| *** mjturek has quit IRC | 00:03 | |
| *** phuongnh has joined #openstack-dib | 01:07 | |
| *** chhavi__ has joined #openstack-dib | 01:11 | |
| *** chhavi__ has quit IRC | 01:16 | |
| *** andreas_s has joined #openstack-dib | 03:25 | |
| ianw | tonyb: if around, mind a sanity check on https://review.openstack.org/#/c/558362/ ? | 03:28 |
|---|---|---|
| *** andreas_s has quit IRC | 03:30 | |
| *** chhavi__ has joined #openstack-dib | 03:54 | |
| openstackgerrit | Doug Hellmann proposed openstack/diskimage-builder master: add lower-constraints job https://review.openstack.org/555412 | 05:23 |
| *** phuongnh has quit IRC | 06:35 | |
| *** phuongnh has joined #openstack-dib | 06:37 | |
| *** andreas_s has joined #openstack-dib | 06:40 | |
| *** andreas_s_ has joined #openstack-dib | 06:41 | |
| *** andreas_s has quit IRC | 06:45 | |
| *** phuongnh has quit IRC | 06:46 | |
| *** phuongnh has joined #openstack-dib | 06:47 | |
| *** phuongnh has quit IRC | 07:08 | |
| *** phuongnh has joined #openstack-dib | 07:09 | |
| *** phuongnh has quit IRC | 07:23 | |
| *** phuongnh has joined #openstack-dib | 07:23 | |
| *** phuongnh has quit IRC | 07:40 | |
| openstackgerrit | Merged openstack/diskimage-builder master: add lower-constraints job https://review.openstack.org/555412 | 07:49 |
| *** phuongnh has joined #openstack-dib | 07:52 | |
| *** phuongnh has left #openstack-dib | 09:19 | |
| *** jesusaur has quit IRC | 09:26 | |
| *** jesusaur has joined #openstack-dib | 09:48 | |
| *** vmlinuz has joined #openstack-dib | 13:11 | |
| *** vmlinuz has joined #openstack-dib | 13:11 | |
| *** hwoarang has quit IRC | 13:51 | |
| *** mjturek has joined #openstack-dib | 13:52 | |
| *** hwoarang has joined #openstack-dib | 13:55 | |
| *** yolanda_ is now known as yolanda | 15:20 | |
| *** openstackgerrit has quit IRC | 15:33 | |
| *** openstackgerrit has joined #openstack-dib | 16:36 | |
| openstackgerrit | Merged openstack/diskimage-builder master: Formalise saving of /etc/resolv.conf https://review.openstack.org/558362 | 16:36 |
| *** dhill__ has quit IRC | 17:27 | |
| *** dhill__ has joined #openstack-dib | 17:27 | |
| *** chhavi__ has quit IRC | 19:02 | |
| *** dhill__ has quit IRC | 19:02 | |
| *** vmlinuz has quit IRC | 21:37 | |
| johnsom | ianw This https://review.openstack.org/#/c/558362/ is breaking our gates | 21:50 |
| johnsom | http://logs.openstack.org/51/558951/1/gate/octavia-v1-dsvm-scenario/96c2c78/logs/devstacklog.txt.gz#_2018-04-06_20_01_38_268 | 21:51 |
| clarkb | johnsom: really? it was just a doc change (other tahn using -e instead of -L and -f) | 21:51 |
| johnsom | It's suspect given the log. | 21:52 |
| johnsom | We are running the non-minimal ubuntu where that file comes in via a package. | 21:52 |
| johnsom | I'm actually doing the minimal patch right now | 21:52 |
| clarkb | I'm jsut having a hard time understand how that could cause a failure since the behavior shouldn't change | 21:53 |
| clarkb | it even skips over it because that file doesn't exist (it doesn't do the mv) | 21:54 |
| clarkb | johnsom: do you have an older job run we can compare again? | 21:54 |
| johnsom | http://logs.openstack.org/79/553879/1/check/octavia-v1-dsvm-scenario/facdf95/logs/devstacklog.txt.gz | 21:55 |
| johnsom | clarkb It's because it's a symbolic link on 16.04 with the ubuntu element | 21:57 |
| clarkb | hrm I think we may need https://review.openstack.org/#/c/558358/1/diskimage_builder/lib/common-functions in to see the tracing we see where it fails | 21:57 |
| clarkb | johnsom: ya so in the old and new behavior it should still move that aside right? | 21:57 |
| clarkb | oh unless does -e not match on symlinks? | 21:57 |
| * clarkb tests | 21:57 | |
| johnsom | Yeah, actually, that should be the same, I think.... | 21:57 |
| clarkb | oh I Know why | 21:57 |
| clarkb | I bet the target of the symlink doesn't exist and -e is follwing it | 21:57 |
| clarkb | I will test that too | 21:57 |
| johnsom | That is likely the case | 21:58 |
| johnsom | resolv.conf -> ../run/resolvconf/resolv.conf | 21:58 |
| clarkb | ya that is what the problem is | 21:59 |
| clarkb | -e works as expected if the target is valid but fails if target is invalid | 21:59 |
| johnsom | Yeah, the /run stuff shouldn't be there at that point in the image build | 21:59 |
| clarkb | or at all, its created by systemd on boot iirc | 21:59 |
| clarkb | so we go back to -f and -L I guess | 22:00 |
| clarkb | I'll push up a patch | 22:00 |
| clarkb | and you can depends on to confirm it is working | 22:00 |
| johnsom | Ok, sounds good | 22:00 |
| openstackgerrit | Clark Boylan proposed openstack/diskimage-builder master: Don't use -e to test for what might be broken symlink https://review.openstack.org/559436 | 22:06 |
| clarkb | johnsom: ^ try that | 22:06 |
| johnsom | https://review.openstack.org/559437 | 22:06 |
| johnsom | If any one of those DSVMs pass we are good | 22:07 |
| johnsom | clarkb Two of the DSVM gates have passed, it looks good to me. | 23:01 |
| clarkb | johnsom: cool, sorry for the problem now we just need ianw and pabelanger to review it | 23:02 |
| johnsom | Ok | 23:02 |
| johnsom | Too bad there isn't a DIB gate that builds an ubuntu image. | 23:03 |
| clarkb | johnsom: there is | 23:05 |
| clarkb | but it uses ubuntu-minimal | 23:05 |
| johnsom | Ah, yeah, that doesn't load the package with that file in it. I'm working to move us over right now. | 23:05 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!