Perl 6 - the future is here, just unevenly distributed

IRC log for #fuel, 2016-08-01

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
01:32 fatdragon joined #fuel
01:48 ilbot3 joined #fuel
01:48 Topic for #fuel is now Fuel 8.0 (Liberty) https://www.fuel-infra.org/ | Paste here http://paste.openstack.org/ | IRC logs http://irclog.perlgeek.de/fuel/
02:00 fritchie joined #fuel
02:24 fatdragon joined #fuel
02:32 Liuqing joined #fuel
02:43 Liuqing_ joined #fuel
03:39 Egyptian[Home] joined #fuel
03:50 code-R_ joined #fuel
04:00 fritchie joined #fuel
04:25 code-R joined #fuel
05:01 code-R joined #fuel
05:08 code-R_ joined #fuel
05:24 Zer0Byte__ joined #fuel
05:32 Zer0Byte__ joined #fuel
06:00 fritchie joined #fuel
06:12 code-R joined #fuel
06:13 Miouge joined #fuel
06:16 javeriak joined #fuel
06:33 fatdragon joined #fuel
06:59 vkulanov joined #fuel
07:02 javeriak joined #fuel
07:25 NickCao_ joined #fuel
08:00 fritchie joined #fuel
08:27 yassine joined #fuel
08:28 fatdragon joined #fuel
09:01 javeriak joined #fuel
09:09 ikalnitsky akasatkin: ikutukov: asvechnikov: guys, please review and merge if it's ok for you https://review.openstack.org/#/c/332019/
09:13 eric_lopez joined #fuel
09:14 ikutukov Patch is ok, but we can't merge, BVT is red
09:18 HeOS joined #fuel
09:24 Miouge_ joined #fuel
09:26 ikalnitsky ikutukov: is someone working on it?
09:28 ikutukov I'm looking, will take some time
09:28 ekosareva joined #fuel
09:29 fatdragon joined #fuel
09:44 ikutukov this patch could be the reason https://github.com/openstack/fuel-library/commit/8e2f824cbbd043217d617aeabe98ba50aafbd4b9
09:47 ikalnitsky ikutukov: don't we run BVT test on each fuel-library patch?
09:50 ddmitriev joined #fuel
10:00 fritchie joined #fuel
10:08 romcheg ikalnitsky: ping, around?
10:08 ikalnitsky yep
10:09 romcheg I have a question regarding tasks (aka transactions) in nailgun
10:10 romcheg I wonder why start and end datetimes aren't stored in the DB?
10:10 romcheg IIRC the UI team needs that
10:10 romcheg and they even filed a bug for this https://bugs.launchpad.net/fuel/+bug/1593753
10:12 romcheg ikalnitsky: having a start and an end time for every task looks pretty straightforward for me
10:14 ikalnitsky I don't know why you asking me. :) We didn't need that earlier, that's why we probably don't have those fields.
10:17 romcheg ikalnitsky: so basically there were no problem adding that?
10:17 romcheg that caused us to refrain from having that information in the DB
10:19 romcheg jaranovich: ping
10:23 ikutukov This patch in theory could fix BVT https://review.openstack.org/#/c/348931/
10:27 ikalnitsky romcheg: i think we can set start time in receiver's "task_in_orchestration", and end time in deploy_resp.
10:28 romcheg yup
10:28 romcheg ikalnitsky: do we need to announce this as an API change?
10:34 ikutukov ikalnitsky we do need this fields now for the deployment history
10:34 ikutukov UI
10:41 NickCao_ joined #fuel
10:42 NickCao joined #fuel
10:51 ikalnitsky romcheg: i don't know. it's sort of backward compatible change, shouldn't be critical.
10:54 ikalnitsky romcheg: do you plan to work on these? https://review.openstack.org/#/c/298791/ https://review.openstack.org/#/c/298783/ if not, maybe abandon them?
10:55 romcheg ikalnitsky: I will abandon them for now
10:55 romcheg have more high priority tasks to work on ATM
10:55 romcheg will restore later
11:18 DaveJ__ joined #fuel
11:30 fatdragon joined #fuel
11:37 zubchick Hi guys, -2K LOC patch ready to go. Please merge :) https://review.openstack.org/#/c/341585/
11:37 javeriak joined #fuel
11:41 johnavp19893 joined #fuel
11:55 javeriak joined #fuel
11:57 yottatsa joined #fuel
11:58 yottatsa joined #fuel
12:00 fritchie joined #fuel
12:16 jackdastripper Good morning. I keep getting the following error when I try to deploy using the L3 HA feature enabled.  (/Stage[main]/Cluster::Mysql/Exec[wait-initial-sync]/returns) ERROR 2002 (HY000): Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2). It seems mysql and maybe other services are not starting automatically. Even if I manually start them, it still fails.
12:19 falavinha joined #fuel
12:32 ekosareva joined #fuel
12:44 johnavp19891 joined #fuel
12:46 javeriak joined #fuel
12:52 Liuqing joined #fuel
12:53 flerfb0rt1 joined #fuel
13:03 romcheg vkulanov: could you please either comment or resolve comments here? https://review.openstack.org/#/c/340030/10
13:03 jaranovich so, romcheg ikutukov can this issue be fixed in the nearest future? https://bugs.launchpad.net/fuel/+bug/1593753 it affects enough UI development.
13:03 romcheg jaranovich: yes, I will try to accomplish that today
13:03 romcheg I'll keep you posted
13:11 romcheg jaranovich: if it blocks you, you may consider bumping its priority
13:13 romcheg vkulanov: could you please change topics of your patches that are related to deprecating the old CLI to deprecate_old_cli?
13:14 romcheg it's going to make keeping track of them easier
13:16 yottatsa joined #fuel
13:32 fatdragon joined #fuel
13:32 vkulanov romcheg, I've already done this. Here is the only patch for now https://review.openstack.org/#/c/346454/ that dealing with 'deprecate_old_cli'.
13:33 romcheg vkulanov: ah, thanks!
13:47 fritchie joined #fuel
13:50 thumpba joined #fuel
13:59 johnavp1989 joined #fuel
14:00 fritchie joined #fuel
14:05 code-R joined #fuel
14:06 code-R_ joined #fuel
14:06 zefciu joined #fuel
14:09 teran joined #fuel
14:09 maximov joined #fuel
14:14 jaranovich joined #fuel
14:15 romcheg joined #fuel
14:15 athinkingmeat joined #fuel
14:16 justif joined #fuel
14:16 elemoine_ joined #fuel
14:17 dpyzhov joined #fuel
14:17 kate_pimenova joined #fuel
14:17 aglarendil joined #fuel
14:17 dklenov joined #fuel
14:17 aroma joined #fuel
14:17 alex_didenko joined #fuel
14:17 holser_ joined #fuel
14:17 smakar joined #fuel
14:17 fedruantine joined #fuel
14:17 akasatkin joined #fuel
14:18 asaprykin joined #fuel
14:18 geekinutah joined #fuel
14:18 ogelbukh joined #fuel
14:18 zubchick joined #fuel
14:18 thiagolib joined #fuel
14:18 kgalanov joined #fuel
14:18 igorbelikov joined #fuel
14:18 kozhukalov joined #fuel
14:19 omolchanov_ joined #fuel
14:19 serverascode joined #fuel
14:19 akislitsky__ joined #fuel
14:21 tkhno joined #fuel
14:21 agordeev joined #fuel
14:45 thumpba joined #fuel
14:48 GMAzrael joined #fuel
14:49 eric_lopez joined #fuel
14:51 thiagolib joined #fuel
14:53 xarses joined #fuel
15:10 fatdragon joined #fuel
15:21 jaranovich romcheg thank you! will raise the issue priority to High
15:29 thiagolib Hello someone already set the fuel of cinder-volume work on mirantis OpenStack in active / passive
15:32 code-R joined #fuel
15:34 ekosareva folks, please review https://review.openstack.org/#/c/341355/
15:34 code-R_ joined #fuel
15:50 venkat joined #fuel
15:52 vvalyavskiy joined #fuel
15:55 yottatsa joined #fuel
16:07 Zer0Byte__ joined #fuel
16:12 yottatsa joined #fuel
16:21 sneha joined #fuel
16:26 Reepicheep_ joined #fuel
16:33 elopez_ joined #fuel
16:35 yottatsa joined #fuel
16:36 javeriak joined #fuel
17:01 sneha I am facing an issue in fuel-8.0 depolyment  wherein when I add a controller node to an already existing deployment; my deployment on compute node fails. I purged out the default ovs and install my own as a part of the initial deployment; so this failure is expected..Is there any way I can bypass this Fuel check for default ovs module being present? So at
17:01 sneha the time of first deployment when i configure my compute node; i make sure i get rid of the default ovs and install my own. As a result of this when i re-run the deployment after adding a controller node; Fuel puppet manifests complain on this default ovs package being not present.
17:01 sneha How can I pass mypackage name which is not the default package name during l23network module initialization ?
17:02 sneha is there a way to override params.pp file values so as to provide custom package and service names for neutron plugins?
17:03 sneha is there a way to override params.pp file values so as to provide custom package and service names for neutron plugins?
17:13 Billias joined #fuel
17:22 Billias joined #fuel
17:25 mwhahaha sneha: for which package
17:30 fritchie joined #fuel
17:31 sneha openvswitch-switch
17:34 sneha is there any provision to pass "ovs_service_name" to l23 network class ?
17:36 mwhahaha so if you update the master's version of the l23network params.pp you could change it
17:36 mwhahaha the modules get synced from the master so if you want to change it there, that'll get applied correctly going forward
17:36 mwhahaha https://github.com/openstack/fuel-library/blob/stable/8.0/deployment/puppet/l23network/manifests/params.pp#L10
17:49 thumpba joined #fuel
17:57 Miouge_ joined #fuel
17:59 Billias joined #fuel
18:02 sneha is there a way to override params.pp file values so as to provide custom package and service names for neutron plugins?
18:03 sneha how do we modify those parameters using hiera so that our plugin uses the custom package/servce names we want
18:23 permalac joined #fuel
18:33 Miouge joined #fuel
18:34 mwhahaha i don't think we expose being able to override those via hiera
18:36 mwhahaha yea the only way to do it at the moment would be to update the params.pp with what you want for the service name
18:36 mwhahaha the common package name you could do some hiera data binding for l23network::let::ovs_common_package_name i think
18:43 brig-man I got an error from fuel when adding a single node to an existing cluster.  It is showing error on the rest of the cluster but not the added node.  How do I get fuel to tell me what failed.  It is just telling me to check the other nodes.
18:45 mwhahaha brig-man: what version of fuel
18:45 mwhahaha brig-man: also you should be able to check the astute.log for error information
18:54 sneha re  you suggesting me to make the change in the Fuel node to update params.pp?
18:55 brig-man mwhahaha: Fuel-8 upgraded to 8.0.3
18:55 mwhahaha sneha: it's only way currently available to do what you need to as we do have a way to target those changes
19:01 eric_lopez joined #fuel
19:02 brig-man mwhahaha: The last thing in the astute log "update hosts status success"
19:02 brig-man mwhahaha: followed by casting message to nailgun
19:02 mwhahaha any references to 'error'?
19:03 mwhahaha it probably wont be the last lines
19:06 brig-man mwhahaha: For all nodes that were not new I get status message: "Node is not ready for deployment"
19:07 brig-man mwhahaha: "mcollective has not answered"
19:07 mwhahaha what does 'mco ping' return when run on the master
19:07 mwhahaha it should get responses back from all your nodes
19:08 brig-man mwhahaha: master fuel node?
19:08 mwhahaha yea
19:08 brig-man It only comes back with the new node and master.
19:09 mwhahaha well that's a problem
19:09 mwhahaha you may want to try restarting mcollective on all the nodes
19:18 Miouge joined #fuel
19:20 brig-man mwhahaha: OK. Done.
19:21 brig-man mwhahaha: try redeploy?\
19:21 brig-man Or something else to look at?
19:21 mwhahaha did you rerun mco ping?
19:23 brig-man mwhahaha: yes.  All nodes now respond
19:26 mwhahaha yea try again now
19:32 brig-man mwhahaha:  It looks like it wiping and re-installing my existing nodes!
19:33 mwhahaha based on what?
19:34 brig-man mwhahaha: message from Fuel UI: "Installing ubuntu: 80%" and the horizon web interface being unresponsive.
19:34 mwhahaha when you initially deployed, did you have any errors?
19:35 brig-man During the initial deployment of the four nodes, no.
19:36 mwhahaha so it shouldn't wipe the nodes, but i have no idea the state of your environment or the history
19:54 brig-man The nodes rebooted and deployment failed again. mco ping responses on master, "added node" and controller(after the deployment reboot).  Compute nodes(3) don't respond.
19:55 brig-man It looks like at least the controller information is intact.  I just need to get the system back online.
20:05 brig-man From the logs, it looks like fuel provisioning is waiting for reboot but the nodes are taking longer than what is expected (ie the fuel timeout is too short).
20:06 brig-man astute nodes failed to reboot. but they did finally show up.
20:06 mwhahaha reboot timeout is configurable in astute
20:07 mwhahaha for 8, it's in a docker container so it's a little trickier
20:07 brig-man OK.  Which container?
20:07 mwhahaha astute
20:08 mwhahaha easiest way might be to edit /etc/puppet/modules/nailgun/templates/astute.conf.erb and uncomment the reboot_timeout setting
20:08 mwhahaha then restart the container
20:08 mwhahaha because i believe the puppet gets rerun on restarting of the container
20:08 mwhahaha dockerctl restart astute
20:09 HeOS joined #fuel
20:11 yottatsa joined #fuel
20:30 brig-man l
21:01 xarses mwhahaha: puppet will run again if the container is terminated
21:02 jackdastripper mwhahaha: Are you aware of 9.0 deployments failing when selecting L3 HA? I can't get a successful deployment. The main issue seems to be that MySQL isn't starting correctly. I can manually start mysql after a failed deployment and get a bit further.
21:02 mwhahaha no i'm not aware of such a thing
21:03 mwhahaha if you provide logs, we can look into it
21:21 xarses jackdastripper: what did you use as a source of your 9.0?
21:24 jackdastripper The master fuel server was installed using the 9.0 ISO and I setup the local repository using fuel-mirror.
21:31 xarses hmm, thats odd. As mwhahaha noted; can you provide a diagnostic snapshot to look at? it would probably also be helpful to attach it to a bug on launchpad https://bugs.launchpad.net/fuel/+filebug
21:40 jackdastripper okay. The deployment just failed a third time. I'll get the logs uploaded.
21:42 brig-man Is there a way to upgrade to fuel-9 without wiping out the master node?
22:03 brig-man I'm seeing this debug output in my astute logs: "connect to host 10.20.0.2 port 22: Connection refused" where 10.20.0.2 is the ip address of the master for the admin network.  The rest of the message show it was from the master.  Is this a configuration error?
22:13 HeOS joined #fuel
22:19 xarses brig-man: it would technically be possible, however quite complex. the Containerized components was removed in 9 so there isn't a straight upgrade path. As it stands currently you'd have to backup the database, destroy all the containers, and then restore the database and have it migrate from 8.0 to 9.0 and reconfigure-install all of the components w/o containers
22:20 xarses brig-man: if you changed the fuel master ip from 10.20.0.2 yes, its a config error of some kind
22:20 xarses if thats the master IP, then there is some iptables rule blocking the connection
22:21 brig-man No.  That was the original IP address.  It just isn't listening except on the external interface not the internal admin interface.
22:22 brig-man It's listening on the external but not the internal/admin.
22:37 Egyptian[Home] joined #fuel
22:48 goldenfri joined #fuel
22:51 thumpba joined #fuel
23:17 brig-man I did follow http://docs.openstack.org/developer/fuel-docs/userdocs/fuel-install-guide/install/install_change_network_interface.html to change the admin network.
23:25 thumpba joined #fuel

| Channels | #fuel index | Today | | Search | Google Search | Plain-Text | summary