| *** tosky has quit IRC | 00:01 | |
| *** macza has quit IRC | 00:23 | |
| *** weezS has quit IRC | 00:44 | |
| *** jawad_axd has joined #openstack-ansible | 00:47 | |
| *** jawad_axd has quit IRC | 00:51 | |
| *** dave-mccowan has joined #openstack-ansible | 01:15 | |
| *** ansmith has joined #openstack-ansible | 01:38 | |
| *** markvoelker has quit IRC | 01:41 | |
| *** hwoarang has quit IRC | 01:48 | |
| *** jawad_axd has joined #openstack-ansible | 01:49 | |
| *** ansmith has quit IRC | 01:52 | |
| *** jawad_axd has quit IRC | 01:53 | |
| *** hwoarang has joined #openstack-ansible | 01:53 | |
| *** cshen has quit IRC | 02:09 | |
| *** jawad_axd has joined #openstack-ansible | 02:51 | |
| *** jawad_axd has quit IRC | 02:55 | |
| *** dave-mccowan has quit IRC | 03:52 | |
| *** hwoarang has quit IRC | 03:56 | |
| *** hwoarang has joined #openstack-ansible | 04:03 | |
| *** udesale has joined #openstack-ansible | 04:17 | |
| *** dhellmann has quit IRC | 05:19 | |
| *** dhellmann has joined #openstack-ansible | 05:20 | |
| *** radeks has joined #openstack-ansible | 05:20 | |
| *** cshen has joined #openstack-ansible | 05:50 | |
| *** hamzaachi has quit IRC | 05:58 | |
| *** ThiagoCMC has quit IRC | 05:59 | |
| *** nurdie has joined #openstack-ansible | 06:05 | |
| *** hamzaachi has joined #openstack-ansible | 06:05 | |
| nurdie | I have an Openstack Ocata deployment that was bootstrapped with OSA and all controller services containerized (python envs, etc). It was just upgraded to Pike with yum. Is there any way to come back from that? | 06:06 |
|---|---|---|
| *** hwoarang has quit IRC | 06:16 | |
| *** hwoarang has joined #openstack-ansible | 06:22 | |
| *** cshen has quit IRC | 06:25 | |
| *** hamzaachi has quit IRC | 06:34 | |
| chandan_kumar | odyssey4me: jrosser evrardjp Hello | 06:47 |
| chandan_kumar | odyssey4me: jrosser evrardjp can we get reviews on this patch https://review.openstack.org/622999 ? | 06:47 |
| chandan_kumar | arxcruz: kopecmartin|off odyssey4me: jrosser evrardjp will i merge this two patch https://review.openstack.org/623187 and https://review.openstack.org/623413 in a single patch? | 06:48 |
| *** aedc has joined #openstack-ansible | 07:01 | |
| *** cshen has joined #openstack-ansible | 07:11 | |
| *** aedc has quit IRC | 07:14 | |
| *** cshen_ has joined #openstack-ansible | 07:17 | |
| *** cshen has quit IRC | 07:19 | |
| *** openstackgerrit has quit IRC | 07:29 | |
| *** ivve has quit IRC | 07:30 | |
| *** jawad_axd has joined #openstack-ansible | 07:34 | |
| *** sum12 has quit IRC | 07:35 | |
| jrosser | odyssey4me: they require each other? | 07:36 |
| *** sum12 has joined #openstack-ansible | 07:37 | |
| *** hamzaachi has joined #openstack-ansible | 07:48 | |
| *** kopecmartin|off is now known as kopecmartin | 07:48 | |
| *** trident has quit IRC | 07:58 | |
| *** trident has joined #openstack-ansible | 08:00 | |
| *** ahosam has joined #openstack-ansible | 08:01 | |
| *** ahosam has quit IRC | 08:01 | |
| *** ahosam has joined #openstack-ansible | 08:01 | |
| *** ahosam has quit IRC | 08:02 | |
| *** shardy has joined #openstack-ansible | 08:05 | |
| evrardjp | chandan_kumar: having a look now | 08:11 |
| *** shardy has quit IRC | 08:21 | |
| *** shardy has joined #openstack-ansible | 08:22 | |
| jrosser | Oh, hmm ignore me, irc phone client confusion, I’m busy till late afternoon today | 08:32 |
| *** markvoelker has joined #openstack-ansible | 08:44 | |
| *** tosky has joined #openstack-ansible | 08:46 | |
| *** markvoelker has quit IRC | 08:49 | |
| *** jonher_ has joined #openstack-ansible | 08:57 | |
| *** fresta_ has joined #openstack-ansible | 08:57 | |
| *** jonher has quit IRC | 09:00 | |
| *** jonher_ is now known as jonher | 09:00 | |
| *** fresta has quit IRC | 09:01 | |
| *** ivve has joined #openstack-ansible | 09:04 | |
| *** DanyC has joined #openstack-ansible | 09:19 | |
| admin0 | \o | 09:26 |
| admin0 | morning all | 09:27 |
| admin0 | nurdie, you meant an OSA deployment got (automatically) upgraded via yum upgrade ? | 09:28 |
| admin0 | did it work :D ? | 09:28 |
| *** openstackgerrit has joined #openstack-ansible | 09:37 | |
| openstackgerrit | Arx Cruz proposed openstack/openstack-ansible-os_tempest master: Better tempest black and whitelist management https://review.openstack.org/621605 | 09:37 |
| *** markvoelker has joined #openstack-ansible | 09:39 | |
| openstackgerrit | Christian Zunker proposed openstack/openstack-ansible-os_cloudkitty master: Add missing libxml2 packages https://review.openstack.org/624618 | 09:52 |
| *** ivve has quit IRC | 09:52 | |
| *** sm806 has quit IRC | 09:54 | |
| *** sm806 has joined #openstack-ansible | 09:55 | |
| *** sm806 has quit IRC | 09:55 | |
| *** sm806 has joined #openstack-ansible | 09:55 | |
| *** electrofelix has joined #openstack-ansible | 10:04 | |
| *** gkadam has joined #openstack-ansible | 10:05 | |
| *** gkadam has quit IRC | 10:05 | |
| nurdie | admin0: No it definitely didn't work, lol. It's still broken | 10:08 |
| nurdie | Having to weed through all of the old Ocata bugs to even get the OSA play to run cleanly. It broke the network, services, alll of it | 10:09 |
| admin0 | someone accidently nuked the galera cluster of an osa cluster which had paying customers .. not sure if your work is more or mine fixing this mess :) | 10:14 |
| admin0 | so the instances are running .. rest all = gone | 10:14 |
| admin0 | will make a nice blog post on how to re-add the base image as image and copy the disk delta to shift the workload from now defunt to new cluster | 10:15 |
| nurdie | admin0: ouch! We do nightly backups of our galera cluster | 10:19 |
| admin0 | in mine, i do hourly | 10:20 |
| admin0 | but i retain only 24 hours back | 10:20 |
| admin0 | nurdie, why are you stil in ocata ? | 10:24 |
| admin0 | what i have seen is .. during the told you hold to not upgrade, something changes in the OS (apt/yum update) that brings in new libs .. so later when you try to upgrade, it starts giving issues which are not documented because those use-changes or changes were not there when the new release was done .. so holding back for too long = more issues | 10:25 |
| admin0 | during the period* | 10:25 |
| admin0 | so now give or take its 1.5+ years .. if you say nothing has changed in your system during this time, then the same osa/ocata might play well ( with the original issues ) .. but if things have changed, you might be facing those issues which only you will face | 10:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/queens: Fix SSL support https://review.openstack.org/624633 | 10:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-galera_server stable/queens: Fix Galera self-signed SSL functionality https://review.openstack.org/624482 | 10:31 |
| *** DanyC has quit IRC | 10:33 | |
| odyssey4me | nurdie the only thing that really matters in the openstack deployment if openstack is hosed is the database - if that's safe, you can recover it | 10:34 |
| odyssey4me | nurdie so you have a choice - but personally I'd recommend upgrading from ocata to pike, then pike to rocky asap... rocky really is the first stable centos deployment we've had | 10:35 |
| nurdie | odyssey4me: If I have a copy of the the database after all VMs were shutdown, can I redeploy and use that DB? It has a Ceph backend, and I'm terrified of losing all of the VMs | 10:36 |
| odyssey4me | nurdie you may hit trouble if your repositories were upgraded to the new centos 7.6 or even 7.5 - we might have to port back some fixes for that to work in ocata or pike | 10:37 |
| odyssey4me | if they're on 7.0 then I think you're ok, but I don't use centos so I can't fully vouch for it | 10:37 |
| odyssey4me | nurdie ok, so are all the vm's running and they still have network connectivity? | 10:38 |
| nurdie | odyssey4me: Everything is down except Ceph | 10:38 |
| nurdie | We turned off all VMs before upgrade attempt, then backed up the DB, again, when we realized it was going south | 10:39 |
| nurdie | Multiple compute hosts, 3 controllers (all running containerized services) | 10:39 |
| odyssey4me | nurdie ok, and is ceph deployed with the OSA deployment or seperately? | 10:40 |
| *** DanyC has joined #openstack-ansible | 10:42 | |
| nurdie | separately | 10:42 |
| odyssey4me | nsmeds I noticed that https://review.openstack.org/624482 brought in part of https://review.openstack.org/572939, so I've ported that back ( https://review.openstack.org/624633 ) and done a clean pick of your patch on top of it | 10:43 |
| odyssey4me | nurdie well, if you have some way of testing the process to iron out the kinks and to understand what you're in for then I'd encourage you to do that first - otherwise you're in for a risky and wild ride and could lose things | 10:46 |
| admin0 | when does 18.1.1 be released .. so that i do not have to cherry pick that ovs patch again and again | 10:46 |
| admin0 | plus there might be more changes | 10:47 |
| odyssey4me | admin0 whenever https://review.openstack.org/624406 is approved, 18.1.1 will be released | 10:47 |
| odyssey4me | once that's done, then https://review.openstack.org/624407 will be clear to merge | 10:47 |
| odyssey4me | nurdie what exactly happened - you did 'yum upgrade' from 7.0 to 7.6 and everything fell apart? | 10:48 |
| nurdie | yum install centos-openstack-pike; yum update | 10:50 |
| nurdie | lol | 10:50 |
| nurdie | On an OSA deploy* | 10:50 |
| odyssey4me | nurdie oh dear | 10:50 |
| odyssey4me | nurdie was that done on all the hosts, or only part of them? | 10:50 |
| nurdie | All of them | 10:50 |
| odyssey4me | well, that was silly :p | 10:51 |
| nurdie | Mhmm. Wasn't my idea | 10:51 |
| odyssey4me | I don't know what 'yum install centos-openstack-pike' actually installs, but is it possible to remove it? | 10:51 |
| odyssey4me | And then does 'yum update' do an upgrade to 7.6, or what does it do exactly? | 10:52 |
| nurdie | But I learned a lot about OSA today lol | 10:52 |
| nurdie | Yes, yum update pulled in the openstack pike updates + the centos updates | 10:53 |
| odyssey4me | nurdie ok, so as long as your /etc/openstack_deploy folder is intact, and your database is intact, the control plane can pretty much be rebuilt from scratch - if you rebuild the hosts using the same MAC/IP's then it should be fine | 10:54 |
| odyssey4me | from there your concern is the compute hosts | 10:55 |
| odyssey4me | there you should be able to uninstall the openstack packages and make sure you purge all their config, then run the OSA deployment over the top of them to get them running again | 10:55 |
| nurdie | I still need to go through the upgrade process, because of DB schemas, etc, right? Which means I still need a working OS deployment lol | 10:56 |
| nurdie | Je pense, but je ne connais pas | 10:56 |
| odyssey4me | I would recommend just going with Pike - OSA will handle the database upgrades and all that. | 10:56 |
| nurdie | hmm | 10:56 |
| odyssey4me | You'll probably have to handle the OSA changes yourself by hand though. | 10:56 |
| odyssey4me | these parts: https://docs.openstack.org/openstack-ansible/pike/user/manual-upgrade.html | 10:57 |
| nurdie | Haha! I just saw a page like that a little while ago that had a disclaimer about the upgrade script being not production safe | 10:58 |
| nurdie | Safer than a yum update on a CentOS box? :p | 10:58 |
| odyssey4me | the upgrade process is a one size fits all, so it's not guaranteed for every environment which is why we had that disclaimer | 10:58 |
| odyssey4me | and, quite honestly, I don't know if it works on CentOS - it's quite possible that one or two things will fall apart | 10:59 |
| odyssey4me | but in your case if you're wiping the env, then there's only a small part of that process which matters... | 11:00 |
| odyssey4me | https://docs.openstack.org/openstack-ansible/pike/user/manual-upgrade.html#clean-up-old-facts | 11:00 |
| odyssey4me | https://docs.openstack.org/openstack-ansible/pike/user/manual-upgrade.html#update-configuration-and-environment-files | 11:00 |
| odyssey4me | https://docs.openstack.org/openstack-ansible/pike/user/manual-upgrade.html#update-user-secrets-file | 11:00 |
| odyssey4me | https://docs.openstack.org/openstack-ansible/pike/user/manual-upgrade.html#clean-up-the-ceph-ansible-galaxy-namespaced-roles | 11:00 |
| odyssey4me | with that your /etc/openstack_deploy will be prepped | 11:01 |
| odyssey4me | everything else is just the same as a standard deployment - except in your case you'll be stopping after setup-hosts.yml to ensure that you restore your database to the galera containersbefore continuing with setup-infrasructure/setup-openstack | 11:02 |
| *** rodolof has joined #openstack-ansible | 11:02 | |
| *** udesale has quit IRC | 11:02 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-os_glance stable/rocky: Fix permissions on /var/lib/glance/cache https://review.openstack.org/624662 | 11:03 |
| *** udesale has joined #openstack-ansible | 11:03 | |
| odyssey4me | nurdie I would definitely recommend putting together a few VM's for a test environment to work through the process to make sure you're familiar with it. | 11:04 |
| nurdie | Much gratitude for the tips, odyssey4me! We may stab that route, because the Ocata OSA repos seem like a giant rabbit hole | 11:05 |
| odyssey4me | nurdie yeah, it's old | 11:05 |
| odyssey4me | nurdie as admin0 recommended earlier, it's better to keep up with releases as quickly as possible - and especially when something like a new centos releases, make sure you're up to date before using it... best to actually maintain your own centos mirrors and only update when you're ready for it | 11:06 |
| nurdie | We have DB copies of the Galera cluster and critical infra VMs before the upgrade. Our stack is already essentially a giant test environment now o_0 | 11:07 |
| admin0 | nurdie, i still find my cluster with a nuked db more fun :) | 11:07 |
| odyssey4me | admin0 that, sir, is a mess | 11:07 |
| nurdie | admin0: hahahaha you have fun with that, and your hourly rates :p | 11:08 |
| odyssey4me | hehehe, yeah - when paid by the hour in that situation, as tough as it is, there is a silver lining | 11:08 |
| odyssey4me | mnaser jrosser I wonder whether it'd be prudent for us to backport https://review.openstack.org/623515 so that centos builds using pike/queens also work with centos 7.6 | 11:09 |
| odyssey4me | let me just do it | 11:10 |
| admin0 | the silver lining here is that you can re-upload the base file as an image, create a new instance with the same name and just copy the disk | 11:10 |
| *** jawad_axd has quit IRC | 11:10 | |
| admin0 | now i am inclined to create a test cluster with ceph, nuke the db and see how far we can go | 11:10 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/queens: prep: remove old machinectl workarounds https://review.openstack.org/624665 | 11:11 |
| odyssey4me | admin0 hopefully the networks in the projects aren't too complex | 11:12 |
| odyssey4me | but still - restoring the VM is the easy part... getting all the networking back to what it was, and all the security groups, etc is only really possible if they have some sort of documentation, which they always do - right?! :p | 11:13 |
| *** sum12 has quit IRC | 11:18 | |
| jrosser | odyssey4me: i think it would be good yes - we're on 7.6 hosts for everything arent we now? | 11:18 |
| *** sum12 has joined #openstack-ansible | 11:19 | |
| *** sum12 has quit IRC | 11:19 | |
| *** sum12 has joined #openstack-ansible | 11:20 | |
| admin0 | lots of scripting .. to dumpxml every running instance, to get the username project and image and then map the security groups using iptables | 11:20 |
| admin0 | a lot of work, but turned out not to be scary .. as all the client instances are running | 11:21 |
| openstackgerrit | lei zhang proposed openstack/openstack-ansible master: Add a note about stale env.d files https://review.openstack.org/624667 | 11:22 |
| openstackgerrit | Chandan Kumar proposed openstack/openstack-ansible-os_tempest master: Added support for running tempestconf from packages https://review.openstack.org/622999 | 11:29 |
| *** markvoelker has quit IRC | 11:30 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-lxc_hosts stable/pike: prep: remove old machinectl workarounds https://review.openstack.org/624674 | 11:30 |
| *** ansmith has joined #openstack-ansible | 11:50 | |
| *** jamesdenton has quit IRC | 11:56 | |
| *** jamesdenton has joined #openstack-ansible | 12:00 | |
| *** markvoelker has joined #openstack-ansible | 12:05 | |
| *** ansmith has quit IRC | 12:05 | |
| gundalow | https://github.com/ansible/ansible/pull/37113 `Added support for missing options capabilities and root_device in properties of os_ironic.py ansible module` Could I get a final review from someone here? Thanks in advance. | 12:06 |
| odyssey4me | gundalow looking, although I may not have enough understanding to know whether it's the right thing to do | 12:30 |
| gundalow | Thanks :) | 12:31 |
| odyssey4me | ah, I see that julia has verified it which works as a SME review for me | 12:31 |
| odyssey4me | gundalow I see you've done some feedback, do you want a revision before merging? the functionality seems right to me | 12:32 |
| gundalow | OK, I'll edit that and merge. Thanks :) | 12:33 |
| jamesdenton | odyssey4me Quick chat about https://review.openstack.org/#/c/454450/ if you have a sec. I could not find the referenced (upcoming) documentation for post-upgrade (P->Q), specifically how to migrate from container to bare metal. Seems to me like nothing changes w/r/t existing containers, but NEW network_hosts will have agents deployed on BM. This sort of conflicts with the language of the release notes IMO. -> | 12:35 |
| jamesdenton | https://docs.openstack.org/releasenotes/openstack-ansible/queens.html#relnotes-17-0-0-stable-queens-upgrade-notes | 12:35 |
| odyssey4me | jamesdenton yeah, unfortunately cloudnull never put anything together as a guide - and it would be really nice to have something in docs to help people | 12:37 |
| odyssey4me | as I understand it, there will - after the upgrade, be agents on bare metal and in containers - although when I was working on the patch to remove the old containers I did not see that | 12:38 |
| odyssey4me | I'm not all that familiar with neutron, so I decided to leave it to others to figure out. | 12:38 |
| odyssey4me | in theory you should already have new agents and can work through scheduling (using the at&t tool perhaps, or native tooling) all routers/networks/etc to the ones on bare metal, disable the ones in containers, then once they're all moved over you can wipe the old container | 12:39 |
| jamesdenton | RIGHT. My experience so far is that the old containers remained (expected) and nothing changed in the inventory, so agents were not deployed on BM. I did add a new network_host, though, and agents were deployed on BM to that. I suspect that some inventory munging will need to be done to force BM deploy to existing infra nodes during the upgrade, and then the container removal exercise post-upgrade | 12:39 |
| odyssey4me | it should be totally automatable, but it needs someone with the time and expertise to work that out | 12:40 |
| jamesdenton | the router and network shuffle, as you say | 12:40 |
| *** dave-mccowan has joined #openstack-ansible | 12:40 | |
| jamesdenton | well, i'll give the docs a workup. and the automation may need to be left to more capable hands | 12:41 |
| odyssey4me | aha, ok - that sort of makes sense... so the issue is that there is already an inventory entry for that network_host... so it keeps deploying to the same one | 12:41 |
| odyssey4me | hmm, I wonder if there's a way we can do it online somehow | 12:41 |
| jamesdenton | I'm trying to get some additional info from this bug report to see if/how they deviated from the upgrade notes: https://bugs.launchpad.net/openstack-ansible/+bug/1804770 | 12:42 |
| openstack | Launchpad bug 1804770 in openstack-ansible "Pike -> Queens Upgrade: Documentation for OVS setup and neutron agent rebalancing missing" [High,Confirmed] - Assigned to James Denton (james-denton) | 12:42 |
| odyssey4me | It seems that it'd be possible to remove the container from the inventory, and with the new env.d when re-running ansible it won't recreate the inventory entry for the container - but then we lose any ways of running ansible against it. :/ | 12:42 |
| jamesdenton | so you think removing the neutron_agents_containers from inventory won't result in a new entry for the respective BM host? | 12:44 |
| odyssey4me | jamesdenton I don't think it will - but the trick is that we need to ensure that the new agents are on the BM host before we remove the cotainer. | 12:45 |
| pabelanger | morning! Do you think we can approve https://review.openstack.org/624407/ today, to update SHAs for stable/rocky for 18.1.2? I guess that needs a release patch first? | 12:45 |
| odyssey4me | pabelanger that won't go into gate until https://review.openstack.org/624406 merges | 12:45 |
| odyssey4me | and it's likely that will only happen once tonyb has seen it, which is likely late tonight for you | 12:46 |
| odyssey4me | early morning for him :) | 12:46 |
| odyssey4me | jamesdenton that's a nice bug report :) | 12:48 |
| pabelanger | odyssey4me: ack, thanks | 12:49 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ops master: MNAIO: Recovery the galera cluster with no gvwstate.dat file https://review.openstack.org/624687 | 12:53 |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: nspawn: correct task names https://review.openstack.org/622183 | 12:53 |
| *** vakuznet has joined #openstack-ansible | 13:07 | |
| *** hamzaachi has quit IRC | 13:19 | |
| *** hamzaachi has joined #openstack-ansible | 13:20 | |
| *** markvoelker has quit IRC | 13:28 | |
| *** vollman has joined #openstack-ansible | 13:30 | |
| openstackgerrit | Merged openstack/openstack-ansible-ops master: MNAIO: Recovery the galera cluster with no gvwstate.dat file https://review.openstack.org/624687 | 13:31 |
| *** rjgibson has joined #openstack-ansible | 13:44 | |
| *** ansmith has joined #openstack-ansible | 13:45 | |
| *** markvoelker has joined #openstack-ansible | 14:01 | |
| arxcruz | odyssey4me: evrardjp do you guys mind take a look at https://review.openstack.org/#/c/621605/ ? | 14:13 |
| arxcruz | when you have time | 14:13 |
| odyssey4me | arxcruz um, what the heck? I get the impression that you're not aware of how to make use of group_vars/host_vars/extra_vars ... | 14:14 |
| odyssey4me | arxcruz also, part of the point of the joint working on this role is to have a common black/whitelist rather than managing two | 14:15 |
| *** fresta has joined #openstack-ansible | 14:17 | |
| arxcruz | odyssey4me: okay... sorry... | 14:17 |
| *** jonher_ has joined #openstack-ansible | 14:17 | |
| arxcruz | we can have a common black/whitelist | 14:18 |
| *** maharg101 has joined #openstack-ansible | 14:19 | |
| *** rodolof has quit IRC | 14:20 | |
| *** jonher has quit IRC | 14:21 | |
| *** jonher_ is now known as jonher | 14:21 | |
| *** rodolof has joined #openstack-ansible | 14:21 | |
| *** fresta_ has quit IRC | 14:21 | |
| openstackgerrit | Andy Smith proposed openstack/openstack-ansible master: Add qdrouterd role for rpc messaging backend deployment https://review.openstack.org/624184 | 14:22 |
| maharg101 | hi, just reading https://bugs.launchpad.net/openstack-ansible/+bug/1657518 regarding firewalld, has anyone built up a list of firewall requirements for openstack that could be shared ? | 14:32 |
| openstack | Launchpad bug 1657518 in openstack-ansible "Hosts role should disable or remove firewalld" [High,Fix released] - Assigned to Major Hayden (rackerhacker) | 14:32 |
| *** irclogbot_0 has quit IRC | 14:36 | |
| *** irclogbot_0 has joined #openstack-ansible | 14:51 | |
| *** cshen_ has quit IRC | 14:54 | |
| *** Miouge has quit IRC | 14:55 | |
| *** Miouge has joined #openstack-ansible | 14:57 | |
| *** DanyC has quit IRC | 15:07 | |
| *** DanyC has joined #openstack-ansible | 15:07 | |
| *** DanyC has quit IRC | 15:11 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: Remove the log container in the AIO config https://review.openstack.org/619526 | 15:14 |
| noonedeadpunk | I've got a bunch of small patches for review, so if someone has time for it - you're welcome https://goo.gl/8QpssC | 15:14 |
| *** markvoelker has quit IRC | 15:18 | |
| odyssey4me | noonedeadpunk I'll look through them now. | 15:18 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_masakari stable/rocky: Add missing docs/releasenotes tests https://review.openstack.org/616635 | 15:23 |
| *** sum12 has quit IRC | 15:28 | |
| *** sum12 has joined #openstack-ansible | 15:28 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_horizon master: Adds ability to enable domain dropdown list on login page https://review.openstack.org/622256 | 15:32 |
| openstackgerrit | kourosh vivan proposed openstack/openstack-ansible-os_trove stable/rocky: Add segmentation option https://review.openstack.org/624741 | 15:33 |
| *** DanyC has joined #openstack-ansible | 15:33 | |
| *** DanyC has quit IRC | 15:34 | |
| *** DanyC has joined #openstack-ansible | 15:35 | |
| *** rodolof has quit IRC | 15:38 | |
| *** rodolof has joined #openstack-ansible | 15:39 | |
| noonedeadpunk | odyssey4me: regarding your comment here https://review.openstack.org/#/c/623643/ - I've faced with the problem during upgrade to the Rocky, and this workaround worked for me, and didn't brought any problems. However offered option will be defenitely safer. | 15:39 |
| odyssey4me | noonedeadpunk ok, I can approve it - but if you think my suggestion is safer then could you push up a patch to change it in master | 15:39 |
| noonedeadpunk | So, should I abandon this cherry-pick, place offered patch for the master, and then squash them and backport to rocky? | 15:39 |
| noonedeadpunk | yep, sure:) | 15:40 |
| openstackgerrit | caoyuan proposed openstack/openstack-ansible-os_zaqar stable/rocky: Add description section https://review.openstack.org/624743 | 15:41 |
| odyssey4me | noonedeadpunk after testing that it works for you, of course :) | 15:41 |
| *** vollman has quit IRC | 15:52 | |
| *** kopecmartin is now known as kopecmartin|off | 16:05 | |
| *** cshen has joined #openstack-ansible | 16:05 | |
| *** gyee has joined #openstack-ansible | 16:19 | |
| *** udesale has quit IRC | 16:19 | |
| nsmeds | odyssey4me: saw your change - understood. thanks for helping | 16:20 |
| noonedeadpunk | odyssey4me: I've tried offered option (and several common as well), and every ends up with this error: http://paste.openstack.org/show/737147/ | 16:24 |
| odyssey4me | noonedeadpunk argh.. must be a bug in it somewhere - if you're concerned enough to figure it out, please do - I don't mind either way | 16:26 |
| spotz | about if needed | 16:27 |
| spotz | had 2 days where the channel was broke for me:( | 16:27 |
| noonedeadpunk | Actually, I ended up with these quotes, as wasn't able to find more proper solution. But I'll do another try | 16:28 |
| *** cshen has quit IRC | 16:30 | |
| noonedeadpunk | I think, the point is, that when jinja tries to get an address from dict, ansible_ib1.87d2 should be quoted (or dot somehow escaped) that it was interpreted as a single key | 16:32 |
| *** macza has joined #openstack-ansible | 16:34 | |
| *** cshen has joined #openstack-ansible | 16:46 | |
| *** radeks has quit IRC | 16:49 | |
| *** cshen has quit IRC | 16:51 | |
| jrosser | mnaser: one for you here https://review.openstack.org/#/c/624336/ | 16:52 |
| *** electrofelix has quit IRC | 16:53 | |
| *** cshen has joined #openstack-ansible | 16:56 | |
| mnaser | jrosser: thats perfect thanks | 17:18 |
| jrosser | that should unlock lxc_container_create tests i think | 17:19 |
| *** DanyC has quit IRC | 17:22 | |
| openstackgerrit | Merged openstack/openstack-ansible-os_masakari stable/rocky: Add missing docs/releasenotes tests https://review.openstack.org/616635 | 17:23 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cloudkitty master: Add missing libxml2 packages https://review.openstack.org/624618 | 17:26 |
| mnaser | jrosser: it will | 17:27 |
| mnaser | i think it will just unlock the patch that removes that tho | 17:27 |
| openstackgerrit | Merged openstack/openstack-ansible-os_horizon master: Adds ability to enable domain dropdown list on login page https://review.openstack.org/622256 | 17:29 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/rocky: Ensure that LDAP config is deployed on all keystone hosts https://review.openstack.org/624430 | 17:30 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts master: Ensure that systemd-resolved is present in the Centos container https://review.openstack.org/624336 | 17:37 |
| *** spatel has joined #openstack-ansible | 17:38 | |
| openstackgerrit | Jonathan Rosser proposed openstack/openstack-ansible-lxc_hosts stable/rocky: Ensure that systemd-resolved is present in the Centos container https://review.openstack.org/624771 | 17:38 |
| spatel | jamesdenton: morning! | 17:38 |
| *** ianychoi has joined #openstack-ansible | 17:38 | |
| spatel | i have neutron VLAN question if you around | 17:39 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/queens: Ensure that LDAP config is deployed on all keystone hosts https://review.openstack.org/624433 | 17:40 |
| spatel | I should say afternoon :) | 17:40 |
| openstackgerrit | Merged openstack/openstack-ansible-os_cinder master: Fix ansible deprecation warnings https://review.openstack.org/621656 | 17:41 |
| openstackgerrit | Merged openstack/openstack-ansible-os_neutron master: Adds ability to set no-resolv for dnsmasq. https://review.openstack.org/620076 | 17:44 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts stable/queens: prep: remove old machinectl workarounds https://review.openstack.org/624665 | 17:44 |
| jamesdenton | hello spatel. whats up? | 17:44 |
| jrosser | mnaser: http://logs.openstack.org/34/624434/1/gate/openstack-ansible-functional-centos-7/1dc28e7/logs/host/lxc-cache-prep-commands.log.txt.gz | 17:45 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_ceilometer master: Add ability to override meters.yaml and event_definitions.yaml https://review.openstack.org/623239 | 17:47 |
| spatel | jamesdenton: I have vlan_200 on my openstack for public and its near to full and i am working with ISP to get new public pool, so is it possible i can add new public subnet to existing vlan ? | 17:49 |
| spatel | If not that i have to create new VLAN on all physical switch ( I have 200 switches in network :( ) | 17:49 |
| jamesdenton | you can. the upstream router/gateway for that network needs to be able to handle multiple L3 subnets on a single interface, though. | 17:50 |
| jamesdenton | most routers can do that. firewall devices (like Cisco ASA) maybe not so much | 17:50 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible master: [WIP] Clean-up and simplify the major upgrade https://review.openstack.org/624773 | 17:51 |
| *** hamzaachi has quit IRC | 17:51 | |
| noonedeadpunk | folks, does anyone know if ceilometer and telemetry projects are alive? As nobody cares about reviews and, actually, filled bugs there... | 17:51 |
| spatel | Yes. new subnet i have to setup on my gateway switches.. | 17:51 |
| jamesdenton | spatel so then it's just a matter of 'openstack subnet create --network EXISTING_VLAN_NET ...' | 17:52 |
| spatel | This is what i did to create my first VLAN network and subnet | 17:52 |
| spatel | neutron subnet-create net_pub_vlan_200 71.xxx.xxx.0/24 --name sub_pub_vlan_200 --allocation-pool start=71.xxx.xxx.2,end=71.xxx.xxx.254 --dns-nameservers 8.8.8.8 --gateway=71.xxx.xxx.1 | 17:52 |
| spatel | Can i use same command to add 72.xxx.xxx.0/24 subnet? | 17:52 |
| jamesdenton | yes. just give it a different name | 17:53 |
| jamesdenton | gateway will be different, start/end, etc. | 17:53 |
| spatel | Different name for subnet right but keep "net_pub_vlan_200" same ? | 17:53 |
| jamesdenton | right | 17:53 |
| spatel | something like this.. | 17:54 |
| spatel | neutron subnet-create net_pub_vlan_200 71.xxx.xxx.0/24 --name sub_new_vlan_200 --allocation-pool start=72.xxx.xxx.2,end=72.xxx.xxx.254 --dns-nameservers 8.8.8.8 --gateway=72.xxx.xxx.1 | 17:54 |
| openstackgerrit | Merged openstack/openstack-ansible-os_trove master: Add segmentation option https://review.openstack.org/623533 | 17:54 |
| jamesdenton | be mindful of the cidr you're passing, that would need to be 72.xxx also | 17:54 |
| spatel | jamesdenton: typo..... :) | 17:55 |
| spatel | yes it should be 72.... | 17:55 |
| spatel | jamesdenton: thank you!!! | 17:55 |
| jamesdenton | looks good to me. sure | 17:55 |
| spatel | I have almost 175 compute node in openstack :) how do i find out what will be the limit of cloud? | 17:56 |
| *** rodolof has quit IRC | 17:56 | |
| jamesdenton | push it till it breaks, then N-1 | 17:56 |
| spatel | touch wood everything looks good so far.. load on infra nodes around 2 to 3%.. | 17:56 |
| jrosser | spatel: you doing any vrrp or anything for the gateway of those subnets? allocating .2 to .254 leaves you little wiggle room for that sort of thing | 17:57 |
| spatel | I have place order for 100 more compute nodes... so wondering should i build new cloud or keep adding in existing one :) | 17:57 |
| jamesdenton | good catch jrosser | 17:57 |
| jrosser | i tend to leave .1 to .9 for such things and have a rule that .5 is vrrp gateway in all /24's | 17:58 |
| spatel | jrosser: 1, 2, 3 first ip will be used by HRSP or VRRP.. | 17:58 |
| jrosser | then you have room for two routers, and spare to add/change if you need to upgrade or stuff like that | 17:58 |
| noonedeadpunk | odyssey4me: It seems, that 'ansible_{{ hostvars[inventory_hostname] | json_query(find_bridge) | replace('-','_') }}' has to be quoted, as the problem occurs inside json_query(), when it see extra dot inside of the interface name (like ansible_ib1.87d2.ipv4.address), while the query should be kinda "'ansible_ib1.87d2'.ipv4.address" | 17:58 |
| jrosser | i guess in this case i'm a little old-school in assigning actual ip to the router interfaces in addition to the vrrp vip | 18:00 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone stable/ocata: Ensure that LDAP config is deployed on all keystone hosts https://review.openstack.org/624436 | 18:00 |
| spatel | jrosser: is that VRRP for DHCP? | 18:01 |
| jamesdenton | spatel your physical gateway device | 18:02 |
| jamesdenton | is it HA? | 18:02 |
| jrosser | no, to make the subnet gateway H/A | 18:02 |
| spatel | i am using physical router gateway so i don't need any VRRP for routing.. | 18:02 |
| openstackgerrit | Merged openstack/openstack-ansible-os_keystone master: Force force-tlsv12 only https://review.openstack.org/623240 | 18:03 |
| spatel | I don't have any vRouter on openstack.. | 18:03 |
| jrosser | no this is on your hardware | 18:03 |
| jrosser | if you have two routers you really should arrange with something like VRRP that the gw ip will fail over between them | 18:03 |
| spatel | Yes that is true... floating ip ( I have HSRP setup ) | 18:04 |
| spatel | but you only need 3 ip address to setup VRRP or HSRP | 18:04 |
| spatel | just trying to understand i am on same page.. or i am missing details here | 18:05 |
| jamesdenton | yeah, what he's implying is that make sure your dhcp allocation range doesn't overlap with those IPs | 18:05 |
| jamesdenton | otherwise neutron may have out an IP configured on the router itself | 18:05 |
| spatel | jamesdenton: :) got it.. | 18:05 |
| jamesdenton | *hand | 18:05 |
| spatel | That command i posted was just example.. i didn't realized .2 to .254 range i specified there | 18:06 |
| jamesdenton | ahh ok | 18:06 |
| spatel | jamesdenton: also DHCP instance also take 3 IPs right from public pool? | 18:06 |
| spatel | in 3 node cluster | 18:06 |
| jamesdenton | if you have 3 DHCP agents, yes | 18:07 |
| spatel | perfect! that is what i thought | 18:07 |
| spatel | jamesdenton: how do i measure openstack limit ? any components i should keep eye on it.. before i say *NOT MORE* compute node | 18:08 |
| spatel | right now i am keep adding compute node every weeks... | 18:08 |
| jamesdenton | others may be better equipped to answer that. if you're not using l3 agent, l2 pop, etc. then the number would be higher. i would say watch DB and Rabbit | 18:09 |
| jamesdenton | i've always heard ~250 is a good spot | 18:09 |
| spatel | hmm! okay.. currently i have 175 so i will keep eye.. | 18:10 |
| spatel | jamesdenton: at some point i want to migrate my existing LinuxBridge to OVS (to use dpdk ) | 18:11 |
| spatel | don't know how hard its going to be but.. i have to go that path.. SR-IOV is just extra pain with performance :) | 18:11 |
| jamesdenton | i think we're a ways out from that. | 18:11 |
| jamesdenton | have you kicked the tires on OVS+DPDK to see if you get performance close to that of SR-IOV? | 18:12 |
| spatel | No i didn't | 18:12 |
| spatel | i will soon but.. 2019 goal | 18:13 |
| *** maharg101 has quit IRC | 18:14 | |
| *** maharg101 has joined #openstack-ansible | 18:17 | |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron stable/rocky: Adds ability to set no-resolv for dnsmasq. https://review.openstack.org/624779 | 18:23 |
| openstackgerrit | Matthew Thode proposed openstack/openstack-ansible-ops master: Chage cpu model to allow for nested virt to work https://review.openstack.org/624780 | 18:23 |
| prometheanfire | odyssey4me: if around, this should fix things... https://review.openstack.org/624780 | 18:24 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron stable/queens: Adds ability to set no-resolv for dnsmasq. https://review.openstack.org/624781 | 18:29 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron stable/queens: Adds ability to set no-resolv for dnsmasq. https://review.openstack.org/624781 | 18:31 |
| *** shardy has quit IRC | 18:35 | |
| *** goldenfri has joined #openstack-ansible | 18:44 | |
| openstackgerrit | Merged openstack/openstack-ansible master: Added masakari-dashboard repo to openstack_services https://review.openstack.org/619062 | 18:55 |
| jrosser | noonedeadpunk: I don’t think it’s true that no one cares about the ceilometer roles, it’s just they’ve been in a bad place for so long that core reviewers like me don’t necessarily have it running in our labs, so it’s hard to get up to speed and be able to make a judgement on patches | 18:58 |
| jrosser | FWIW I think it’s really great that they are getting some attention | 18:59 |
| openstackgerrit | Dmitriy Rabotjagov (noonedeadpunk) proposed openstack/openstack-ansible-os_neutron stable/pike: Adds ability to set no-resolv for dnsmasq. https://review.openstack.org/624786 | 19:01 |
| openstackgerrit | Merged openstack/openstack-ansible stable/pike: Update all SHAs for 16.0.24 https://review.openstack.org/624427 | 19:05 |
| openstackgerrit | Merged openstack/openstack-ansible stable/queens: Update all SHAs for 17.1.6 https://review.openstack.org/624414 | 19:05 |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Add retries for utility container distro package installs https://review.openstack.org/624024 | 19:05 |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Zuul: Simplify the integrated test playbooks https://review.openstack.org/624385 | 19:05 |
| noonedeadpunk | jrosser: I meant about ceilometer itself to be honest:) I will never say that anyone here doesn't care about smth | 19:05 |
| *** spatel has quit IRC | 19:07 | |
| noonedeadpunk | IRC #openstack-telemetry is deserted, bugs are not reviewed for month, even if they have related fixes in gerrit... | 19:08 |
| jrosser | Oh that’s sad | 19:08 |
| noonedeadpunk | So I'm just became worried about the destiny of the project, as we're building billing based on it right now. | 19:09 |
| jrosser | I would like to have more of this working in my cloud | 19:09 |
| jamesdenton | https://julien.danjou.info/lessons-from-openstack-telemetry-deflation/ | 19:09 |
| jamesdenton | i wouldn't put your eggs in that basket | 19:11 |
| noonedeadpunk | jamesdenton: That's really sad... | 19:12 |
| jamesdenton | agreed | 19:12 |
| mnaser | noonedeadpunk: you should start getting more involved with the project and take the lead :D | 19:13 |
| *** AB2019 has joined #openstack-ansible | 19:14 | |
| jamesdenton | cheers for noonedeadpunk - OpenStack Telemetry PTL | 19:15 |
| noonedeadpunk | I'm not so perfect at Python to take a leadership:) I'd wish I had | 19:16 |
| jrosser | You might find even small involvement surprisingly fruitful, particularly with as a deployer with a use case, and a foot here in a deployment project. A lot could get fixed quickly.... | 19:17 |
| noonedeadpunk | Yep, probably you're right | 19:20 |
| noonedeadpunk | Oh, I've read the book written by this guy - never knew that he's a telemetry PTL | 19:21 |
| *** spatel has joined #openstack-ansible | 19:24 | |
| spatel | Folks... i have quick question, i want to change something in /etc/rsyslog.d/50-default.conf file so how do i rollout my change? | 19:25 |
| spatel | I am seeing this file is static file and not using any template variables | 19:25 |
| spatel | # /etc/ansible/roles/rsyslog_client/files/50-default.conf | 19:26 |
| spatel | what are the options i have to edit this file and rollout to make sure next update won't wipe out my changes? | 19:26 |
| jamesdenton | what are you needing to change? | 19:27 |
| spatel | *.*;local7,auth,authpriv,cron,daemon,mail,news.none -/var/log/syslog | 19:28 |
| spatel | CentOS using /var/log/messages ( my monitoring agent reading this files ) | 19:29 |
| spatel | i want to change /var/log/syslog to /var/log/messages | 19:29 |
| *** vollman has joined #openstack-ansible | 19:29 | |
| spatel | i want to change /var/log/syslog to /var/log/messages | 19:29 |
| jamesdenton | Hmm. symlink? | 19:29 |
| -spatel- [root@ostack-compute-75 ~]# ls -l /var/log/syslog | 19:30 | |
| -spatel- ls: cannot access /var/log/syslog: No such file or directory | 19:30 | |
| spatel | Something is broken in syslog deployment of OSA | 19:30 |
| spatel | its not logging any logs.. | 19:30 |
| spatel | This is what i found if i remove "$PrivDropToUser syslog" & "$PrivDropToGroup syslog" from /etc/rsyslog.conf file then it start logging.. | 19:32 |
| spatel | We have all centOS folks and when they see no /var/log/messages they freaked out :( | 19:33 |
| jamesdenton | may be an oversight for centos deploys on our part. | 19:34 |
| spatel | why don't in OSA make simple variable to symlink it or change path of file.. | 19:34 |
| jamesdenton | https://bugs.launchpad.net/openstack-ansible/+filebug | 19:34 |
| *** maharg101 has quit IRC | 19:35 | |
| spatel | I have find /etc/ansible/roles/rsyslog_client/templates/rsyslog.conf.j2 where i can change that but no variable :( | 19:35 |
| spatel | jamesdenton: i did file bug last week but no activity.. yet | 19:36 |
| *** maharg101 has joined #openstack-ansible | 19:38 | |
| jamesdenton | who is the owner of /var/log/messages? | 19:41 |
| jamesdenton | root:root? | 19:41 |
| vakuznet | spatel related bug https://bugs.launchpad.net/openstack-ansible/+bug/1805239 what bug did you file? | 19:42 |
| openstack | Launchpad bug 1805239 in openstack-ansible "missing logrotate of the syslog, daemon.log " [Medium,Confirmed] - Assigned to Vadim Kuznetsov (vakuznet) | 19:42 |
| jamesdenton | https://bugs.launchpad.net/openstack-ansible/+bug/1807268 | 19:42 |
| openstack | Launchpad bug 1807268 in openstack-ansible "CentOS rsyslog bug" [Undecided,New] | 19:42 |
| spatel | vakuznet: looking.. | 19:48 |
| spatel | jamesdenton: | 19:48 |
| -spatel- [root@ostack-osa templates(keystone_admin)]# ls -l /var/log/messages | 19:48 | |
| -spatel- -rw------- 1 root root 710528 Dec 12 14:40 /var/log/messages | 19:48 | |
| jrosser | Wasn’t there a patch for that? | 19:50 |
| spatel | https://bugs.launchpad.net/openstack-ansible/+bug/1807268 | 19:53 |
| openstack | Launchpad bug 1807268 in openstack-ansible "CentOS rsyslog bug" [Undecided,New] | 19:53 |
| spatel | Here you go vakuznet | 19:53 |
| redkrieg | Does anyone know if this is still the best way to do multiple regions? I'd be fine with only having one keystone and pointing the other region to it if that's practical, I just don't see a good way to do it. https://developer.rackspace.com/blog/keystone-to-keystone-federation-with-openstack-ansible/ | 19:53 |
| *** Adri2000 has quit IRC | 19:54 | |
| *** Adri2000 has joined #openstack-ansible | 19:56 | |
| *** rodolof has joined #openstack-ansible | 19:56 | |
| spatel | jamesdenton: you got it | 19:57 |
| spatel | what are the option i have now ? | 19:57 |
| spatel | Or should i write custom ansible module to roll out my change outside OSA ? | 19:58 |
| spatel | it will be ugly! | 19:58 |
| *** nurdie has quit IRC | 20:07 | |
| *** cshen has quit IRC | 20:34 | |
| *** cshen has joined #openstack-ansible | 20:36 | |
| openstackgerrit | James Denton proposed openstack/openstack-ansible-rsyslog_client master: [WIP] Templatize rsyslog configuration files https://review.openstack.org/624805 | 20:36 |
| jamesdenton | spatel You can try that: https://review.openstack.org/624805 | 20:37 |
| jamesdenton | but i can't guarantee it's the right approach, or there may be something in the works | 20:37 |
| *** nurdie has joined #openstack-ansible | 20:38 | |
| spatel | oh! i got it what you did, looks good | 20:40 |
| spatel | should i do that by hand until patch get merge ? | 20:40 |
| jrosser | mnaser: i'm sure there was some discussion previously about this rsyslog/centos stuff - was there a patch? | 20:40 |
| spatel | all your changes.. | 20:40 |
| jamesdenton | spatel there may already be something in the works by mnaser so it may be in vain at this point. | 20:41 |
| spatel | hmm! | 20:42 |
| jrosser | i've not found anything so might have been imagining it - but it was certainly discussed - maybe in the meeting bug triage actually | 20:43 |
| *** nurdie has quit IRC | 20:43 | |
| jamesdenton | For you, today, you might need to roll changes by hand until something official comes down. I think you can create a symlink for /var/log/syslog, and it would appear in /var/log/messages | 20:43 |
| spatel | I am writing quick ansible module to roll out those changes.. | 20:46 |
| spatel | jamesdenton: thanks again! | 20:47 |
| jamesdenton | sure. lemme know! | 20:47 |
| spatel | jamesdenton: this is what i did.. and rollout to all node http://paste.openstack.org/show/737166/ | 21:04 |
| jamesdenton | after adding the vars? | 21:06 |
| jamesdenton | or that was your playbook after downloading the patch | 21:07 |
| spatel | I did quick hack.. manually rollout those file.. i didn't use your patch | 21:10 |
| spatel | i thought may be something in on the way so waiting for final merge | 21:10 |
| jamesdenton | ok so you changed the values in the files then | 21:10 |
| spatel | in short create ansible module outside OSA and rollout two files and restart rsyslog | 21:11 |
| jamesdenton | cool | 21:11 |
| spatel | Yes!! i did | 21:11 |
| spatel | everything working now | 21:11 |
| spatel | logs are flowing now | 21:11 |
| jamesdenton | subsequent playbook runs may overwrite that, just FYI | 21:11 |
| spatel | Yes i know that :) so for now i made those changes in role/rsyslog_client module | 21:13 |
| spatel | so new box will get correct fix.. until we have something in upstream | 21:14 |
| jamesdenton | good deal | 21:14 |
| spotz | noonedeadpunk: I thought ceilometer was dead, at least the API was removed. How long have things been sitting? Folks may be on holiday | 21:15 |
| spatel | I have internal git to check-in those changes so later i can review what changes i mad overtime.. | 21:15 |
| *** vakuznet has quit IRC | 21:21 | |
| *** vnogin has joined #openstack-ansible | 21:25 | |
| *** hw_wutianwei_ has quit IRC | 21:25 | |
| *** vnogin has quit IRC | 21:30 | |
| *** rodolof has quit IRC | 21:32 | |
| *** rodolof has joined #openstack-ansible | 21:32 | |
| *** ansmith has quit IRC | 21:33 | |
| *** spatel has quit IRC | 21:53 | |
| *** DanyC has joined #openstack-ansible | 22:02 | |
| *** dcdamien has joined #openstack-ansible | 22:02 | |
| *** mmercer has quit IRC | 22:22 | |
| *** AB2019_ has joined #openstack-ansible | 22:22 | |
| *** DanyC_ has joined #openstack-ansible | 22:25 | |
| openstackgerrit | Merged openstack/openstack-ansible stable/rocky: Fixed ability of usage interfaces/bridges with dots (aliases) https://review.openstack.org/623643 | 22:27 |
| *** AB2019 has quit IRC | 22:29 | |
| *** andreykurilin has quit IRC | 22:29 | |
| *** eumel8 has quit IRC | 22:29 | |
| *** logan- has quit IRC | 22:29 | |
| *** AB2019_ is now known as AB2019 | 22:29 | |
| *** DanyC has quit IRC | 22:29 | |
| *** logan- has joined #openstack-ansible | 22:31 | |
| *** trident has quit IRC | 22:32 | |
| *** trident has joined #openstack-ansible | 22:34 | |
| *** cshen has quit IRC | 22:40 | |
| openstackgerrit | Antony Messerli proposed openstack/openstack-ansible-ops master: Better handling for resuming leapfrog upgrade https://review.openstack.org/624455 | 23:14 |
| *** dave-mccowan has quit IRC | 23:19 | |
| *** rodolof has quit IRC | 23:21 | |
| *** rodolof has joined #openstack-ansible | 23:21 | |
| *** DanyC_ has quit IRC | 23:24 | |
| *** mmercer has joined #openstack-ansible | 23:35 | |
| *** rodolof has quit IRC | 23:35 | |
| *** rodolof has joined #openstack-ansible | 23:35 | |
| *** rodolof has quit IRC | 23:37 | |
| *** rodolof has joined #openstack-ansible | 23:38 | |
| *** DanyC has joined #openstack-ansible | 23:40 | |
| *** DanyC has quit IRC | 23:44 | |
| *** dcdamien has quit IRC | 23:47 | |
| *** mmercer is now known as Daemoen | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!