*** edcragg has joined #baserock | 05:46 | |
*** paulw has joined #baserock | 06:30 | |
*** edcragg has quit IRC | 06:30 | |
*** tiagogomes has joined #baserock | 07:17 | |
*** paulw has quit IRC | 07:25 | |
*** paulw has joined #baserock | 07:25 | |
*** pedroalvarez has quit IRC | 07:41 | |
*** cyndis has quit IRC | 07:41 | |
*** kejiahu has quit IRC | 07:41 | |
*** ctgriffiths has quit IRC | 07:41 | |
*** pedroalvarez has joined #baserock | 07:42 | |
*** cyndis has joined #baserock | 07:42 | |
*** kejiahu has joined #baserock | 07:42 | |
*** ctgriffiths has joined #baserock | 07:42 | |
*** hitchcock.freenode.net sets mode: +v pedroalvarez | 07:42 | |
*** kejiahu has quit IRC | 07:45 | |
*** cyndis has quit IRC | 07:46 | |
*** pedroalvarez has quit IRC | 07:46 | |
*** ctgriffiths has quit IRC | 07:46 | |
*** pedroalvarez has joined #baserock | 07:47 | |
*** ChanServ sets mode: +v pedroalvarez | 07:47 | |
*** jonathanmaw has joined #baserock | 07:47 | |
*** ctgriffiths has joined #baserock | 07:48 | |
*** kejiahu has joined #baserock | 07:53 | |
*** CTtpollard has quit IRC | 07:58 | |
*** cyndis has joined #baserock | 07:59 | |
*** CTtpollard has joined #baserock | 08:04 | |
*** bashrc has joined #baserock | 08:11 | |
*** mdunford has joined #baserock | 09:21 | |
*** brlogger has joined #baserock | 09:24 | |
*** paulw has quit IRC | 09:53 | |
*** paulw has joined #baserock | 09:57 | |
pedroalvarez | SotK: :( | 10:42 |
---|---|---|
paulsherwood | does anyone here have recommendations for how to do tagging in gerrit? | 10:42 |
pedroalvarez | I know someone tried to integrate it, (ctgriffiths?) but I think it wasn't ready | 10:42 |
pedroalvarez | paulsherwood: have you already tried to tag anything? | 10:44 |
paulsherwood | no, i'm trying to understand 'the rules'. this is for another project, to be fair | 10:44 |
paulsherwood | but given we have lots of gerrit experience here i thought i might be able to get guidance | 10:45 |
pedroalvarez | I *think* that for creating tags, you only need the right permissions in gerrit | 10:46 |
pedroalvarez | If one has those permissions, he will be able to push tags, using git | 10:47 |
pedroalvarez | I don't think you can create tags using the UI, but I may be wrong | 10:48 |
SotK | pedroalvarez: :( indeed | 10:50 |
*** paulw has quit IRC | 10:50 | |
SotK | I don't know what will be the best way to split it either | 10:51 |
pedroalvarez | I didn't expect storyboard to have common dependencies with OpenStack strata | 10:52 |
SotK | I worry about ending up with really fragmented definitions | 10:52 |
SotK | it depends on stuff like SQLAlchemy and Pecan | 10:52 |
SotK | plus a bunch of oslo stuff | 10:53 |
* SotK finds the list | 10:53 | |
SotK | http://git.openstack.org/cgit/openstack-infra/storyboard/tree/requirements.txt | 10:53 |
pedroalvarez | that list is nothing after dealing with all the requirements of openstack :) | 10:54 |
SotK | :) | 10:54 |
* SotK also wonders what to do with the webclient, which depends on a bunch of npm packages, some of which we already have the xstatic equivalent of in git.b.o | 10:55 | |
*** mdunford has quit IRC | 10:56 | |
pedroalvarez | some versions of oslo components conflict with the requirements of kilo | 10:59 |
pedroalvarez | or the version of kilo that we have integrated now | 10:59 |
*** paulw has joined #baserock | 11:00 | |
pedroalvarez | e.g: oslo.db>=1.7.0,<1.8.0 for Kilo, oslo.db>=1.8.0 for Storyboard | 11:02 |
* pedroalvarez has a list of all the requirements for Kilo | 11:03 | |
SotK | oh hm :/ | 11:03 |
paulsherwood | moving forward is usually least pain | 11:04 |
paulsherwood | imo | 11:04 |
paulsherwood | presumably pstream would be happy for folks to report Kilo working (or how to fix to work) with 1.8.0 or later | 11:04 |
pedroalvarez | from my experience, in OpenStack that's not the case | 11:05 |
*** paulw has quit IRC | 11:06 | |
pedroalvarez | They know that the libraries are going to change in some version, and that's why they put the '<' there | 11:07 |
pedroalvarez | Actually they only do that with the libraries that they control, like oslo* keystonemiddleware, python-{swift|ceilometer|heat|...}*clients | 11:09 |
pedroalvarez | Also, fixing upstream in this context would be fixing a lot of code | 11:10 |
*** paulw has joined #baserock | 11:10 | |
pedroalvarez | Maybe we could just move to OpenStack Liberty | 11:10 |
*** mdunford has joined #baserock | 11:10 | |
pedroalvarez | meh, the extensions/strip-gplv3.configure doesn't work anymore | 11:24 |
paulsherwood | erk | 11:24 |
pedroalvarez | It's not stripping anything, and then it fails with: http://paste.baserock.org/efotobasoh | 11:24 |
pedroalvarez | I'll try to fix this | 11:25 |
pedroalvarez | I see, this might be failing since we dropped cliapp from the extensions | 11:49 |
pedroalvarez | but why Mason is able to deploy it? this doesn't make any sense :( | 11:53 |
tiagogomes | no, it makes sense, you still just didn't crack it | 11:56 |
pedroalvarez | :( | 11:58 |
pedroalvarez | right, fixed | 12:06 |
tiagogomes | cool, so what didn't make sense and now makes sense? | 12:11 |
pedroalvarez | still doesn't make sense | 12:12 |
pedroalvarez | Mason and I are using different versions of morph though | 12:12 |
pedroalvarez | I found that the regular expresion for finding the .meta files didn't work for subprocess, but I guess it did work for cliapp runcmd | 12:14 |
pedroalvarez | I just had to remove the escaping characters '\' | 12:15 |
* richard_maw raises eyebrow | 12:15 | |
richard_maw | I'd be more inclined to believe it's a difference between GNU find and busybox find… since neither do anything more to the path than pass it directly on to the find command | 12:17 |
pedroalvarez | is it possible that subprocess scapes them? | 12:17 |
richard_maw | pedroalvarez: only on windows, where their exec syscall equivalent takes a single string, rather than a list of arguments | 12:18 |
* pedroalvarez tries with busybox find | 12:20 | |
pedroalvarez | I want to believe that subprocess is escaping '\' making the regex to fail | 12:31 |
pedroalvarez | subprocess.check_output(['find', '/baserock/', '-regex', "/baserock/gcc-[^-]\+\.meta"]) | 12:39 |
pedroalvarez | That doesn't show anything in my system | 12:39 |
pedroalvarez | oh, no that works in Mason, that explains why it works there | 12:42 |
richard_maw | if I remove the \ it doesn't work on my system, but if I put them it, it works | 12:42 |
pedroalvarez | I guess I failed to check busysbox find | 12:43 |
richard_maw | I'm using busybox find btw. | 12:43 |
*** tiagogomes has quit IRC | 12:43 | |
pedroalvarez | Mason only has busybox find too | 12:43 |
richard_maw | ok, so the difference appears to be whether it uses extended, or basic regexes | 12:44 |
richard_maw | find uses basic btw | 12:44 |
richard_maw | hmm, -regextype | 12:44 |
pedroalvarez | ok! I was pondering forcing the use of busybox find | 12:45 |
richard_maw | bah, busybox find doesn't recognise regextype | 12:45 |
richard_maw | ah, hm, if it has both, then that ought to work | 12:46 |
richard_maw | argh! "The POSIXLY_CORRECT environment variable does not affect the behaviour of the -regex or -iregex tests because those tests aren't specified in the POSIX standard." | 12:48 |
pedroalvarez | Using posix-basic fixes it, but it blows up if the system doesn't have GNU find | 12:50 |
richard_maw | yep, which is why I was annoyed that it wasn't specified in POSIXLY_CORRECT | 12:50 |
richard_maw | since then you could set it that way for GNU find, and leave busybox find at the default | 12:50 |
pedroalvarez | I see | 12:51 |
richard_maw | so assuming find is still built into busybox, running `busybox find`, or doing the traversal in-python are your best choices | 12:51 |
pedroalvarez | richard_maw: thanks for raising that eyebrow and helping | 12:52 |
richard_maw | ☺ | 12:52 |
*** paulw has quit IRC | 13:04 | |
perryl | i'm currently having issues with setting up wifi on baserock; wlan0 appears to be up but not running; can anyone assist? | 13:07 |
*** paulw has joined #baserock | 13:10 | |
pdar | perryl: have you had any luck with your wireless? | 13:23 |
perryl | pdar: yes, was just about to say i've sorted the issue :) | 13:23 |
pdar | awesome! What was the problem? | 13:24 |
*** tiagogomes has joined #baserock | 13:29 | |
*** paulw has quit IRC | 13:30 | |
pedroalvarez | I've never done that before | 13:36 |
perryl | the issue i was having was no output running "iw dev wlan0 scan" but it turns out i needed to run "iwconfig wlan0 up" first... | 13:37 |
* richard_maw keeps forgetting that too when he has to set up interfaces manually | 13:39 | |
*** paulw has joined #baserock | 13:59 | |
*** petefoth has joined #baserock | 14:07 | |
pdar | tiagogomes: i was just testing your staging area patches | 14:42 |
pdar | the lock files for successfully built things seem to remain after their staging area is removed | 14:43 |
pdar | is this intended? | 14:46 |
richard_maw | I don't believe it is, though I understood the v2 of those patches used the directory rather than a lock file | 14:47 |
tiagogomes | pdar not it is not intended, but you seem to be testing an old version. The current version of the patch is v3 | 14:48 |
tiagogomes | s/not it/not it | 14:48 |
pdar | tiagogomes: oh yes, ti seems im testing stuff from the 3rd of august, whoops | 14:59 |
pdar | sorry | 14:59 |
* pedroalvarez sends a couple of patches Genivi-related | 15:00 | |
pedroalvarez | https://gerrit.baserock.org/1040 | 15:00 |
pedroalvarez | https://gerrit.baserock.org/1041 | 15:00 |
richard_maw | pedroalvarez: voted | 15:03 |
pedroalvarez | thanks | 15:04 |
*** paulw has quit IRC | 15:21 | |
*** paulw has joined #baserock | 15:22 | |
*** fay_ has quit IRC | 16:32 | |
*** mdunford has quit IRC | 16:44 | |
*** paulw has quit IRC | 16:50 | |
*** jonathanmaw has quit IRC | 17:01 | |
*** mdunford has joined #baserock | 17:47 | |
*** mdunford has quit IRC | 17:52 | |
*** rdale has quit IRC | 19:39 | |
*** paulw has joined #baserock | 21:14 | |
*** paulw has quit IRC | 21:24 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!