IRC logs for #baserock for Friday, 2015-07-24

*** zoli__ has joined #baserock05:23
*** zoli__ has quit IRC05:24
*** zoli__ has joined #baserock05:24
*** paulw has joined #baserock06:01
*** mariaderidder has joined #baserock07:47
*** bashrc has joined #baserock08:06
*** mdunford has joined #baserock08:07
*** jonathanmaw has joined #baserock08:23
*** ssam2 has joined #baserock08:42
*** ChanServ sets mode: +v ssam208:42
*** edcragg has joined #baserock08:53
*** a1exhughe5 has joined #baserock09:41
pedroalvarezhrf! this busybox has been linked agains libpcre.so.309:45
pedroalvarezwhich is not present in my system..09:45
rjekoops09:45
ssam2split rules, or missing stratum in the system .morph?09:45
pedroalvarezbut but..09:46
ssam2lack of automated runtime dependency tracking makes this sort of thing common when constructing new systems09:46
pedroalvarezthis is busybox in build essential09:46
ssam2hmm, true09:46
*** mdunford has quit IRC09:51
paulsherwoodssam2: this build-system thing....09:53
*** sambishop has quit IRC09:54
ssam2go on09:54
paulsherwoodnote that adding build-system: in itself would be a simple non-invasive change.09:54
paulsherwoodit's calling it a new VERSION that creates the friction09:54
paulsherwoodboth ybd and morph would be happy with the new definitions, if you don't change the version number09:55
ssam2isn't VERSION just a tool for indicating incompatible changes?09:55
paulsherwoodthis change is not incompatible09:55
ssam2adding build-system: doesn't gain anything09:55
paulsherwoodi think it does?09:55
ssam2*requiring* build-system is the interesting part09:55
paulsherwoodyup09:55
paulsherwoodbut adding does gain something, even so09:56
paulsherwood(it's tidier, and makes it possible to reason about those definitions without probing to autodetect)09:56
ssam2yes, I guess we could try to introduce the change as an optional one09:57
ssam2the complexity involved in taking that approach long-term gives me nightmares, though09:57
ssam2trying to track which features of which tools work on definitions which do which things09:57
ssam2we can say "you have to specify build-system now, and all the tools work" (which is what I proposed)09:58
*** sambishop has joined #baserock09:58
ssam2or we can say "you can specify build-system for everything now, and if you don't, some tools may not work for you in future"09:58
ssam2I guess you prefer the latter?09:59
paulsherwoodthat seems fine  to me :)09:59
ssam2I really think this is the shallow imprint of what will become a massive hole. But I'll think on it09:59
paulsherwoodbut note also i'm managing to avoid most of the complexity you're worried about so far in ybd... and i think i can keep it working09:59
paulsherwoodyou may be right, i may be wrong09:59
ssam2I think it'd be interesting to try the latter approach in YBD10:00
ssam2I may also be wrong, of course :)10:00
* paulsherwood has a meetging10:00
ssam2what do others in this channel think?10:02
tiagogomes_I just don't like this approach of doing small changes and increasing VERSION. I'd had prefer that we though how definitions should be, did the changes and increased the VERSION in one jump. I realize that improving definitions is a iterative process, but we start this version thingy recently and what you propose is already the 6th version! When you bring default split rules to definitions are you planning to increase VERSION10:06
tiagogomes_ again?10:06
ssam2yes, to 710:08
*** mdunford has joined #baserock10:09
ssam2I would prefer to do all the changes in one go, but I think that in practice we'd just discuss things forever and never have the resources to actually do them10:09
ssam2is your problem that there might be a long chain of upgrades that people need to do to catch up with 'master' ?10:11
ssam2or something else?10:11
ssam2turns out that building the baserock-15.02 tag in a Baserock 15.25 reference system doesn't work10:14
pedroalvarezssam2: stage1-gcc?10:14
ssam2with either Morph or YBD. Seems to be that it's trying and failing to build documentation10:14
ssam2stage1-gcc10:15
tiagogomes_catching up with master shouldn't be a problem as it will be done with the migration scripts. Is more like I'd like to see a plan for the definitions, an implementation, and a jump in the version. Not every change, a jump in the version10:15
*** mdunford has quit IRC10:16
*** rdale has joined #baserock10:17
ssam2tiagogomes_: I think that's fair10:17
ssam2tiagogomes_: but I think making such a plan is hard. I don't know exactly I want from them yet, beyond detangling them from Morph's code10:18
pdarwhats the current version of baserock provided on the downloads page of g.b.o?10:19
pedroalvarezlatest :P10:19
ssam215.2510:19
pedroalvarezyes, that one10:19
pdarta ssam210:20
pdarand pedroalvarez10:20
tiagogomes_ssam2 detangling them from Morph's code sounds like a plan to me :) So the next jump in definitions could be doing precisely that, moving split rules to definitions and adding build-system to the strata10:20
tiagogomes_ssam2 also, shoudn't build-system be removed from the chunk morph and moved to the strata as well? It is odd to see it in both strata and chunk morphs10:23
ssam2tiagogomes_: I agree it's a little strange.. but I think it would be worse to have configure-commands, build-commands etc. in one place and build-system in another10:24
*** rdale has quit IRC10:30
*** rdale has joined #baserock10:31
tiagogomes_I am not sure on that one10:40
pedroalvarezuh10:55
pedroalvarezif something fails to build sometimes, should I max-jobs=1?10:55
pedroalvarezI had this one: http://paste.baserock.org/zugipilaze10:56
pedroalvarezdidn't know why10:56
pedroalvarezand then it just worked10:56
ssam2yeah, could be that their Makefile logic is a broken when running in parallel10:58
*** ssam2 has quit IRC11:03
*** tiagogomes_ has quit IRC11:04
*** tiagogomes_ has joined #baserock11:04
*** franred has joined #baserock11:11
*** rdale has quit IRC11:17
*** franred has quit IRC11:19
*** rdale has joined #baserock11:19
pdarWas baserock 15.25 prior sams patches getting rid of workspaces?11:23
nowsterwhen uploading a new patch to a gerrit ref, is it sufficient to send a diff against the existing patch, or should the whole thing be merged down into a single patchset again?11:32
*** mdunford has joined #baserock11:33
pedroalvarezpdar: probably11:37
tiagogomes_nowster single patch-set again with same change-id11:38
pedroalvareznowster: I'd prefer the fixes merged in one patch-set11:38
pedroalvarezbut I've seen recently a +2 in a patch series that had fixes in later commits11:38
nowsterso rebase -i and "squash commits"?11:39
pedroalvarezyup, making sure you keep the change-id of the original patch-set11:40
nowsterlet's see what it does11:40
nowster...fixup11:40
nowsterit seems to have kept it11:41
pdarAre there any instructions on w.b.o regarding upgrading a system ina chroot11:49
pedroalvarezI believe that upgrading one is not possible11:50
pdaralas11:50
pedroalvarezonly deploying a new one and using it11:50
pedroalvarezyou can upgrade morph only if that's what you are looking for11:51
tiagogomes_is not possible to update a chroot system using the ssh+rsync extension, using localhost?11:53
tiagogomes_But I think it is easier to deploy a new one and make it use the same /src of the older one11:54
pedroalvareztiagogomes_: nope, the upgrades system depends on the baserock btrfs layout11:54
tiagogomes_right11:56
pdarthanks pedroalvarez and tiagogomes_, I shall do a new one and switch the src11:57
*** franred has joined #baserock12:06
*** zoli__ has quit IRC12:59
*** zoli__ has joined #baserock13:02
*** zoli__ has quit IRC13:04
*** zoli__ has joined #baserock13:04
*** zoli__ has quit IRC13:06
*** zoli__ has joined #baserock13:12
*** zoli___ has joined #baserock13:13
*** zoli__ has quit IRC13:13
*** zoli___ has quit IRC13:13
*** zoli__ has joined #baserock13:17
*** zoli__ has quit IRC13:20
*** zoli__ has joined #baserock13:26
*** zoli___ has joined #baserock13:27
*** zoli__ has quit IRC13:27
*** zoli___ has quit IRC13:29
*** zoli__ has joined #baserock13:29
*** zoli__ has quit IRC13:31
*** zoli__ has joined #baserock13:34
*** zoli__ has joined #baserock13:41
*** paulw has quit IRC13:42
*** zoli__ has quit IRC13:42
*** zoli__ has joined #baserock13:43
*** zoli__ has quit IRC13:45
*** zoli__ has joined #baserock13:47
*** zoli__ has quit IRC13:49
*** zoli__ has joined #baserock13:52
*** zoli__ has quit IRC13:54
*** zoli__ has joined #baserock13:55
*** zoli___ has joined #baserock13:56
*** zoli__ has quit IRC13:56
*** zoli___ has quit IRC13:58
*** zoli__ has joined #baserock13:59
*** zoli__ has joined #baserock14:01
*** mdunford has quit IRC14:02
*** zoli__ has quit IRC14:02
*** zoli__ has joined #baserock14:02
*** zoli__ has quit IRC14:04
*** zoli__ has joined #baserock14:05
*** zoli__ has quit IRC14:05
*** zoli__ has joined #baserock14:08
*** zoli__ has quit IRC14:09
*** zoli__ has joined #baserock14:09
*** zoli__ has quit IRC14:11
*** zoli__ has joined #baserock14:14
*** zoli__ has quit IRC14:16
pedroalvarezextlinux question of today: why, when creating a baserock ext4 disk image, with one partition, extlinux doesn't just work?14:16
*** mdunford has joined #baserock14:16
pedroalvarezI have to "dd if=/usr/lib/syslinux/mbr.bin of=$disk` to make it work14:17
pedroalvarezahh14:18
pedroalvarezit's on the partition14:18
pedroalvarezuseful command of the day: `xxd -l512 $disk`14:19
*** zoli__ has joined #baserock14:21
*** zoli__ has quit IRC14:22
*** zoli__ has joined #baserock14:22
*** zoli___ has joined #baserock14:23
*** zoli__ has quit IRC14:23
*** zoli___ has quit IRC14:24
*** zoli__ has joined #baserock14:25
*** zoli__ has quit IRC14:27
*** zoli___ has joined #baserock14:27
*** zoli__ has joined #baserock14:33
*** zoli__ has quit IRC14:35
*** zoli__ has joined #baserock14:35
*** zoli__ has quit IRC14:37
*** zoli__ has joined #baserock14:40
*** zoli__ has quit IRC14:42
*** zoli__ has joined #baserock14:44
*** zoli__ has quit IRC14:46
*** zoli__ has joined #baserock14:46
*** zoli___ has joined #baserock14:47
*** zoli___ has quit IRC14:48
*** zoli__ has joined #baserock14:48
*** zoli__ has quit IRC14:51
*** zoli__ has joined #baserock14:53
*** zoli___ has joined #baserock14:53
*** zoli___ has quit IRC14:54
*** zoli__ has joined #baserock14:57
*** zoli__ has joined #baserock15:07
*** zoli___ has joined #baserock15:10
*** zoli__ has quit IRC15:12
*** zoli___ has quit IRC15:12
*** zoli__ has joined #baserock15:12
*** zoli___ has joined #baserock15:14
*** zoli__ has quit IRC15:14
*** zoli___ has quit IRC15:16
*** zoli__ has joined #baserock15:16
*** zoli__ has quit IRC15:18
*** zoli__ has joined #baserock15:18
*** zoli__ has quit IRC15:20
*** zoli__ has joined #baserock15:21
*** zoli__ has quit IRC15:22
*** zoli___ has joined #baserock15:22
*** zoli__ has joined #baserock15:23
*** zoli___ has joined #baserock15:23
*** zoli__ has quit IRC15:23
*** zoli___ has quit IRC15:24
*** zoli__ has joined #baserock15:25
*** zoli___ has joined #baserock15:26
*** zoli__ has quit IRC15:26
*** zoli___ has quit IRC15:27
*** mariaderidder has quit IRC15:27
*** zoli__ has joined #baserock15:30
*** zoli___ has joined #baserock15:31
*** zoli__ has joined #baserock15:31
*** zoli__ has quit IRC15:34
*** zoli__ has joined #baserock15:38
*** zoli__ has joined #baserock15:42
*** mariaderidder has joined #baserock15:42
*** zoli__ has quit IRC15:42
*** jonathanmaw has quit IRC15:48
*** CTtpollard has quit IRC15:50
*** zoli__ has joined #baserock15:51
*** zoli__ has joined #baserock15:54
*** zoli__ has quit IRC15:56
*** zoli___ has joined #baserock15:56
*** zoli___ has quit IRC16:01
*** franred has quit IRC16:03
*** rdale has quit IRC16:04
*** zoli__ has joined #baserock16:05
*** a1exhughe5 has quit IRC16:07
*** zoli___ has joined #baserock16:17
*** zoli__ has quit IRC16:20
*** mariaderidder has quit IRC16:30
*** zoli___ has quit IRC16:47
*** zoli__ has joined #baserock16:47
*** bashrc has quit IRC16:49
*** sambishop has quit IRC16:57
*** zoli__ has quit IRC16:58
*** zoli__ has joined #baserock16:58
*** gary_perkins has quit IRC17:01
*** zoli__ has quit IRC17:07
*** mdunford has quit IRC17:07
*** tiagogomes_ has quit IRC17:37
*** tiagogomes has joined #baserock17:37
*** edcragg has quit IRC17:44

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