03:59:44  * piccoloaiutantequit (Remote host closed the connection)
09:57:37  * seishunjoined
10:25:06  * mylesborinsquit (Quit: farewell for now)
10:25:37  * mylesborinsjoined
14:26:20  * node-ghjoined
14:26:20  * node-ghpart
14:29:00  * tniessenjoined
14:30:33  <tniessen>CI is acting weird again: https://github.com/nodejs/build/issues/761
14:32:17  * node-ghjoined
14:32:17  * node-ghpart
14:43:57  * node-ghjoined
14:43:58  * node-ghpart
15:05:27  <seishun>I see that issue got closed, but I still can't run a node-test-pull-request job
15:06:30  <refack>It was closed for that specific time (then it was that master run out of space), now we just have a very long backlog
15:09:37  <seishun>where can I see the backlog? it looks like my job gets cancelled immediately
15:10:47  <refack>https://ci.nodejs.org/job/node-test-commit/build?delay=0sec I don't think they are cancelled, it's just they dodn't even start
15:11:35  <refack>https://usercontent.irccloud-cdn.com/file/cJ4H9GcA/image.png
15:11:56  <refack>it says `pending` not `cancelled` and still has the stop button
15:14:24  <seishun>alright
15:15:30  <tniessen>Guess the C&L PRs were too many :)
15:26:37  <seishun>can't say I'm too happy about being held up by a bunch of trivial changes
15:31:45  <tniessen>same here...
18:09:48  <seishun>about node-test-commit-linuxone: I don't suppose this is something I could test locally? cc joaocgreis
18:11:07  <refack>You can get free linuxone access for 120 days https://developer.ibm.com/linuxone/
18:17:17  <seishun>oh cool. next question is how they are deployed. rhel72 isn't listed here https://github.com/nodejs/build/blob/master/ansible/roles/baselayout/vars/main.yml so I'd assume something under setup/?
18:18:00  <refack>That's probably IBM black magic
18:30:41  <seishun>also, shouldn't we retire the centos5 boxes? it's EOL since 31 March 2017
18:34:41  <refack>https://github.com/nodejs/build/issues/718 the decision is to keep testing on EOL platforms for LTS versions just so they don't break
18:34:54  <refack>but they are "not supported"
18:40:03  <seishun>I don't see the decision about testing, only about building LTS node
19:43:27  * evanlucasquit (Remote host closed the connection)
19:54:32  <seishun>joaocgreis: how is centos6 deployed? I get `fatal: [test-softlayer-centos6-x64-1]: FAILED! => {"changed": false, "failed": true, "msg": "Aborting, target uses selinux but python bindings (libselinux-python) aren't installed!"}` at [baselayout : disable sftp]
19:58:47  <seishun>after installing libselinux-python manually, I get `fatal: [test-softlayer-centos6-x64-1]: FAILED! => {"changed": false, "failed": true, "msg": "missing required arguments: name"}` at [baselayout : repo : add scl devtoolset]
20:24:09  * node-ghjoined
20:24:09  * node-ghpart
21:09:31  * node-ghjoined
21:09:31  * node-ghpart
21:11:01  <Trott>tniessen and seishun: The timing of our CI speed woes is unfortunate given C + L. (V8 snapshots, we miss you!) I'll definitely go and cancel a bunch of jobs if they're still holding up things if no one beats me to it. (I'll probably get to it in about 20 minutes or less.)
21:11:34  <tniessen>Trott: I think Vse has already done some things in that area, not sure though
21:16:41  * evanlucasjoined
21:25:46  <Trott>Since the hold up is really just Raspberry Pi 1 testing, I'm going to cancel the Pi 1 tests on C + L PRs where everything else has already run. That ought to fix our backlog CI issue, I think.
21:36:51  <Trott>Canceled a few. At least the scary-looking pending message is gone on the left hand side. Will try to clean up a bit more. I see one of seishun's jobs failed spectacularly due to infra/build issues and will need to be kicked off. I think it's the middle of the night in their time zone, so it would be nice if we had it re-queued and finished before their morning. :-D
21:37:47  <seishun>the failure was expected, I ran it to see which machines have old gcc versions
21:40:31  <Trott>Ah, yeah, I'm looking now and seeing lambda this and that.
21:40:35  <Trott>Which is what the PR is about.
21:40:38  <Trott>So, never mind. :-D
21:51:14  * lanceballquit (Excess Flood)
21:51:39  * lanceballjoined
22:09:17  * node-ghjoined
22:09:17  * node-ghpart
22:14:31  * seishunquit (Ping timeout: 255 seconds)
22:26:37  * piccoloaiutantejoined
22:39:48  <refack>So Trott how would you sum it up? From PR quantity and percent of follow ups I'd say huge success
22:41:42  <Trott>Gotta pack, check out, get breakfast, and head to the airport, but I'd say it went really well. I over-estimated potential communication issues and under-estimated how prepared a lot of attendees were going to be (although as always, it totally ran the gamut--some people, we were happy to just have all their compile issues worked out after 4 hours).
22:42:43  <Trott>For me personally, it was great to meet some folks I only ever interact with online. (Did you know Jackson Tian has two unbelievably adorable daughters?! He brought them.)
22:44:45  <Trott>And it was a real group effort. Joyee Cheung and Timothy Gu were especially valuable. (And not just because they're both fluent in both Chinese and English, although that was great because it meant they smoothed over all communication issues.) .
22:44:47  <Trott>I'm glad Bryan English came because he dealt with Windows issues, and it's never good when someone has Windows problems and we're like, "Oh, I have no idea what's going on with Windows."
22:45:17  <Trott>And Anna as always is crucial in a million ways, large and small.
22:45:18  <Trott>Oh, sorry, wrong channel. I thought I was in node-dev. Sorry, build folks!
22:46:03  <Trott>Anyway, the attendees were great. Enthusiastic, largely prepared, etc. etc. Everyone wanted a photo with us too, which was fun.