| 01:12.04 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 01:12.04 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 01:20.59 | *** part/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 01:44.27 | *** join/#asterisk-dev superscrat (~asanders@173-21-89-217.client.mchsi.com) |
| 02:01.28 | *** join/#asterisk-dev mjordan (mjordan@nat/digium/x-rglfsvbljnjcgzsk) |
| 02:01.28 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 02:01.44 | mjordan | file: check out who did the verify: https://gerrit.asterisk.org/#/c/738/ |
| 02:07.59 | coreyfarrell | congrats! so is jenkins going to end up doing the testsuite as well or is that staying on bamboo? |
| 02:10.07 | coreyfarrell | BTW your experience setting up jenkins makes me want to never try :( |
| 02:23.07 | mjordan | heh |
| 02:23.11 | mjordan | it's been a long road |
| 02:23.17 | mjordan | so yes, the plan is eventually it will replace bamboo |
| 02:23.28 | mjordan | the reason being: you can define jobs/scripts to run in YAML/python/bash |
| 02:23.40 | mjordan | which opens up what gets run and when to everyone |
| 02:24.28 | mjordan | coreyfarrell: uses a combo of http://docs.openstack.org/infra/jenkins-job-builder/ and http://docs.openstack.org/infra/zuul/zuul.html#layout-yaml |
| 02:26.02 | coreyfarrell | am I correct to assume that it has to be run in an openstack cluster? |
| 02:26.06 | mjordan | I will say that jenkins wasn't the worst. Zuul took the cake. |
| 02:26.08 | mjordan | nope! |
| 02:26.15 | mjordan | all of this runs independent of openstack |
| 02:26.26 | mjordan | it's just the openstack project that has been the primary force behind developing these projects |
| 02:26.37 | mjordan | I'm going to write a wiki page documenting all of the stuff |
| 02:26.42 | mjordan | but generally, you have: |
| 02:26.53 | mjordan | Gerrit (code review + git repo). Changes there - of any type are picked up by... |
| 02:27.22 | mjordan | Zuul: organizes jobs. Monitors Gerrit through SSH. Zuul actually consists of two parts: the server, which monitors Gerrit and issues jobs to 'things' using a GearMan server, and ... |
| 02:27.53 | mjordan | Zuul-Merger - this creates another Git repo for build agents to access. It basically grabs the reviews, does speculative merge checks, and creates a repo for build agents to use instead of Gerrit (reducing load on that Git server) |
| 02:28.07 | mjordan | Finally, Jenkins runs the jobs that have been issues to it via Zuul's Gearman server |
| 02:28.41 | mjordan | the sticky point was Zuul, as for whatever reason - probably a networky thingy at our colo - the SSH connection between Zuul and Gerrit was super flakey |
| 02:29.01 | mjordan | so I had to move Zuul from zuul.asterisk.org over to gerrit.asterisk.org, while keeping zuul-merger running on zuul.asterisk.org |
| 02:29.38 | mjordan | the job definitions still have a ways to go. |
| 02:29.58 | mjordan | and we'll want to be careful about what all we make a 'voting' change on Gerrit, as I don't want to annoy people with a lot of -1 verifications due to bouncy tests |
| 02:30.21 | coreyfarrell | it will be major overkill but I'd like to setup gerrit and zuul for my private projects, not sure I have the system resources to do it yet though |
| 02:30.35 | mjordan | you could run it all on one box |
| 02:30.48 | mjordan | I mean, not a 'great idea', but doable |
| 02:31.00 | mjordan | the one nice thing is that the openstack project has published their puppet manifests for all of these |
| 02:31.16 | mjordan | even if you don't use puppet, it helps to outline where things should live, what dependencies you need, etc. |
| 02:31.24 | coreyfarrell | I have a VM server, but it has 8GB ram total, probably 4GB is the max I could allocate to this |
| 02:31.31 | mjordan | it would work |
| 02:31.44 | mjordan | HD space is sometimes more important, given how many freakin git repos it creates |
| 02:33.40 | coreyfarrell | right now I just use gitolite which is good enough to host git, but the ability to follow the full process would be useful to me, even though it'll just be me reviewing my own pull requests the next day |
| 02:39.32 | mjordan | yeah |
| 02:39.48 | mjordan | generally, I've used gitolite (which is a minimal system); GitHub; and Stash |
| 02:40.07 | mjordan | and Gerrit - despite the somewhat clunky UI - has my favorite workflow |
| 02:40.31 | mjordan | but the setup of all the pieces has been non-trivial. If I had to do it over again, knowing the work involved, I might have just gone with Stash + Bamboo. |
| 02:40.52 | mjordan | The only good thing about it is that Bamboo hasn't stopped working with Gerrit, so that gave us some breathing room to try and figure out what was going on with the rest of the pieces |
| 02:41.34 | coreyfarrell | my biggest complaint against gerrit is that the web session times out after less than a day |
| 02:42.10 | mjordan | heh |
| 02:42.34 | mjordan | suspects there's a parameter for that somewhere |
| 02:42.37 | coreyfarrell | the way it handles it is annoying too |
| 03:40.56 | coreyfarrell | mjordan: I keep forgetting to ask, is it possible to tell bamboo to keep the reference debug artifacts for 7 days? right now the artifacts are only available for a couple days a week. |
| 04:15.41 | *** join/#asterisk-dev doome_ (~doome@94-21-238-35.pool.digikabel.hu) |
| 05:42.01 | *** join/#asterisk-dev superscrat (~asanders@173-21-89-217.client.mchsi.com) |
| 06:02.05 | *** join/#asterisk-dev doome_ (~doome@94-21-238-35.pool.digikabel.hu) |
| 07:51.35 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.dynamic.se.alltele.net) |
| 07:51.36 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 09:25.04 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.dynamic.se.alltele.net) |
| 09:25.04 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 09:56.44 | *** join/#asterisk-dev pchero (~pchero@0x555140b5.adsl.cybercity.dk) |
| 09:58.01 | *** join/#asterisk-dev Stefan26 (~Stefan0@h105n2-s-oe-a15.ias.bredband.telia.com) |
| 10:25.36 | *** join/#asterisk-dev tzafrir (~tzafrir@bzq-179-40-172.cust.bezeqint.net) |
| 10:25.36 | *** mode/#asterisk-dev [+o tzafrir] by ChanServ |
| 10:38.41 | *** join/#asterisk-dev doome_ (~doome@94-21-238-35.pool.digikabel.hu) |
| 14:02.58 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 14:02.59 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 14:29.34 | *** join/#asterisk-dev [TK]D-Fender (~joe@64.235.216.2) |
| 14:32.52 | coreyfarrell | mjordan: missed you last night, can you tell bamboo to keep the reference debug job artifacts for 7 days? |
| 14:33.10 | *** join/#asterisk-dev oej (~oej@h87-96-134-129.dynamic.se.alltele.net) |
| 14:33.10 | *** mode/#asterisk-dev [+o oej] by ChanServ |
| 15:13.18 | mjordan | coreyfarrell: I'm not sure if the setting for that is plan specific; I'll have to check |
| 15:13.44 | mjordan | The reason we cranked it down to where it was was due to running out of disk space in various places |
| 15:13.55 | mjordan | I'm pretty sure it's a global setting |
| 15:15.00 | mjordan | yeah, it's global |
| 15:15.04 | mjordan | I can bump it up a bit and see what happens |
| 15:15.34 | coreyfarrell | I'd say don't risk bringing bamboo down |
| 15:17.10 | mjordan | well, we set those policies awhile back, and some of the plans and things have changed since then |
| 15:17.20 | mjordan | I'll tell it to keep it for 5 days, as well as the last build result |
| 15:17.23 | mjordan | and we'll see what happens |
| 15:17.35 | mjordan | I'll watch it over the next few days, and if it looks like we're tanking, I'll push it back down |
| 15:17.41 | coreyfarrell | last build result? |
| 15:19.35 | coreyfarrell | if you could tell it to retain artifacts from the most current test regardless of age - that would solve it. the problem is that the test only runs every 7 days but the logs are deleted after like 2 days. |
| 15:44.23 | *** join/#asterisk-dev mjordan (mjordan@nat/digium/x-xanzktecezjpwinb) |
| 15:44.23 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |
| 16:06.03 | *** part/#asterisk-dev mjordan (mjordan@nat/digium/x-xanzktecezjpwinb) |
| 16:40.47 | *** join/#asterisk-dev Stefan26 (Stefan0@h105n2-s-oe-a15.ias.bredband.telia.com) |
| 18:55.16 | *** join/#asterisk-dev pchero (~pchero@0x555140b5.adsl.cybercity.dk) |
| 22:50.59 | *** join/#asterisk-dev mjordan (~mjordan@75.76.55.191) |
| 22:50.59 | *** mode/#asterisk-dev [+o mjordan] by ChanServ |