15:02:20 #startmeeting XenAPI 15:02:25 Meeting started Wed Dec 17 15:02:20 2014 UTC and is due to finish in 60 minutes. The chair is johnthetubaguy. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:02:26 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:02:28 The meeting name has been set to 'xenapi' 15:02:40 #topic XenServer CI 15:02:51 so, wondering how things are going in the world of CI 15:03:03 might wait a little bit for matel to join 15:03:18 So generally things are OK 15:03:22 although we're a bit worried 15:03:27 images aren't rebuilding at the moment 15:03:30 and we don't understand why 15:03:40 there is some weird puppet interaction with iptables 15:03:50 oh dear, thats annoying 15:03:55 but figuring out the root cause for that is hard 15:04:00 Very short term it's not a problem 15:04:05 because we have active images in each region 15:04:06 yeah, thats stuff is a nightmare to debug 15:04:10 but they can quickly get stale 15:04:14 agreed 15:04:42 I think that's all I can say for that though 15:04:54 did we mention last week that we've rebuilt the CI? 15:05:06 mahusive rebuild to pick up the latest nodepool etc 15:05:09 yeah, or at least, its now easily automated 15:05:15 yup 15:05:21 ah, cool, not sure you said that was complete 15:05:28 But the rebasing on nodepool should have given us some more stability in the image generation 15:05:32 and reduce the bill for RAX ;) 15:05:51 {'': 33, 'Failed': 29, 'Passed': 224, None: 2} 15:05:57 Failure rate of 10% is not ideal though 15:06:01 Too high for my liking 15:06:32 yeah 10% seems high 15:06:45 unless people are uploading lots of very broken patches… which is possible I guess 15:07:00 did you add in those extra tests? 15:07:09 just wondering if there is a common cause for the failure 15:07:17 rebuild + resize server are failing ish 15:07:24 Failures 15:07:25 ------------------- 15:07:25 19 No tempest failures detected 15:07:25 3 Fewer than 2 duplicates 15:07:25 3 tempest.scenario.test_shelve_instance.TestShelveInstance.test_shelve_instance 15:07:28 2 tempest.api.compute.servers.test_disk_config.ServerDiskConfigTest.test_resize_server_from_manual_to_auto 15:07:31 2 tempest.api.compute.admin.test_servers.ServersAdminTest.test_rebuild_server_in_error_state 15:07:33 2 tempest.api.compute.servers.test_server_actions.ServerActionsTest.test_rebuild_server 15:07:37 2 tempest.api.compute.servers.test_server_actions.ServerActionsTest.test_resize_server_revert 15:07:40 2 tempest.api.compute.servers.test_disk_config.ServerDiskConfigTest.test_resize_server_from_auto_to_manual 15:07:43 2 tempest.api.compute.servers.test_disk_config.ServerDiskConfigTest.test_rebuild_server_with_auto_disk_config 15:07:46 2 tempest.api.compute.servers.test_server_actions.ServerActionsTest.test_rebuild_server_in_stop_state 15:07:48 hmm, thats a worry 15:07:49 2 tempest.api.compute.servers.test_disk_config.ServerDiskConfigTest.test_rebuild_server_with_manual_disk_config 15:07:52 2 tempest.api.compute.servers.test_disk_config.ServerDiskConfigTest.test_update_server_from_auto_to_manual 15:07:55 but the 19 no-tempest-failures are also frustraing. 15:08:00 implies that devstack failed 15:08:05 sorry for being late. 15:08:09 or we couldn't collect logs (often seems to be a RAX network failure) 15:08:31 but perhaps adding a retry to collect logs would fix the network failures 15:08:52 BobBall: yeah, was just about to say a retry, swift can be a bit "fussy" now and then 15:09:50 I thought the issue was with downloading from the guest to the CI machine 15:09:55 but I'm not certain 15:10:02 I might have a look at adding a retry at some point 15:12:27 ok 15:12:40 I guess that's the CI. 15:12:41 matel: just CI updates 15:12:47 so yeah, all done I guess 15:12:55 #topic Open Discussion 15:12:59 Indeed - I don't have anything else to say on OSCI 15:13:04 our internal CI is very poorly 15:13:04 so any more we need to cover? 15:13:35 We redeployed it but it's not come up so we're having to put a lot of effort into trying to fix various things 15:14:07 yeah. 15:14:39 and Mate has done a stunning job on docs 15:15:54 awesome stuff, there is certainly "room for growth" in that area... 15:16:04 appreciate that getting some attention 15:16:12 it is sure to be hurting adoption 15:18:20 yeah, we no longer have the warning in xenserver docs 15:18:24 Which is good. 15:18:33 cool 15:18:42 any more for any more? 15:19:10 no, not this year. 15:19:14 OK, one bit more... 15:19:17 Watch this space for fun things happening next yar 15:19:19 year* 15:19:29 next week, no meeting I guess? 15:19:39 I'm on vacation 15:19:39 probably same the week after? 15:19:44 Indeed. 15:20:00 #info next meeting on 7th Jan 15:20:10 OK thanks all 15:20:17 matel: :) build-xenserver-core-master-centos worked. 15:20:25 matel: thats for all your hard work on this, really is appreciated :) 15:20:42 Happy christmas and happy new year folks 15:20:45 #endmeeting