IRC logs for #baserock for Saturday, 2015-03-14

*** pacon has quit IRC00:31
*** inara has quit IRC02:26
*** inara has joined #baserock02:34
*** zoli__ has joined #baserock05:55
*** zoli__ has quit IRC07:48
*** zoli__ has joined #baserock07:48
*** zoli__ has joined #baserock07:49
paulsherwoodis it me, or does gerrit.baserock.org not work with openid.baserock.org ?08:09
pedroalvarezIt does work with our open ID09:08
pedroalvarezAre you getting any specific error?09:10
*** pacon has joined #baserock09:20
paulsherwoodi get provider is not supported when i enter http://openid.baserock.org as provider at http://gerrit.baserock.org/login/q/status:open09:51
*** sadcat_ is now known as straycat10:05
*** rdale has quit IRC11:32
pedroalvarezAh, that's because you have to use your personal openid url (e.g. http://openid.baserock.org/openid/pedroalvarez/ )11:47
paulsherwoodurghh :-)11:47
paulsherwoodok11:47
pedroalvarezyou can see yours here: http://openid.baserock.org/11:48
paulsherwoodi think we need some wiki instructions on this... i'll do it if i get time11:48
paulsherwoodthanks pedroalvarez11:48
pedroalvareznp :)11:48
paulsherwoodin other news, thanks to richard ipsum's review, i've confirmed that m4-tarball needs stage2-linux-api-headers, but also it needs stage2-gawk!! no way that's a 'transitive dependency' i think :-)11:49
paulsherwoodstraycat: ^^11:49
SotKIs distbuild failing silently when the controller raises an Exception a known bug?11:54
paulsherwoodi don't know, i'm afraid12:02
straycatSotK, you mean the morph 'client' exits without an error when the controller raises an exception?12:03
SotKstraycat: yep12:04
*** pacon has quit IRC12:04
SotKIt also causes false positives in morph's test suite12:04
straycati don't think that's known12:05
*** pacon has joined #baserock12:05
straycatpaulsherwood, cool :)12:06
*** pacon has joined #baserock12:18
paulsherwood:-)12:24
bjdooksI could relly do with baserock big-endian released12:45
bjdooksZynq works12:45
bjdookstegra works with patches12:45
bjdooksit would be nice to have a baserock be rootfs available12:45
paulsherwood+112:53
pedroalvarezAfter thinking how to install and configure apache-httpd properly, I came to the conclusion that I want support in the chunk morphologies for creating users/groups and changing owners of some files13:10
persiaFor a release, I think it would be wise to restrict ABIs to those for which the project has distbuild clusters.13:12
persiaThe point being that the experience of working without shared cache means rebuilding everything anyway, so it isn't that much less painful than cross-bootstrapping.13:12
* paulsherwood thinks he's managed to submit his patch via gerrit13:13
persiaa3781d3ad5aaeda64b0771914a4e56d1f278465b ?13:14
persiaIf so, I can certainly see it.13:14
paulsherwoodyup :-)13:14
pedroalvarezpaulsherwood: how was your experience13:15
pedroalvarez?13:15
paulsherwoodsome initial confusion on setting up13:15
persiaI believe the prior inference was that everything listed previously in a stratum was available as a build dependency for anything listed later in a stratum.13:16
* straycat just came back to find gcc randomly building itself13:16
persiaIf we're making it explicit, it probably makes sense to remove that feature from morph.13:16
straycatlooks like it could be a connection issue: http://sprunge.us/jTeM something similar happened a couple of days ago13:16
paulsherwoodpersia: that inference is not made, i'm pretty sure13:16
persiastraycat: Indeed.  Did you previously have a cached gcc locally?13:17
paulsherwoodwhen i add foo, then bar, to a stratum, i often forget to add foo as a build-depend for bar, and morph dutifully errors13:17
straycatpersia, no13:18
persiaHrm.  That contradicts what I was told a long time ago, but it was a long time, so I believe it could no longer be true.13:18
persiastraycat: Then yes, entirely a connectivity issue.13:18
paulsherwoodpedroalvarez: also, my patch was made before installed the hook, so i needed to --amend it... but otherwise it worked :-)13:19
persiaGrr.  My password for openid.baserock.org seems to have been reset :/13:19
persiaThat said, I notice the signin page for gerrit.baserock.org doesn't suggest openid.baserock.org.13:20
* paulsherwood notices gerrit has the words '... uploaded patch set 1' so it must deal with patch series somehow?13:21
*** pacon has quit IRC13:21
paulsherwoodpersia: yup, that's odd :)13:21
persiastraycat: Seems there was some maintenance from the hosting organisation today: http://status.datacentred.io/pages/history/53e53a3caf49cecb5f00014e  This may have been the cause of the connectivity issue.13:22
pedroalvarezpersia: you are allowed to use other openid providers13:23
*** edcragg has joined #baserock13:23
pedroalvarezregarding your password in openid.baserock.org, I can't tell :/13:23
persiapedroalvarez: That's a good thing.  Still, some promotion of the Baserock provider seems in order, no?13:23
straycatpersia, hrm ok13:24
persiapedroalvarez: No worries on the password.  I can reset it again.13:24
persiaThat said, in doing so I end up with a page with the title "User test", which seems wrong.13:24
pedroalvarezpersia: agree with adding info about our openid and removing the google one :P13:26
persiaWhy remove the google one?13:37
*** edcragg has quit IRC14:22
paulsherwoodpersia: they're killing it themselves i believe14:31
* paulsherwood discovers that ybd's stage2-gcc output is just *slightly* different from morph's, and screams inside15:45
persiapaulsherwood: Ah, right.  Now I remember reading that.17:01
*** zoli__ has quit IRC17:52
*** zoli__ has joined #baserock17:53
*** zoli__ has joined #baserock17:54
*** zoli__ has quit IRC17:55
*** zoli__ has joined #baserock17:56
*** wdutch__ has quit IRC17:59
*** wdutch has joined #baserock17:59
*** De|ta has quit IRC18:04
*** De|ta has joined #baserock18:04
*** zoli__ has quit IRC20:05
paulsherwoodhttps://developers.google.com/accounts/docs/OpenID220:40
*** zoli__ has joined #baserock21:00
*** zoli__ has quit IRC22:13

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!