| rm_work | umm maybe it is related | 00:14 |
|---|---|---|
| rm_work | it seems to me that the directory there exists on boot <_< | 00:15 |
| *** ducttape_ has joined #openstack-lbaas | 00:17 | |
| *** klindgren_ has joined #openstack-lbaas | 00:20 | |
| *** klindgren has quit IRC | 00:22 | |
| *** rcernin has quit IRC | 00:22 | |
| johnsom | Umm, no. That would be creepy | 00:24 |
| rm_work | nm, yeah | 00:24 |
| rm_work | i don't even | 00:24 |
| rm_work | trying to figure this out | 00:24 |
| *** harlowja has joined #openstack-lbaas | 00:30 | |
| rm_work | ok got it | 00:32 |
| rm_work | so it's getting: | 00:32 |
| rm_work | OSError: [Errno 9] Bad file descriptor | 00:32 |
| rm_work | from: | 00:32 |
| rm_work | ipr.link('set', index=idx, net_ns_fd=consts.AMPHORA_NAMESPACE, | 00:33 |
| rm_work | IFLA_IFNAME=primary_interface) | 00:33 |
| rm_work | :/ | 00:33 |
| johnsom | Fixed here: https://review.openstack.org/#/c/409953/ | 00:33 |
| *** diogogmt has quit IRC | 00:35 | |
| rm_work | uhh wut | 00:36 |
| rm_work | i have that | 00:36 |
| johnsom | I am really wondering if the image built is based on the patches or just master. | 00:37 |
| rm_work | yeah it's fine on the amp | 00:37 |
| rm_work | oh wait no it isn't | 00:37 |
| rm_work | yeah the amp doesn't have that code | 00:38 |
| rm_work | but i have it on my stack | 00:38 |
| rm_work | :/ | 00:38 |
| johnsom | With that new cache stuff, I have to do the following | 00:38 |
| johnsom | git status in /opt/stack/octavia and check the log, everything has to be committed or the cache stuff doesn't pick it up. | 00:39 |
| rm_work | yeah i mean | 00:39 |
| rm_work | i let stack.sh fetch it | 00:39 |
| rm_work | it's on FETCH_HEAD but it's there | 00:39 |
| johnsom | Also, sometime I have to rm -rf /opt/stack/.cache and the caching stuff gets confused. | 00:39 |
| rm_work | this was from a fresh stack :/ | 00:39 |
| johnsom | But that usually blows up the build with "missing commit" | 00:39 |
| rm_work | :/ | 00:40 |
| rm_work | amp is still pulling code from some specific hash or tag or something i bet | 00:40 |
| *** yuanying has joined #openstack-lbaas | 00:44 | |
| korean101 | rm_work: hi! rm. Can you tell when Newton .deb packages? | 00:49 |
| korean101 | rm_work: and this issue resolved? (https://review.openstack.org/#/c/356590/) | 00:49 |
| johnsom | korean101 We are actively working on those diskimage-builider tripleo issues | 00:50 |
| korean101 | johnsom: OK. thanks. got it | 00:51 |
| korean101 | johnsom: and another question... | 00:51 |
| korean101 | johnsom: fixed DVR bugs?? (https://bugs.launchpad.net/neutron/+bug/1583694) | 00:52 |
| openstack | Launchpad bug 1583694 in neutron "[RFE] DVR support for Allowed_address_pair port that are bound to multiple ACTIVE VM ports" [Wishlist,Triaged] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan) | 00:52 |
| johnsom | korean101 You will need to ask in the neutron channel. That is a neutron bug | 00:52 |
| rm_work | johnsom: for YOUR build was the amp getting the updated code? | 00:52 |
| *** yuanying has quit IRC | 00:52 | |
| rm_work | i don't THINK it's something on my side like a cache... | 00:52 |
| *** yuanying has joined #openstack-lbaas | 00:52 | |
| johnsom | rm_work Yes, my image has the patches I checked out into the /opt/stack/octavia and /opt/stack/disk-imagebuilder | 00:53 |
| *** ducttape_ has quit IRC | 00:53 | |
| korean101 | johnsom: yes. thanks for your kindly answer | 00:53 |
| johnsom | korean101 To my knowledge (and sadness) I don't think it is fixed yet | 00:53 |
| rm_work | weird | 00:53 |
| rm_work | yeah mine does not. | 00:54 |
| rm_work | i am unclear as to why | 00:54 |
| korean101 | johnsom: OMG... so sad. | 00:54 |
| johnsom | rm_work git log octavia: commit 95d7dcf8bf55cb2bd626bf23cf281c9b1ad790be | 00:54 |
| johnsom | git log DIB: commit ad09e02e1688749985369f690dff5e7a6d0acdbb | 00:54 |
| johnsom | rm_work Oh, you are running it by hand right? | 00:55 |
| johnsom | export DIB_REPOLOCATION_amphora_agent=$OCTAVIA_DIR | 00:56 |
| johnsom | export DIB_REPOREF_amphora_agent=$(git -C "$OCTAVIA_DIR" log -1 --pretty="format:%H") | 00:56 |
| johnsom | Otherwise it pulls master | 00:56 |
| johnsom | I pulled that from the plugin.sh | 00:56 |
| rm_work | ah fff | 00:58 |
| rm_work | k | 00:58 |
| rm_work | i hope this is right | 00:59 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Fix some python3 string issues in amphora agent https://review.openstack.org/410345 | 00:59 |
| rm_work | k | 00:59 |
| rm_work | johnsom: those go in ... | 01:00 |
| rm_work | johnsom: my bashrc? | 01:00 |
| johnsom | If you want | 01:00 |
| rm_work | k | 01:00 |
| rm_work | $OCTAVIA_DIR ? | 01:00 |
| rm_work | is that /opt/stack/octavia | 01:00 |
| johnsom | yep | 01:00 |
| johnsom | Ha, I thought you were part of setting that up.... | 01:01 |
| johnsom | grin | 01:01 |
| rm_work | yeah i was T_T | 01:01 |
| rm_work | it's been a few months | 01:01 |
| johnsom | Yeah, I hear you | 01:02 |
| *** ducnc has joined #openstack-lbaas | 01:08 | |
| rm_work | urg, amphora agent debug comes from maind ebug | 01:20 |
| johnsom | yep | 01:21 |
| openstackgerrit | Adam Harwell proposed openstack/octavia: Amphora-agent should log to a distinct location https://review.openstack.org/410487 | 01:23 |
| rm_work | and fixed my tooling to use dib path | 01:25 |
| johnsom | FYI, those will need logrotate.d files... Otherwise we have self destructing amps. | 01:26 |
| rm_work | ugh | 01:32 |
| rm_work | yeah ok | 01:32 |
| rm_work | i'll poke at that in a bit once i figure this out | 01:32 |
| *** yamamoto has quit IRC | 01:33 | |
| rm_work | bbl grabbing dinner while stack stacks | 01:34 |
| *** diogogmt has joined #openstack-lbaas | 01:36 | |
| *** yamamoto has joined #openstack-lbaas | 01:52 | |
| *** ducttape_ has joined #openstack-lbaas | 01:53 | |
| *** ducttape_ has quit IRC | 01:55 | |
| *** ducttape_ has joined #openstack-lbaas | 01:55 | |
| *** ducttape_ has quit IRC | 02:09 | |
| *** fnaval_ has quit IRC | 02:09 | |
| *** fnaval has joined #openstack-lbaas | 02:13 | |
| *** fnaval has quit IRC | 02:15 | |
| *** fnaval has joined #openstack-lbaas | 02:16 | |
| rm_work | johnsom: uhh that git command is ... | 02:17 |
| rm_work | Unknown option: -C | 02:18 |
| rm_work | --git-dir= ? | 02:18 |
| rm_work | nope | 02:19 |
| rm_work | gmm | 02:19 |
| rm_work | *hmm | 02:19 |
| *** ducttape_ has joined #openstack-lbaas | 02:21 | |
| *** kevo has quit IRC | 02:22 | |
| *** chenhb has quit IRC | 02:36 | |
| *** chenhb has joined #openstack-lbaas | 02:36 | |
| *** yatin has joined #openstack-lbaas | 02:44 | |
| *** yuanying has quit IRC | 02:53 | |
| *** yuanying has joined #openstack-lbaas | 02:54 | |
| *** ducttape_ has quit IRC | 02:56 | |
| *** ryu25 has joined #openstack-lbaas | 02:58 | |
| *** yuanying has quit IRC | 02:58 | |
| *** kevo has joined #openstack-lbaas | 03:16 | |
| *** gongysh has joined #openstack-lbaas | 03:19 | |
| *** reedip has joined #openstack-lbaas | 03:23 | |
| *** yuanying has joined #openstack-lbaas | 03:52 | |
| *** ducttape_ has joined #openstack-lbaas | 03:57 | |
| *** ducttape_ has quit IRC | 04:01 | |
| *** ducnc has quit IRC | 04:22 | |
| *** ducnc has joined #openstack-lbaas | 04:24 | |
| *** ducttape_ has joined #openstack-lbaas | 04:33 | |
| *** ducttape_ has quit IRC | 04:44 | |
| *** chlong has joined #openstack-lbaas | 04:44 | |
| *** links has joined #openstack-lbaas | 04:57 | |
| *** anilvenkata has joined #openstack-lbaas | 05:03 | |
| *** ducnc has quit IRC | 05:08 | |
| *** ducttape_ has joined #openstack-lbaas | 05:44 | |
| *** ducttape_ has quit IRC | 05:49 | |
| *** diogogmt has quit IRC | 05:58 | |
| *** diogogmt has joined #openstack-lbaas | 05:58 | |
| *** Alex_Stef has joined #openstack-lbaas | 06:12 | |
| *** amotoki has joined #openstack-lbaas | 06:12 | |
| *** amotoki has quit IRC | 06:14 | |
| *** amotoki has joined #openstack-lbaas | 06:17 | |
| korean101 | i use stable/newton branches (http://paste.openstack.org/show/592295/) | 06:25 |
| korean101 | but still got the same errors... (http://paste.openstack.org/show/592296/) | 06:25 |
| korean101 | Thank you for your efforts. | 06:26 |
| korean101 | i'm still waiting | 06:26 |
| *** kobis has joined #openstack-lbaas | 06:28 | |
| *** ducttape_ has joined #openstack-lbaas | 07:04 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 07:09 | |
| *** eezhova has joined #openstack-lbaas | 07:15 | |
| *** yamamoto has quit IRC | 07:18 | |
| *** ducttape_ has quit IRC | 07:24 | |
| *** gongysh has quit IRC | 07:47 | |
| *** eezhova has quit IRC | 07:52 | |
| *** rcernin has joined #openstack-lbaas | 07:56 | |
| *** yamamoto has joined #openstack-lbaas | 07:56 | |
| *** rcernin has quit IRC | 07:56 | |
| *** rcernin has joined #openstack-lbaas | 07:57 | |
| *** pcaruana has joined #openstack-lbaas | 07:59 | |
| *** ducnc has joined #openstack-lbaas | 08:07 | |
| *** reedip has quit IRC | 08:08 | |
| *** ducnc1 has joined #openstack-lbaas | 08:10 | |
| *** ducnc has quit IRC | 08:12 | |
| *** ducnc1 is now known as ducnc | 08:12 | |
| *** ihrachys has joined #openstack-lbaas | 08:17 | |
| *** reedip has joined #openstack-lbaas | 08:22 | |
| *** ducttape_ has joined #openstack-lbaas | 08:25 | |
| *** ducttape_ has quit IRC | 08:30 | |
| *** eezhova has joined #openstack-lbaas | 08:30 | |
| *** openstackgerrit has quit IRC | 08:48 | |
| *** amoralej|off is now known as amoralej | 08:56 | |
| *** strigazi1AFK has joined #openstack-lbaas | 08:58 | |
| *** strigazi_AFK has quit IRC | 09:00 | |
| *** csomerville has quit IRC | 09:02 | |
| *** cody-somerville has joined #openstack-lbaas | 09:03 | |
| *** strigazi_AFK has joined #openstack-lbaas | 09:04 | |
| *** strigazi1AFK has quit IRC | 09:07 | |
| *** openstackgerrit has joined #openstack-lbaas | 09:26 | |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology - Initial Cluster Manager https://review.openstack.org/405238 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology Initial Distributor Noop Driver https://review.openstack.org/313006 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology - create distributor network flow https://review.openstack.org/409763 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: Active-Active Topology - register/uregister amphorae tasks https://review.openstack.org/409765 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: Active-Active Topology - Cluster DB Tasks https://review.openstack.org/409764 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology OVS-based Distributor Driver https://review.openstack.org/317629 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology OVS-based Distributor Backend https://review.openstack.org/320422 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology - Distributor related tasks https://review.openstack.org/406951 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE - controller network tasks https://review.openstack.org/323481 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology - Distributor image creation https://review.openstack.org/403594 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE - network driver related changes https://review.openstack.org/322494 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE - distributor certificate tasks https://review.openstack.org/406952 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology - distributor creation flow https://review.openstack.org/406953 | 09:26 |
| openstackgerrit | Valleriya Perelman proposed openstack/octavia: ACTIVE-ACTIVE Topology - create shared distributor https://review.openstack.org/406954 | 09:26 |
| *** ducttape_ has joined #openstack-lbaas | 09:56 | |
| *** ducttape_ has quit IRC | 10:00 | |
| *** gongysh has joined #openstack-lbaas | 10:04 | |
| *** strigazi_AFK is now known as strigazi | 10:26 | |
| *** links has quit IRC | 10:38 | |
| *** links has joined #openstack-lbaas | 10:38 | |
| *** gongysh has quit IRC | 10:43 | |
| *** yamamoto has quit IRC | 11:12 | |
| *** gongysh has joined #openstack-lbaas | 11:24 | |
| *** yamamoto has joined #openstack-lbaas | 11:26 | |
| *** ducttape_ has joined #openstack-lbaas | 11:26 | |
| *** ducttape_ has quit IRC | 11:31 | |
| *** gongysh has quit IRC | 12:19 | |
| *** amoralej is now known as amoralej|lunch | 12:55 | |
| *** ducttape_ has joined #openstack-lbaas | 12:57 | |
| *** ducttape_ has quit IRC | 13:01 | |
| *** ducttape_ has joined #openstack-lbaas | 13:09 | |
| *** chlong has quit IRC | 13:32 | |
| *** ducttape_ has quit IRC | 13:45 | |
| *** amoralej|lunch is now known as amoralej | 14:07 | |
| *** ducttape_ has joined #openstack-lbaas | 14:15 | |
| *** fnaval has quit IRC | 14:17 | |
| xgerman | o/ | 14:33 |
| *** ducttape_ has quit IRC | 14:34 | |
| *** links has quit IRC | 14:57 | |
| *** fnaval has joined #openstack-lbaas | 14:57 | |
| *** anilvenkata has quit IRC | 14:58 | |
| *** fnaval_ has joined #openstack-lbaas | 14:59 | |
| greghaynes | johnsom: rm_work ok I think all the dib fixes landed | 15:00 |
| *** fnaval__ has joined #openstack-lbaas | 15:01 | |
| xgerman | hooray! | 15:01 |
| *** fnaval has quit IRC | 15:01 | |
| *** fnaval has joined #openstack-lbaas | 15:02 | |
| *** fnaval_ has quit IRC | 15:04 | |
| *** fnaval_ has joined #openstack-lbaas | 15:04 | |
| *** fnaval__ has quit IRC | 15:05 | |
| *** fnaval has quit IRC | 15:08 | |
| *** anilvenkata has joined #openstack-lbaas | 15:14 | |
| *** ducttape_ has joined #openstack-lbaas | 15:16 | |
| *** chlong has joined #openstack-lbaas | 15:21 | |
| *** anilvenkata has quit IRC | 15:21 | |
| *** csomerville has joined #openstack-lbaas | 15:52 | |
| *** diogogmt has quit IRC | 15:53 | |
| *** cody-somerville has quit IRC | 15:55 | |
| *** _ducttape_ has joined #openstack-lbaas | 15:57 | |
| *** ducttape_ has quit IRC | 16:00 | |
| johnsom | korean101 Yeah, the diskimage-builder project does not have stable branches | 16:02 |
| greghaynes | well, not yet, we havent had a second major release so there isnt a stable branch that could potentially exist | 16:02 |
| greghaynes | once we have a v2 though we likely will | 16:03 |
| greghaynes | (hopefully pretty soon) | 16:03 |
| *** kevo has quit IRC | 16:13 | |
| *** rcernin has quit IRC | 16:14 | |
| *** eezhova has quit IRC | 16:18 | |
| johnsom | greghaynes Having a stable branch would really help us out. I.e. the transition of default release versions causes instability in our stable branches. | 16:28 |
| greghaynes | yea, we will have one once there is a stable branch that can potentially exist | 16:28 |
| greghaynes | I dont understand transition of default release versions? | 16:29 |
| johnsom | trusty->xenial | 16:30 |
| *** outofmemory has joined #openstack-lbaas | 16:30 | |
| greghaynes | ah. So that isnt really something tied to dib's stable api since were really just an installer. If you want to pin that its really simple to do and the interface to do so wont change | 16:31 |
| greghaynes | just DIB_RELEASE=trusty | 16:32 |
| johnsom | Yeah, I put patches in to pin the default release on our side now. | 16:32 |
| *** _ducttape_ has quit IRC | 16:32 | |
| *** outofmemory is now known as reedip_ | 16:32 | |
| johnsom | It's just with elements changing, etc. I would be much more comfortable with having a stable version of DIB paired to a stable release of Octavia | 16:32 |
| *** ducttape_ has joined #openstack-lbaas | 16:33 | |
| greghaynes | right, so I think what youre asking for isnt so much a stable branch but to consider default release and package versions as part of our stable api - right now if we branched off something we'd still change these things there since they arent considered part of our stable api | 16:35 |
| greghaynes | is there an example other than the default release thing? | 16:36 |
| johnsom | Yeah, to me, in the case of DIB, stable branch means more than just API. | 16:36 |
| johnsom | Yeah, the changes we just made for example. Until now xenial was using python2.7 by default right? So our Newton stable branch might have the same issues our master branch is having. | 16:37 |
| greghaynes | its a bit more complex, basically it was just a huge dib bug WRT py3 - xenial was definitely using py3k but dib was failing to install python3 deps because it was unaware | 16:38 |
| greghaynes | so dib was installing python2 deps, and then xenial already had python3, so some things were using py2 and some were using py3 | 16:38 |
| johnsom | Yeah, I get it. It's just one of those changes that can impact us. Another example is the removal of that dependency that started all of this. That broke all of our stable branches as well | 16:39 |
| greghaynes | right but that kind of hits at the issue - we (dib) dont really have control over what and when distros release, so in this case we dont get to make the choice of 'this release is now python3', we just have to be aware that this is what the distro has decided. Pinning a release definitely helps but... | 16:40 |
| *** diogogmt has joined #openstack-lbaas | 16:41 | |
| greghaynes | the issue there is all the distros release and maintain support at different cadences, so theres not a practical way we can really pin fedora, suse, and ubuntu while maintaining a sane stable versioning system | 16:41 |
| johnsom | That wasn't a distro change though, it was an element change. | 16:41 |
| greghaynes | it was a default config option change specifically | 16:42 |
| johnsom | https://review.openstack.org/#/c/402250/1/elements/os-refresh-config/element-deps | 16:42 |
| johnsom | That change | 16:42 |
| greghaynes | oh, thats not dib | 16:43 |
| greghaynes | thats a dib downstream | 16:43 |
| johnsom | Fair enough on that point. And I am excited to remove that dependency | 16:43 |
| greghaynes | yea, but youre bringing up a good point and what im wondering is 1) is there a feasible way for dib to pin default distro releases with dib major versinos and 2) is there another thing were doing WRT not having enough of a stable api that breaks folks | 16:44 |
| greghaynes | and I think really fedora is going to be the pain point with the release pinning TBH | 16:45 |
| greghaynes | because they release and deprecate SUPER fast | 16:45 |
| greghaynes | but, ianw may have thoughts on it | 16:45 |
| *** chlong has quit IRC | 16:45 | |
| johnsom | Yeah, I agree, that cycle is fast | 16:45 |
| greghaynes | yep, and we cant have dib with a default thats deprecated so were basically obligated to churn out major releases somewhat frequently, then | 16:46 |
| *** rcernin has joined #openstack-lbaas | 16:46 | |
| greghaynes | ill poke ianw and maybe ML it | 16:47 |
| greghaynes | its a nasty problem | 16:47 |
| greghaynes | the other big thing id ask to keep in mind is - dib is basically the big distro <-> your app integration layer, so kind of our thought has been that this is something that always breaks because dependencies move foward and our role is basically to be some glue between the two, definitely not a stable interface for a distro | 16:49 |
| *** diogogmt has quit IRC | 16:49 | |
| greghaynes | so in other words - youre probably going to break a lot but hopefuly not any more than you would if you were just following whatever the latest release is | 16:49 |
| *** diogogmt has joined #openstack-lbaas | 16:49 | |
| *** kevo has joined #openstack-lbaas | 16:52 | |
| *** Alex_Stef has quit IRC | 16:55 | |
| *** yamamoto has quit IRC | 16:55 | |
| greghaynes | johnsom: on https://review.openstack.org/#/c/409990/ - I think thats something youll definitely want in CI, I wonder if we could send image build logs to another file? | 16:57 |
| *** yamamoto has joined #openstack-lbaas | 16:57 | |
| *** kevo has quit IRC | 16:57 | |
| johnsom | greghaynes The best answer IMHO would be sending the trace logs to another file. | 16:58 |
| *** chlong has joined #openstack-lbaas | 16:58 | |
| *** kevo has joined #openstack-lbaas | 16:59 | |
| greghaynes | not something dib can do :( | 16:59 |
| greghaynes | but yea, that would be great | 16:59 |
| *** diogogmt has quit IRC | 17:00 | |
| greghaynes | alternatively, we could just add the arg to the image build, leave it off by default, and then if you want the debug just add the -x in your change temporarily | 17:01 |
| johnsom | Yeah, that is my feeling. Not having it on by default in the devstack plugin, but leave it as an option for local use | 17:02 |
| *** yamamoto has quit IRC | 17:02 | |
| greghaynes | ok | 17:02 |
| *** kobis_ has joined #openstack-lbaas | 17:03 | |
| *** kobis_ has quit IRC | 17:03 | |
| johnsom | greghaynes We are grumbling about DIB due to this recent issue, but over the last few cycles it really hasn't been that big of a problem for us. | 17:06 |
| greghaynes | glad to hear :) | 17:07 |
| openstackgerrit | greghaynes proposed openstack/octavia: Add support for diskimage-builder tracing https://review.openstack.org/409990 | 17:07 |
| *** reedip_ has quit IRC | 17:08 | |
| johnsom | greghaynes How would you feel about having an octavia test on DIB? A small scenario that shows we are building a good image. | 17:08 |
| rm_work | johnsom: oh yeah so what is the GOAL of that git command, just need the hash of whatever is currently checked out: | 17:11 |
| rm_work | ? | 17:11 |
| johnsom | Yes, it gets the id of the last commit in the log | 17:11 |
| rm_work | k that's what i thought, should be easy enough | 17:12 |
| rm_work | wonder why my git doesn't have -C | 17:12 |
| greghaynes | We've had this same discussion with ironic before, and really theres two components to the coverage - the image build aspect is pretty easy to get good coverage without a co-gate since really its just 'can we install deps, and the app'. For that its pretty easy to add some functional testing to dib of 'can we install a few apps and their deps', not sure theres really anything specific to octavia | 17:12 |
| greghaynes | there that really can break | 17:12 |
| greghaynes | the other big part though is booting, and its kind of a similar situatoin | 17:12 |
| rm_work | well, a very simple octavia scenario test would cover all the way from creating to booting to running our app | 17:13 |
| greghaynes | right, I guess what im saying is - whats a potential dib break youd catch that isnt an octavia break | 17:13 |
| *** ihrachys has quit IRC | 17:14 | |
| rm_work | hmm | 17:14 |
| greghaynes | and that we couldnt get via dib functional testing | 17:14 |
| johnsom | Element changes (like this issue, yes, it was tripleo, but...). Also, there was a change in the past that broke us. That was when doug proposed adding the co-gate. | 17:15 |
| johnsom | It is odd, I understand. Because there is the potential that distro changes could break that gate. But I don't think that has ever happened to us. | 17:16 |
| *** diogogmt has joined #openstack-lbaas | 17:17 | |
| greghaynes | hrmm. can you remember what the previous break was? This has really been the question ive been waiting for - if theres something we can actually prevent with co-gating then I am all for it, what we discovered before is that we didnt catch a single thing by doing it with ironic | 17:17 |
| greghaynes | because the use case there which I think is similar is they just care that they can install things and copy files in, which is something we can cover pretty well in dib | 17:18 |
| greghaynes | booting is where it gets wierd | 17:18 |
| johnsom | Let me search for the bug | 17:18 |
| greghaynes | are you all part of the integrated gate by chance? | 17:20 |
| johnsom | https://bugs.launchpad.net/neutron/+bug/1541670 | 17:20 |
| openstack | Launchpad bug 1541670 in neutron "lbaas tests gate on dib" [High,Invalid] | 17:20 |
| johnsom | That was the bug. Not too descriptive | 17:20 |
| greghaynes | hrm, basically just saying 'we need gating' | 17:21 |
| *** bdeschenes has joined #openstack-lbaas | 17:21 | |
| greghaynes | yea, so how about if you all ever hit something like 'crap we git bit by this bug from dib' and its not something we can just add as a dib functional test for then well co-gate? | 17:22 |
| johnsom | I may have old meeting logs that describe it better, still looking | 17:22 |
| greghaynes | I think a while back I broke you all when hacking on an haproxy element, if thats it | 17:26 |
| greghaynes | although I thnk that was also either in t-i-e or octavia | 17:26 |
| *** fnaval_ has quit IRC | 17:30 | |
| *** bdeschenes has quit IRC | 17:34 | |
| *** fnaval has joined #openstack-lbaas | 17:35 | |
| *** fnaval_ has joined #openstack-lbaas | 17:38 | |
| *** fnaval has quit IRC | 17:41 | |
| *** AlexeyAbashkin has quit IRC | 17:41 | |
| *** diogogmt has quit IRC | 17:49 | |
| *** eezhova has joined #openstack-lbaas | 17:50 | |
| *** yamamoto has joined #openstack-lbaas | 17:56 | |
| *** klindgren__ has joined #openstack-lbaas | 17:57 | |
| *** ihrachys has joined #openstack-lbaas | 17:57 | |
| *** klindgren_ has quit IRC | 17:59 | |
| *** dasanind has joined #openstack-lbaas | 18:00 | |
| *** yamamoto has quit IRC | 18:01 | |
| *** ihrachys has quit IRC | 18:10 | |
| *** ihrachys has joined #openstack-lbaas | 18:11 | |
| *** yamamoto has joined #openstack-lbaas | 18:26 | |
| *** yamamoto has quit IRC | 18:27 | |
| *** yamamoto has joined #openstack-lbaas | 18:27 | |
| *** yamamoto has quit IRC | 18:27 | |
| rm_work | johnsom: wait, do we not have any custom logrotate stuff already? maybe not? | 18:35 |
| *** ducttape_ has quit IRC | 18:39 | |
| johnsom | Yes, it was all rotated before. | 18:39 |
| johnsom | Just not custom | 18:39 |
| *** ducttape_ has joined #openstack-lbaas | 18:50 | |
| *** yatin has quit IRC | 18:50 | |
| *** kevo has quit IRC | 18:52 | |
| *** ihrachys has quit IRC | 19:00 | |
| *** yatin has joined #openstack-lbaas | 19:02 | |
| *** eezhova has quit IRC | 19:05 | |
| johnsom | So for the amp the logs went into either syslog or upstart logs and were rotated through those systems | 19:05 |
| *** ihrachys has joined #openstack-lbaas | 19:06 | |
| *** ihrachys has quit IRC | 19:09 | |
| *** amoralej is now known as amoralej|off | 19:11 | |
| rm_work | johnsom: wtf | 19:12 |
| rm_work | it's still not pulling my changes | 19:12 |
| rm_work | declare -x DIB_REPOLOCATION_amphora_agent="/opt/stack/octavia" | 19:13 |
| rm_work | declare -x DIB_REPOREF_amphora_agent="18a486baaa04f9dfdae2646f2b362f86c4bbca4b" | 19:13 |
| rm_work | that commit hash is correct for what is in /opt/stack/octavia | 19:13 |
| rm_work | but it's still not on the amp | 19:13 |
| rm_work | AHHHH nm | 19:14 |
| rm_work | ffffff i know | 19:14 |
| rm_work | it's correct NOW... *sigh* | 19:14 |
| *** kevo has joined #openstack-lbaas | 19:14 | |
| ankur-gupta-f | haha | 19:14 |
| rm_work | this is how i solve my problems | 19:14 |
| rm_work | i walk through it out loud in IRC and then answer my own question :P | 19:15 |
| ankur-gupta-f | rm_work: want to come help solve mine :| | 19:15 |
| rm_work | heh what's up? | 19:15 |
| ankur-gupta-f | its just crazy dep issues with building devstack stable/newton and stable/mitaka | 19:15 |
| rm_work | hmm | 19:25 |
| rm_work | where does the stack break? | 19:26 |
| ankur-gupta-f | on either the starting of octavia, or installation of horizon and neutron-lbaas-dashboard. | 19:26 |
| ankur-gupta-f | my last approach is to manually install lbaas-dashboard else I give up. | 19:27 |
| ankur-gupta-f | actually then i will try packstack since thats how one of the bugs was reported | 19:27 |
| *** yamamoto has joined #openstack-lbaas | 19:28 | |
| *** nmagnezi has joined #openstack-lbaas | 19:30 | |
| *** yamamoto has quit IRC | 19:36 | |
| *** kobis has quit IRC | 19:48 | |
| rm_work | hmm | 19:52 |
| rm_work | what dep does octavia claim it's missing? | 19:53 |
| rm_work | uuugh wtf, brand new centos package issue as of like, NOW. wtfffff | 19:54 |
| *** kobis has joined #openstack-lbaas | 19:56 | |
| openstackgerrit | Merged openstack/octavia: Add support for diskimage-builder tracing https://review.openstack.org/409990 | 19:57 |
| johnsom | Octavia meeting starting soon on #openstack-meeting-alt | 19:58 |
| *** pcaruana has quit IRC | 20:33 | |
| *** rm_work has quit IRC | 20:34 | |
| *** rm_work has joined #openstack-lbaas | 20:34 | |
| rm_work | oh whoop | 20:45 |
| rm_work | forgot what channel i was in | 20:46 |
| rm_work | meant to mention the centos kemu thing here | 20:46 |
| *** diogogmt has joined #openstack-lbaas | 20:46 | |
| *** kobis has quit IRC | 20:57 | |
| xgerman | nmagnezi quick question | 20:57 |
| nmagnezi | xgerman, sure | 20:57 |
| xgerman | how do I test https://review.openstack.org/#/c/299998/ | 20:58 |
| xgerman | I started two agents on my devstack and the lb went into perpetual pending update | 20:58 |
| *** _ducttape_ has joined #openstack-lbaas | 20:58 | |
| nmagnezi | xgerman, valid question, 1 sec :) | 20:58 |
| nmagnezi | xgerman, i linked this in one of the bug comments: http://paste.openstack.org/show/591851/ | 20:59 |
| nmagnezi | xgerman, a step by step on howto test this | 20:59 |
| xgerman | ok, thanks | 20:59 |
| nmagnezi | xgerman, thank you German | 21:00 |
| nmagnezi | :) | 21:00 |
| *** ducttape_ has quit IRC | 21:02 | |
| *** ipsecguy has quit IRC | 21:06 | |
| *** _ducttape_ has quit IRC | 21:09 | |
| *** ducttape_ has joined #openstack-lbaas | 21:10 | |
| rm_work | yep. awesome. damnit. | 21:20 |
| rm_work | RDO updated the packages for a centos 7.3 release but we use the same RDO repo and i guess the rest of the deps are included in that release but somehow not available in the repos i have T_T | 21:21 |
| ankur-gupta-f | xgerman: johnsom: sorry missed irc meeting. But yea, like Lubosz said. I failed. Still trying, but can't get past this damn installation | 21:22 |
| xgerman | :-( | 21:23 |
| *** catintheroof has joined #openstack-lbaas | 21:24 | |
| *** ipsecguy has joined #openstack-lbaas | 21:28 | |
| johnsom | rm_work Not sure if this is relevant or not: http://lists.openstack.org/pipermail/openstack-dev/2016-December/108884.html | 21:32 |
| johnsom | "Known issue with CentOS 7.3 and qemu-kvm(-ev) 2.6.0" | 21:32 |
| *** ipsecguy has quit IRC | 21:39 | |
| *** bana_k has joined #openstack-lbaas | 21:47 | |
| rm_work | yeah | 21:53 |
| rm_work | that's what they linked me to in -qa | 21:53 |
| rm_work | it isn't exact, but i think it's the same thing | 21:53 |
| rm_work | i think the new centos base image has the right deps, but the older one i'm using doesn't | 21:54 |
| *** ipsecguy has joined #openstack-lbaas | 21:54 | |
| rm_work | and the repo that includes the new qemu-kvm package is included with devstack, but whatever repos include the deps aren't | 21:54 |
| *** catintheroof has quit IRC | 22:19 | |
| *** nmagnezi has quit IRC | 22:46 | |
| *** ducttape_ has quit IRC | 22:51 | |
| *** rcernin has quit IRC | 23:05 | |
| openstackgerrit | Merged openstack/neutron-lbaas: Auto reschedule loadbalancers from dead agents https://review.openstack.org/299998 | 23:06 |
| rm_work | ok, got more stuff fixed from the issue that just appeared this morning | 23:13 |
| rm_work | if i'm SUPER lucky i might be able to build an amp by EOD >_> | 23:13 |
| *** jschwarz has quit IRC | 23:16 | |
| *** jschwarz has joined #openstack-lbaas | 23:21 | |
| *** ducttape_ has joined #openstack-lbaas | 23:24 | |
| *** chlong has quit IRC | 23:37 | |
| *** fnaval has joined #openstack-lbaas | 23:56 | |
| *** fnaval_ has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!