | 16:22 |
evrardjp | but let's keep it that way | 16:22 |
evrardjp | I prefer that way for now | 16:22 |
*** MasterOfBugs has joined #openstack-ansible | 16:22 | |
evrardjp | let's cleanup the lxc container create | 16:22 |
*** pramodrj07 has joined #openstack-ansible | 16:22 | |
mhayden | ohai Bico_Fino | 16:22 |
*** zkynet has joined #openstack-ansible | 16:22 | |
evrardjp | or more: let's not add more variables :) | 16:22 |
Bico_Fino | How do I populate /etc/hosts all over again? | 16:22 |
evrardjp | openstack-hosts role | 16:23 |
evrardjp | openstack_hosts* | 16:23 |
evrardjp | Bico_Fino: ^ | 16:23 |
Bico_Fino | For some reason I got duplicate hosts. o_I | 16:23 |
evrardjp | Bico_Fino: openstack-hosts-setup.yml | 16:23 |
evrardjp | that's the name of the playbook | 16:24 |
evrardjp | mmm | 16:24 |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Remove LXC yum repo from containers https://review.openstack.org/526730 | 16:24 |
evrardjp | is your openstack_user_config fine? | 16:24 |
*** savvas_ has joined #openstack-ansible | 16:25 | |
Bico_Fino | evrardjp Yes. | 16:25 |
Bico_Fino | I remove the factors | 16:25 |
Bico_Fino | removed* | 16:25 |
*** savvas_ has quit IRC | 16:26 | |
Bico_Fino | but if I removed the factors, I had to clean /etc/hosts as well? | 16:26 |
*** savvas_ has joined #openstack-ansible | 16:27 | |
evrardjp | yeah facts are an ansible thing | 16:27 |
evrardjp | Bico_Fino: I am more concerned about duplicated | 16:27 |
evrardjp | duplicates* | 16:27 |
evrardjp | probably a hidden issue in your inventory | 16:27 |
evrardjp | you should ask for more help here, but I have to go. | 16:27 |
evrardjp | because I have to go* | 16:27 |
Bico_Fino | running the playbook | 16:27 |
Bico_Fino | I discovered it because I was logging into rabbitmq container and ended up on heat one. haha | 16:28 |
*** Brew has joined #openstack-ansible | 16:29 | |
mhayden | evrardjp / hwoarang: commented | 16:35 |
hwoarang | thanks! | 16:37 |
mhayden | hwoarang: thankyousir | 16:37 |
*** bhujay has joined #openstack-ansible | 16:38 | |
evrardjp | answered | 16:39 |
evrardjp | I am gone | 16:39 |
evrardjp | ! | 16:39 |
evrardjp | see you laterz | 16:39 |
mhayden | thanks! | 16:41 |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Remove LXC yum repo from containers https://review.openstack.org/526730 | 16:42 |
*** bhujay has quit IRC | 16:49 | |
*** cloudtoko has quit IRC | 16:50 | |
*** cloudtoko has joined #openstack-ansible | 16:50 | |
*** bhujay has joined #openstack-ansible | 16:53 | |
*** chyka has joined #openstack-ansible | 17:00 | |
*** aludwar1 has joined #openstack-ansible | 17:02 | |
*** aludwar1 has quit IRC | 17:03 | |
*** savvas__ has joined #openstack-ansible | 17:07 | |
openstackgerrit | Merged openstack/openstack-ansible stable/ocata: Bump roles SHA https://review.openstack.org/526050 | 17:08 |
*** vnogin has joined #openstack-ansible | 17:09 | |
*** savvas_ has quit IRC | 17:11 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: [TEST] Try making a local LXC RPM repo https://review.openstack.org/526739 | 17:11 |
*** vnogin has quit IRC | 17:14 | |
*** zkynet has quit IRC | 17:15 | |
*** stuartgr has left #openstack-ansible | 17:20 | |
openstackgerrit | Markos Chandras (hwoarang) proposed openstack/openstack-ansible-tests master: test-log-collect.sh: Collect package and repo information from containers https://review.openstack.org/526741 | 17:22 |
*** DanyC has quit IRC | 17:28 | |
*** DanyC has joined #openstack-ansible | 17:29 | |
*** DanyC has quit IRC | 17:33 | |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: [TEST] Try making a local LXC RPM repo https://review.openstack.org/526739 | 17:35 |
*** germs has joined #openstack-ansible | 17:36 | |
*** germs has joined #openstack-ansible | 17:37 | |
*** germs1 has joined #openstack-ansible | 17:39 | |
*** bhujay has quit IRC | 17:39 | |
*** germs has quit IRC | 17:42 | |
*** smatzek has joined #openstack-ansible | 17:47 | |
*** cloudtoko has quit IRC | 17:51 | |
*** vnogin has joined #openstack-ansible | 17:51 | |
*** cloudtoko has joined #openstack-ansible | 17:51 | |
*** ppetit has quit IRC | 17:57 | |
*** openstackgerrit has quit IRC | 18:03 | |
*** vnogin has quit IRC | 18:14 | |
*** openstackgerrit has joined #openstack-ansible | 18:19 | |
openstackgerrit | Merged openstack/openstack-ansible-os_heat master: Update static files https://review.openstack.org/526306 | 18:19 |
*** electrofelix has quit IRC | 18:28 | |
*** pbandark has quit IRC | 18:30 | |
*** openstacking_123 has joined #openstack-ansible | 18:33 | |
openstacking_123 | There any bugs related to pike and neutron dvr? seems like my routers are dying over night | 18:35 |
*** cmart has quit IRC | 18:37 | |
*** funkypenguin has joined #openstack-ansible | 18:44 | |
openstackgerrit | Merged openstack/openstack-ansible-tests master: Skip failures when collecting nic setups https://review.openstack.org/526515 | 18:45 |
openstackgerrit | Merged openstack/openstack-ansible-tests master: Use Ansible 2.4 https://review.openstack.org/525518 | 18:45 |
openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Remove LXC yum repo from containers https://review.openstack.org/526730 | 18:47 |
*** cmart has joined #openstack-ansible | 18:50 | |
*** funkypenguin has quit IRC | 18:52 | |
*** savvas__ has quit IRC | 18:53 | |
*** savvas_ has joined #openstack-ansible | 18:53 | |
mgariepy | mhayden, reposync ? how does it work ? | 18:55 |
mgariepy | sync upstream repo locally ? | 18:55 |
mhayden | mgariepy: it drags down the contents of the repo (just rpms) | 18:55 |
mhayden | and createrepo turns that into something yum can use | 18:55 |
mhayden | oh crap it worked | 18:55 |
mgariepy | will it helps with a copr service that is down ? | 18:55 |
mhayden | well, what we can do is tell ansible to *try* to reposync | 18:56 |
mhayden | if it's an initial deployment, then we're still hosed | 18:56 |
mhayden | but if it's an existing, it should improve things | 18:56 |
mhayden | it should speed up the gates a bit since it's not checking against that slow repo | 18:56 |
mhayden | i'm still trying to find a better home for those packages | 18:56 |
mgariepy | i'm poking stgrabber to see if he would want to build the pkg in github | 18:57 |
mgariepy | but i don't think he will want to ;) haha | 18:57 |
mhayden | the COPR pkgs have some patches applied | 18:57 |
mgariepy | ha, | 18:57 |
*** savvas_ has quit IRC | 18:58 | |
*** vnogin has joined #openstack-ansible | 18:58 | |
mgariepy | reposync from copr if fail from icanhazip ? :P | 18:58 |
*** vnogin has quit IRC | 18:58 | |
mhayden | trololol | 18:58 |
mhayden | reposync is also really fast | 18:58 |
mhayden | < 2s to sync the repo down | 18:58 |
mgariepy | wow. | 18:58 |
*** vnogin has joined #openstack-ansible | 19:11 | |
*** pramodrj07 has quit IRC | 19:13 | |
*** MasterOfBugs has quit IRC | 19:13 | |
mgariepy | mhayden, any idea why repo container restart doesn't work on centos ? | 19:14 |
mhayden | not sure i've seen that | 19:14 |
mhayden | what's the issue? | 19:14 |
openstacking_123 | It stuck on networking? | 19:16 |
mgariepy | http://logs.openstack.org/76/525576/11/check/openstack-ansible-deploy-aio-centos-7/40ef536/job-output.txt.gz#_2017-12-08_17_54_07_986194 | 19:18 |
openstacking_123 | Can you try to stop and then start ? lxc-stop --name aio1_repo_container-5c38af3f , lxc-start --name aio1_repo_container-5c38af3f | 19:22 |
mgariepy | https://github.com/lxc/lxc/issues/1928 | 19:23 |
mgariepy | seems like the stop does work properly.. | 19:24 |
openstacking_123 | Good find | 19:25 |
*** MasterOfBugs has joined #openstack-ansible | 19:27 | |
*** pramodrj07 has joined #openstack-ansible | 19:27 | |
*** vnogin has quit IRC | 19:28 | |
*** funkypenguin has joined #openstack-ansible | 19:33 | |
*** vnogin has joined #openstack-ansible | 19:34 | |
evrardjp | mhayden: could you vote on https://review.openstack.org/#/c/525576/ ? | 19:35 |
evrardjp | finally passing | 19:35 |
evrardjp | this would unblock gates | 19:35 |
evrardjp | then i can issue a release | 19:35 |
mhayden | looking | 19:35 |
mhayden | ah, i looked that one over earlier but was waiting on gates | 19:35 |
evrardjp | thanks | 19:35 |
*** cmart has quit IRC | 19:36 | |
evrardjp | see you in 3 hours I guess. Hoping for something different than a recheck. | 19:36 |
evrardjp | Praying the gate lords for a good instance | 19:36 |
evrardjp | then ansible 2.4 and bare metal! | 19:37 |
evrardjp | wooot | 19:37 |
openstackgerrit | Marc GariƩpy (mgariepy) proposed openstack/openstack-ansible-lxc_hosts master: Add systemd-sysv to be able to stop correctly https://review.openstack.org/526764 | 19:38 |
mgariepy | mhayden, ^^ do you have systemd-sysv in your lxc containers ? | 19:38 |
*** vnogin has quit IRC | 19:38 | |
mhayden | i don't think so | 19:38 |
mgariepy | without it lxc-stop -n waits 30 or 60 sec and then kill the pid. | 19:39 |
mhayden | i don't have it | 19:39 |
mhayden | ORLY | 19:39 |
mgariepy | which version of lxc ? | 19:39 |
mhayden | is that why my containers take a long time to shut off? | 19:39 |
mhayden | 2.0.9-1 | 19:39 |
evrardjp | have we found a way to speed up centos? | 19:40 |
evrardjp | :D | 19:40 |
mgariepy | install the pkg, restart them then it shoudl be pretty quick. | 19:40 |
mgariepy | no, a way to fix a bug haha | 19:40 |
mgariepy | that wasn't there before .. lol | 19:40 |
evrardjp | do you have a trick for the speed up of package install? :p | 19:40 |
evrardjp | Haha | 19:40 |
evrardjp | I see. | 19:40 |
mgariepy | would be nice if we can save 60sec per container haha | 19:41 |
evrardjp | ok going back to normal life for a while | 19:41 |
mgariepy | evrardjp, http://logs.openstack.org/76/525576/11/check/openstack-ansible-deploy-aio-centos-7/40ef536/job-output.txt.gz#_2017-12-08_17_54_07_986194 | 19:41 |
evrardjp | loading | 19:41 |
funkypenguin | morning folks :) | 19:42 |
evrardjp | oh it's on 525576... | 19:42 |
funkypenguin | I solved my slow cloning problem, but I'm stuck on what may be a related issue | 19:42 |
evrardjp | funkypenguin: good night :) | 19:42 |
mgariepy | evrardjp, i've seen it at a lot of place.. | 19:42 |
evrardjp | funkypenguin: what was it? | 19:42 |
evrardjp | mgariepy: on centos I guess? | 19:43 |
funkypenguin | evrardjp - speed issue was a network fault | 19:43 |
funkypenguin | class-ofservice related | 19:43 |
evrardjp | mgariepy: I mean it's purely centos, not suse? | 19:43 |
evrardjp | funkypenguin: I see | 19:43 |
funkypenguin | somehow git cloning makes the issue more obvious | 19:43 |
funkypenguin | http download - 600kb/s | 19:44 |
funkypenguin | git clone : 75kb/s | 19:44 |
funkypenguin | now that it's fixed.. 2MB/s | 19:44 |
mgariepy | http://paste.openstack.org/show/628478/ | 19:44 |
mgariepy | does suse have the same issue ? | 19:44 |
funkypenguin | so, moving onto the next hurdle, galera playbook is failing on this: | 19:44 |
evrardjp | mgariepy: I don't know | 19:45 |
funkypenguin | FAILED - RETRYING: TASK: galera_server : Download the percona-xtrabackup package (1 retries left). | 19:45 |
funkypenguin | fatal: [nbs-dh-14_galera_container-e9a15f4d]: FAILED! => {"attempts": 3, "changed": false, "failed": true, "msg": "Failed to validate the SSL certificate for www.percona.com:443. Make sure your managed systems have a valid CA certificate installed. You can use validate_certs=False if you do not need to confirm the servers identity but this is unsafe and not recommended. Paths checked for this platform: /etc/ssl/certs, /etc/pki/ | 19:45 |
funkypenguin | ca-trust/extracted/pem, /etc/pki/tls/certs, /usr/share/ca-certificates/cacert.org, /etc/ansible"} | 19:45 |
evrardjp | but I never heard of it | 19:45 |
*** woodard has quit IRC | 19:45 | |
evrardjp | anyway I am off for today. mgariepy if you want, you can add the package to the package list to install in the container | 19:46 |
mgariepy | https://review.openstack.org/#/c/526764/ | 19:46 |
mgariepy | LO | 19:46 |
evrardjp | adding it to the base applies everywhere | 19:46 |
mgariepy | LOL | 19:46 |
mgariepy | but i thinkg i;ll push a patch to lxc for that. | 19:46 |
mgariepy | but not this week ;) haha | 19:47 |
mgariepy | have a nice weekend everyone. | 19:47 |
*** cmart has joined #openstack-ansible | 19:49 | |
funkypenguin | hmm | 19:51 |
funkypenguin | that error only occurs on 1/3 containers | 19:51 |
funkypenguin | maybe false alarm, let me whack the host with my reboot stick, and try again... | 19:51 |
mhayden | mgariepy: were you testing those containers via lxc tools? or using "reboot" inside the containers? | 19:53 |
mgariepy | lxc tools | 19:54 |
mgariepy | lxc-stop | 19:54 |
mhayden | hmm, i don't see that delay here :/ | 19:56 |
*** woodard has joined #openstack-ansible | 19:56 | |
mgariepy | maybe it's new-ish ? | 19:56 |
mgariepy | fresh install does the same on gate and my vm | 19:57 |
mgariepy | maybe the pkg set of the template has changed ? | 19:57 |
mgariepy | mhayden, https://github.com/lxc/lxc/issues/1928 << this explains the delay | 19:59 |
*** funkypenguin has quit IRC | 19:59 | |
mgariepy | mhayden, but it wasn't alway like that, so maybe something changed somewhere. | 20:00 |
*** gouthamr has joined #openstack-ansible | 20:04 | |
mgariepy | pike isn't delayed :S http://logs.openstack.org/82/526682/1/check/openstack-ansible-deploy-aio-centos-7/0ce95ab/job-output.txt.gz#_2017-12-08_15_53_20_563757 | 20:04 |
*** savvas has quit IRC | 20:05 | |
mgariepy | master is : http://logs.openstack.org/50/522850/10/check/openstack-ansible-deploy-aio-centos-7/f8c4f70/job-output.txt.gz#_2017-12-08_13_07_43_089544 | 20:05 |
mgariepy | mhayden, ^^ | 20:06 |
*** savvas has joined #openstack-ansible | 20:06 | |
mgariepy | anyway have a nice weekend. | 20:06 |
*** woodard has quit IRC | 20:09 | |
openstackgerrit | Merged openstack/openstack-ansible-os_tempest master: Run tempest with its defaults https://review.openstack.org/526372 | 20:09 |
*** funkypenguin has joined #openstack-ansible | 20:11 | |
*** funkypenguin has quit IRC | 20:15 | |
*** ChanServ has quit IRC | 20:17 | |
captaindave | Hey all. I'm running the repo-install.yml and I'm getting a os_keystone venv build failure of 'No matching distribution found for pyasn1-modules>=0.1.5 (from python-ldap)'. Any thoughs on why that might be happening? | 20:20 |
*** cmart has quit IRC | 20:22 | |
*** ChanServ has joined #openstack-ansible | 20:24 | |
*** barjavel.freenode.net sets mode: +o ChanServ | 20:24 | |
*** funkypenguin has joined #openstack-ansible | 20:27 | |
*** funkypenguin has quit IRC | 20:31 | |
*** openstacking_123 has quit IRC | 20:32 | |
mhayden | mgariepy: oh | 20:36 |
mhayden | i see what you mean | 20:37 |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts stable/pike: Remove LXC yum repo from containers https://review.openstack.org/526768 | 20:38 |
openstackgerrit | Major Hayden proposed openstack/openstack-ansible-lxc_hosts master: Create and use a local LXC RPM repo https://review.openstack.org/526739 | 20:41 |
*** dgonzalez has left #openstack-ansible | 20:46 | |
captaindave | Ah. New python-ldap. | 20:48 |
*** cmart has joined #openstack-ansible | 20:50 | |
SamYaple | does OSA run glance behind apache or nginx? with ot without uwsgi? | 20:51 |
*** funkypenguin has joined #openstack-ansible | 20:57 | |
*** gouthamr has quit IRC | 21:07 | |
openstackgerrit | Kevin Carter (cloudnull) proposed openstack/openstack-ansible-tests stable/pike: Skip failures when collecting nic setups https://review.openstack.org/526775 | 21:07 |
cloudnull | SamYaple: we use uWSGI by default | 21:08 |
*** cloudtoko has quit IRC | 21:08 | |
SamYaple | cloudnull: uwsgi in http mode? or socket with nginx/apache proxy | 21:09 |
*** cloudtoko has joined #openstack-ansible | 21:09 | |
cloudnull | I want to say http mode. | 21:09 |
* cloudnull going to look | 21:09 | |
SamYaple | :) thanks | 21:09 |
SamYaple | im aksing because im having some issues with it and want to source the communities advice | 21:10 |
SamYaple | OSA is typically forfront in these areas | 21:10 |
cloudnull | https://github.com/openstack/openstack-ansible-os_glance/blob/master/templates/glance-uwsgi.ini.j2 | 21:10 |
cloudnull | no NGINX yet | 21:10 |
SamYaple | got it. ok | 21:10 |
cloudnull | I know there was some work to get a cluster of servers running nginx setup w/ router fast + uwsgi however I'm not sure what the state of that is | 21:11 |
SamYaple | cool | 21:11 |
cloudnull | I want to say logan- andymccr were working on that ? | 21:11 |
SamYaple | yea im just having issues with uploading a 2gb image, i have to set the haproxy timeout server 10m to make it work | 21:12 |
SamYaple | which is....no bueno | 21:12 |
cloudnull | that was the spec - https://review.openstack.org/#/c/524812/ - seems like its still being worked on | 21:12 |
cloudnull | yea, that's a bummer. | 21:13 |
cloudnull | is hap severing the connection ? | 21:13 |
cloudnull | sorry it was jmccrory with the spec :D | 21:14 |
SamYaple | well nginx starts returning 499 and that makes haproxy mark the backend down | 21:14 |
SamYaple | thats the route cause there | 21:14 |
SamYaple | but it only does 499 while the image is uploading | 21:15 |
SamYaple | uwsgi is logging | 21:15 |
SamYaple | [pid: 8|app: 0|req: 416/416] 10.1.6.64 () {28 vars in 289 bytes} [Fri Dec 8 20:58:10 2017] OPTIONS / => generated 0 bytes in 0 msecs (HTTP/1.0 200) 2 headers in 0 bytes (0 switches on core 0) | 21:15 |
SamYaple | Fri Dec 8 20:58:10 2017 - uwsgi_response_write_headers_do(): Broken pipe [core/writer.c line 238] during OPTIONS / (10.1.6.64) | 21:15 |
SamYaple | and thats why nginx is 499'ing, but i mind you this all works.... in time | 21:16 |
SamYaple | so its weird | 21:16 |
cloudnull | 499!? | 21:16 |
cloudnull | isn't that a client side timeout ? | 21:16 |
SamYaple | custom nginx code | 21:16 |
SamYaple | oh yea, i see client closed rquest | 21:16 |
cloudnull | that's wierd | 21:17 |
cloudnull | **weird | 21:17 |
SamYaple | so nginx isthe client here, if uwsgi isnt responding faster enough/at all then nginx would close request | 21:17 |
SamYaple | that part makes sense | 21:17 |
SamYaple | hmmm.... thats something to go on at least | 21:17 |
SamYaple | i am running single thread/process everywhere. let me bump those up | 21:18 |
SamYaple | one thread could be locked writting to ceph or something | 21:18 |
cloudnull | yea that can do it, | 21:18 |
SamYaple | thanks man. atleast an ave to explore | 21:19 |
cloudnull | np. | 21:19 |
cloudnull | I'm heading off. | 21:19 |
cloudnull | take care all ! | 21:19 |
SamYaple | take care man | 21:19 |
cloudnull | enjoy the weekend :) | 21:19 |
*** dave-mccowan has quit IRC | 21:25 | |
*** smatzek has quit IRC | 21:27 | |
*** hamza21 has quit IRC | 21:30 | |
*** funkypenguin has quit IRC | 21:31 | |
SamYaple | ok moar threads/processes fixes the 499 | 21:39 |
SamYaple | i see theimage write fully out to ceph, correct size anddata | 21:39 |
SamYaple | but it never snaps/protects it | 21:39 |
*** smatzek has joined #openstack-ansible | 21:57 | |
SamYaple | weeeiirddd. first 440 bytes, it writes out fine. byte 441 is different | 21:59 |
*** smatzek has quit IRC | 22:01 | |
*** cmart has quit IRC | 22:02 | |
*** cmart has joined #openstack-ansible | 22:03 | |
*** vnogin has joined #openstack-ansible | 22:03 | |
*** cmart has quit IRC | 22:03 | |
*** sxc731 has joined #openstack-ansible | 22:05 | |
*** sxc731 has quit IRC | 22:11 | |
*** cmart has joined #openstack-ansible | 22:12 | |
*** gouthamr has joined #openstack-ansible | 22:12 | |
*** ianychoi_ has joined #openstack-ansible | 22:17 | |
*** ianychoi has quit IRC | 22:19 | |
*** indistylo has joined #openstack-ansible | 22:23 | |
*** askb_ has joined #openstack-ansible | 22:24 | |
*** askb_ has quit IRC | 22:29 | |
*** indistylo has quit IRC | 22:30 | |
*** askb_ has joined #openstack-ansible | 22:31 | |
*** DanyC has joined #openstack-ansible | 22:35 | |
*** Brew has quit IRC | 22:41 | |
*** askb_ has quit IRC | 23:02 | |
*** askb_ has joined #openstack-ansible | 23:03 | |
*** esberglu has quit IRC | 23:10 | |
*** germs1 has quit IRC | 23:34 | |
*** japestinho has quit IRC | 23:35 | |
*** vnogin has quit IRC | 23:37 | |
*** vnogin has joined #openstack-ansible | 23:38 | |
*** DanyC has quit IRC | 23:43 | |
*** DanyC has joined #openstack-ansible | 23:43 | |
*** engel75 has quit IRC | 23:45 | |
*** DanyC has quit IRC | 23:47 | |
*** markvoelker has quit IRC | 23:49 | |
*** markvoelker has joined #openstack-ansible | 23:50 | |
*** markvoelker has quit IRC | 23:54 | |
*** vnogin has quit IRC | 23:57 | |
*** cmart has quit IRC | 23:59 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!