| longstaff | No other patches in flight, so please feel free to post patches. | 00:00 |
|---|---|---|
| rm_work | johnsom: err, so then is there no test that actually checks for "option external-check command /var/lib/octavia/ping-wrapper.sh"? | 00:00 |
| rm_work | O_o | 00:00 |
| openstackgerrit | Merged openstack/octavia master: Fix possible DB deadlock scenario https://review.openstack.org/518781 | 00:11 |
| johnsom | Geez, I just noticed we may be running taskflow in serial mode.... | 00:20 |
| rm_work | O_o | 00:20 |
| rm_work | I thought it was specified by the flow | 00:20 |
| johnsom | Well, that was the default anyway, but I think we had the param wrong | 00:21 |
| johnsom | Yeah, but if the overall engine isn't.... | 00:21 |
| rm_work | erk | 00:21 |
| rm_work | how many of our flows actually utilize parallel tasks? | 00:22 |
| johnsom | Like one I think | 00:28 |
| johnsom | It looks like 5 | 00:29 |
| *** fnaval has quit IRC | 00:30 | |
| rm_work | yeah i mean, it seems shocking and like, super dumb... but ... really I don't think it makes that much of a difference | 00:31 |
| johnsom | Yeah, my mistake like three years ago... | 00:32 |
| johnsom | Well, I will fix it now. | 00:32 |
| johnsom | You still have to actually turn it on in the config file | 00:32 |
| rm_work | what does that look like O_o | 00:34 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix TaskFlow loader to honor engine configuration https://review.openstack.org/518812 | 00:34 |
| johnsom | [task_flow] | 00:35 |
| johnsom | # engine = serial | 00:35 |
| johnsom | engine = parallel | 00:35 |
| johnsom | max_workers = 50 | 00:35 |
| *** longstaff has quit IRC | 00:36 | |
| rm_work | cool, i'll have to add that to my config i guess | 00:42 |
| johnsom | Yeah, after that patch merges... | 00:43 |
| rm_work | i mean, i can add it before too, it just won't do anything :P | 00:43 |
| johnsom | true | 00:44 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix TaskFlow loader to honor engine configuration https://review.openstack.org/518812 | 00:57 |
| openstackgerrit | Merged openstack/neutron-lbaas master: More zuul v3 job cleanup after newton EOL https://review.openstack.org/515481 | 01:07 |
| rm_work | johnsom: wait so it was because the arg was wrong?! | 01:15 |
| rm_work | lol | 01:15 |
| johnsom | Yeah, it was kind enough to take it even though it wasn't right/used.... | 01:16 |
| rm_work | T_T | 01:16 |
| johnsom | Who knows, maybe it changed over the history of taskflow | 01:16 |
| rm_work | them **kwards | 01:16 |
| johnsom | And we just didn't notice | 01:16 |
| rm_work | * **kwargs | 01:16 |
| rm_work | heh maybe | 01:16 |
| * rm_work shrugs | 01:16 | |
| rm_work | ok i swear this gentoo building is close | 01:16 |
| rm_work | then i just need to figure out all the network stuff for os_utils | 01:17 |
| rm_work | but gentoo docs/support has historically been phenominal | 01:17 |
| *** sapd_ has joined #openstack-lbaas | 01:24 | |
| *** sapd__ has quit IRC | 01:25 | |
| rm_work | johnsom: are the vars we set in the diskimage-create.sh like "AMP_ARCH" available to be used in stuff like the pre-install.d scripts somehow? | 01:31 |
| rm_work | those are run *inside* the image, so they'd need to be passed through... I wonder if it does that for the ARCH | 01:31 |
| johnsom | I think so if they are exported | 01:31 |
| rm_work | it's a different ent | 01:32 |
| rm_work | *env | 01:32 |
| johnsom | Hmm, I know there is DIB stuff that gets in | 01:32 |
| rm_work | but I am wondering if disk-image-create puts them inside the env since we pass them to it | 01:32 |
| johnsom | But maybe they explicitly set those. | 01:32 |
| rm_work | i guess this is a question for the DIB channel | 01:32 |
| johnsom | Bummer, yeah, I broke centos somehow with the ping patch. I'll have to spin that up local and have a look | 01:34 |
| rm_work | weird O_o | 01:34 |
| rm_work | not sure why/ho | 01:34 |
| rm_work | *how | 01:34 |
| rm_work | unless `command -v` doesn't work | 01:34 |
| johnsom | Yeah, me either | 01:34 |
| rm_work | i wish i could easily ssh to a centos box to check | 01:34 |
| rm_work | <_< | 01:34 |
| rm_work | >_> | 01:34 |
| johnsom | Could be, but that still shouldn't break the agent call like that | 01:34 |
| johnsom | bbl | 01:35 |
| rm_work | kk | 01:35 |
| *** sapd__ has joined #openstack-lbaas | 01:37 | |
| *** sapd_ has quit IRC | 01:37 | |
| *** jdavis has joined #openstack-lbaas | 01:38 | |
| *** jdavis has quit IRC | 01:43 | |
| *** sshank has quit IRC | 01:46 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 01:57 | |
| *** AlexeyAbashkin has quit IRC | 02:01 | |
| *** yamamoto has joined #openstack-lbaas | 03:05 | |
| *** aojea has joined #openstack-lbaas | 03:13 | |
| *** aojea has quit IRC | 03:17 | |
| *** fnaval has joined #openstack-lbaas | 03:20 | |
| openstackgerrit | Merged openstack/neutron-lbaas master: Fix tempest scenario tests waiters https://review.openstack.org/517931 | 03:36 |
| johnsom | https://www.irccloud.com/pastebin/TgfTeKDi/ | 03:42 |
| *** sapd_ has joined #openstack-lbaas | 03:48 | |
| *** sapd__ has quit IRC | 03:48 | |
| johnsom | haproxy-1.5.18-6.el7.x86_64 | 03:49 |
| johnsom | CentOS Linux release 7.4.1708 (Core) | 03:49 |
| johnsom | So, not sure what to do about that. | 04:02 |
| johnsom | 1.5 doesn't have external health checks | 04:03 |
| johnsom | We could go rogue and use: https://dl.iuscommunity.org/pub/ius/stable/CentOS/7/x86_64/repoview/haproxy16u.html | 04:26 |
| johnsom | But it would mean rhel would not work | 04:27 |
| mnaser | johnsom: unpopular opinion, haproxy docker containers | 04:45 |
| mnaser | :p | 04:45 |
| johnsom | Ha, not sure how unpopular | 04:47 |
| mnaser | johnsom: easy cross-os support and interesting ideas surrounding upgrades etc | 04:48 |
| *** ssmith has joined #openstack-lbaas | 05:17 | |
| rm_work | yeah i looked at that | 05:17 |
| rm_work | the custom centos haproxy 1.6 | 05:17 |
| *** slaweq has joined #openstack-lbaas | 05:25 | |
| *** slaweq has quit IRC | 05:29 | |
| *** aojea has joined #openstack-lbaas | 05:35 | |
| *** armax has quit IRC | 05:44 | |
| *** slaweq has joined #openstack-lbaas | 05:46 | |
| *** slaweq has quit IRC | 05:46 | |
| *** slaweq has joined #openstack-lbaas | 05:46 | |
| *** ssmith has quit IRC | 05:50 | |
| *** slaweq has quit IRC | 05:53 | |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 06:02 | |
| *** aojea has quit IRC | 06:33 | |
| *** aojea has joined #openstack-lbaas | 06:39 | |
| *** eN_Guruprasad_Rn has quit IRC | 06:39 | |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 06:40 | |
| *** aojea has quit IRC | 06:50 | |
| *** spectr has joined #openstack-lbaas | 07:09 | |
| *** aojea has joined #openstack-lbaas | 07:14 | |
| *** spectr has quit IRC | 07:20 | |
| *** spectr has joined #openstack-lbaas | 07:21 | |
| *** aojea has quit IRC | 07:29 | |
| *** aojea has joined #openstack-lbaas | 07:32 | |
| *** eN_Guruprasad_Rn has quit IRC | 07:35 | |
| *** eN_Guruprasad_Rn has joined #openstack-lbaas | 07:35 | |
| *** ssmith has joined #openstack-lbaas | 07:47 | |
| *** ssmith has quit IRC | 07:51 | |
| *** rcernin has quit IRC | 07:56 | |
| *** aojea has quit IRC | 07:56 | |
| *** gcheresh has joined #openstack-lbaas | 08:02 | |
| *** annp has joined #openstack-lbaas | 08:20 | |
| *** aojea has joined #openstack-lbaas | 08:24 | |
| *** Alex_Staf_ has joined #openstack-lbaas | 08:28 | |
| *** Yipei has joined #openstack-lbaas | 08:37 | |
| *** Alex_Staf_ has quit IRC | 08:38 | |
| *** gcheresh has quit IRC | 08:41 | |
| *** salmankhan has joined #openstack-lbaas | 08:42 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 08:45 | |
| *** salmankhan has quit IRC | 08:46 | |
| *** sri_ has joined #openstack-lbaas | 08:49 | |
| *** gcheresh has joined #openstack-lbaas | 08:50 | |
| *** aojea has quit IRC | 08:54 | |
| *** rcernin has joined #openstack-lbaas | 09:08 | |
| *** slaweq has joined #openstack-lbaas | 09:12 | |
| *** yamamoto has quit IRC | 09:23 | |
| *** aojea has joined #openstack-lbaas | 09:26 | |
| *** yamamoto has joined #openstack-lbaas | 09:36 | |
| *** ipsecguy_ has joined #openstack-lbaas | 09:44 | |
| *** ipsecguy has quit IRC | 09:45 | |
| isantosp | hi, how could I solve this error? "CertificateGenerationException: Could not sign the certificate request: Failed to load CA Certificate" | 10:07 |
| *** salmankhan has joined #openstack-lbaas | 10:16 | |
| *** aojea has quit IRC | 10:20 | |
| *** aojea has joined #openstack-lbaas | 10:33 | |
| *** gcheresh has quit IRC | 10:38 | |
| *** openstack has joined #openstack-lbaas | 21:55 | |
| *** ChanServ sets mode: +o openstack | 21:55 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 21:55 | |
| *** AlexeyAbashkin has quit IRC | 22:00 | |
| rm_work | johnsom: o/ | 22:01 |
| johnsom | o/ | 22:02 |
| rm_work | my plan was to wake up super early so i'd be around today | 22:02 |
| rm_work | instead i slept for 12h | 22:02 |
| johnsom | Ha, nice. Yeah, travel will do that. | 22:02 |
| rm_work | so ... i'm about to post something that I think qualifies as "interesting" | 22:03 |
| johnsom | I just about have a work around for the CentOS thing done | 22:03 |
| johnsom | Ha, well, it's still a few hours from beer-:30 here, so.... | 22:04 |
| rm_work | i wish centos would just... UPGRADE HAPROXY | 22:04 |
| rm_work | ffff | 22:04 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Enable building gentoo amphora images https://review.openstack.org/518986 | 22:04 |
| johnsom | Yeah, and GIT! | 22:04 |
| rm_work | lol yes | 22:05 |
| rm_work | meanwhile... | 22:05 |
| rm_work | ^^^ | 22:05 |
| rm_work | the image "builds" | 22:05 |
| johnsom | Neat | 22:05 |
| rm_work | with haproxy 1.7.9 woo | 22:12 |
| rm_work | could go with keepalived 1.3 too but dunno if it matters | 22:12 |
| rm_work | right now it's on 1.2 | 22:12 |
| rm_work | johnsom: i am super curious what your centos workaround is :P | 22:13 |
| johnsom | Yeah, I think there is some "optimization with 1.3" , but nothing necessary | 22:14 |
| johnsom | Well, I check the install haproxy version and if it is older than 1.6 I strip the external-check from the config | 22:14 |
| johnsom | i.e. Sucks to be on CentOS | 22:14 |
| rm_work | oh. | 22:15 |
| rm_work | yes, i've been saying that for a while. | 22:15 |
| rm_work | unfortunately that means it sucks to be me for the time being | 22:15 |
| johnsom | It does no harm to the old OS's, but fixes anything not in the stone age | 22:15 |
| rm_work | so it doesn't FIX the ping bug | 22:15 |
| rm_work | for centos | 22:15 |
| johnsom | Nothing stopping you from pulling that 1.6 RPM into your local image | 22:15 |
| rm_work | just makes it not break horribly | 22:15 |
| rm_work | lol yeah well | 22:15 |
| johnsom | Yep | 22:16 |
| johnsom | I put it in the release notes. | 22:16 |
| johnsom | It's all I could come up with other than importing that 1.6 version of HAproxy into our images. | 22:16 |
| rm_work | T_T yeah | 22:16 |
| rm_work | well, so it'll be no change from the previous configs for centos right, so we'll still get at least a TCP connect instead of a PING | 22:17 |
| rm_work | so it'll do SOMETHING | 22:17 |
| johnsom | Correct | 22:17 |
| rm_work | there's a good chance the tcp check will *work* | 22:17 |
| rm_work | and be better than ping | 22:17 |
| rm_work | soooo | 22:17 |
| rm_work | we'll just call it a feature | 22:17 |
| johnsom | Yeah, PING is dumb, so I expect it impacts like 1% of the world | 22:17 |
| rm_work | have you used gentoo before? | 22:21 |
| johnsom | Once a long time ago | 22:22 |
| openstackgerrit | Michael Johnson proposed openstack/octavia master: Fix health monitor type "PING" to use icmp ping https://review.openstack.org/518632 | 22:25 |
| rm_work | heh, k | 22:27 |
| rm_work | package masking is interesting | 22:27 |
| *** cpusmith has quit IRC | 22:27 | |
| rm_work | ugh the osutils bit is going to be the hard part of all this | 22:29 |
| johnsom | Maybe Mathew can help answer the gentoo questions for it | 22:31 |
| rm_work | yes | 22:31 |
| rm_work | I assume he can, but i think he's still traveling / touristing | 22:31 |
| rm_work | he's been responding to me while i'm away and v.v. | 22:31 |
| rm_work | xgerman_ / nmagnezi: want to merge https://review.openstack.org/#/c/518624/ ? | 22:33 |
| rm_work | Simple one. | 22:33 |
| rm_work | I am gonna build on it momentarily i think and try to do the rest | 22:33 |
| johnsom | Nice | 22:34 |
| xgerman_ | K | 22:36 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge keepalived elements https://review.openstack.org/518990 | 22:37 |
| rm_work | whoops | 22:39 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge keepalived elements https://review.openstack.org/518990 | 22:39 |
| *** rcernin has quit IRC | 22:44 | |
| *** AlexeyAbashkin has joined #openstack-lbaas | 22:55 | |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge amphora-agent elements https://review.openstack.org/518992 | 22:58 |
| rm_work | kay | 22:58 |
| rm_work | all done | 22:58 |
| rm_work | that cleanup has been a long time coming | 22:59 |
| *** AlexeyAbashkin has quit IRC | 22:59 | |
| johnsom | Yes | 22:59 |
| johnsom | Hmm, no mention of Octavia in the tripleo project update.... | 23:00 |
| rm_work | :( | 23:00 |
| rm_work | were we supposed to be? | 23:00 |
| rm_work | oh hold up, rhel was very very specific, one sec | 23:00 |
| johnsom | Well, I thought nmagnezi was spending a lot of time integrating octavia into it, so I figured it would have a bullet somewhere | 23:01 |
| *** aojea has quit IRC | 23:03 | |
| rm_work | so for rhel image building we actually pull a PACKAGED version of the amp agent | 23:03 |
| rm_work | but for centos (another in the redhat family) we just used the normal install | 23:04 |
| rm_work | hmmmmmm | 23:04 |
| rm_work | though in centos, these package names are wrong so i'm not sure how that worked | 23:04 |
| rm_work | gonna have to poke at it | 23:04 |
| rm_work | k | 23:08 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge amphora-agent elements https://review.openstack.org/518992 | 23:08 |
| rm_work | AUGH almost | 23:11 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge amphora-agent elements https://review.openstack.org/518992 | 23:11 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Enable building gentoo amphora images https://review.openstack.org/518986 | 23:20 |
| rm_work | ok | 23:20 |
| rm_work | now it's a nice chain | 23:20 |
| rm_work | going to need nmagnezi to verify my work, i forget his work schedule again tho | 23:23 |
| johnsom | He is off now, I think Sunday is his next work day | 23:26 |
| rm_work | k | 23:28 |
| openstackgerrit | Merged openstack/octavia master: Merge haproxy-octavia elements https://review.openstack.org/518624 | 23:41 |
| *** AlexeyAbashkin has joined #openstack-lbaas | 23:55 | |
| rm_work | rebase spreeeee | 23:58 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge keepalived elements https://review.openstack.org/518990 | 23:59 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: Merge amphora-agent elements https://review.openstack.org/518992 | 23:59 |
| openstackgerrit | Adam Harwell proposed openstack/octavia master: WIP: Enable building gentoo amphora images https://review.openstack.org/518986 | 23:59 |
| *** AlexeyAbashkin has quit IRC | 23:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!