[00:55] axw: builders etc on 1.8 now. i tried to merge the earlier ec2 error status related PR and fater an intitial bot issue, it's run with some test failures; i haven't looked in detail; a few url parsing things etc [00:56] wallyworld: yeah thanks, I'm merging develop into the feature branch - I think that should fix those failures [00:56] sgtm [00:56] i forgot that was on the fb [00:56] hence it won't have the 1.8 test fixes [01:21] hello, any news about https://bugs.launchpad.net/juju/+bug/1616704 ? [01:21] Bug #1616704: juju restore-backup does not complete [01:25] seyeongkim: no news as yet as far as I know. there's a performance issue there, but no update yet [01:25] thanks wallyworld [01:27] seyeongkim: there's a comment on bug 1680683 which maybe related. i suspect there's a couple of underlying issues that need to be addressed. some work has already been done as indicated in the comment, but there's still more to do [01:27] Bug #1680683: Poor "juju create-backup" performance [01:29] I see wallyworld [02:32] babbageclunk: belated happy birthday :) [02:32] axw: thanks! [07:55] anyone wants to review this small one? https://github.com/juju/juju/pull/7280 [08:30] good morning.. who has experience with the txn pruning? [08:40] jam: you're needed :) [08:40] axw: What about CpuPower? [08:41] 13ae680e4ac environs/manual/detection.go (Andrew Wilkins 2013-08-15 17:21:16 +0800 156) // TODO(axw) calculate CpuPower. What algorithm do we use? [08:41] wpk: I wouldn't worry about it, it's not as widely used as cpu-cores [08:41] (AFAIK) [08:42] ashipika: what's up? [08:43] jam: txn cleanup failed.. error while iterating.. io timeout.. is it safe to retry? [08:43] ashipika: what command were you running? [08:44] it is generally ok, except you can't expect it to actually work the second time if it failed the first [08:44] jam: we're migrating our DB.. and prunning txns.. so a custom command.. using the CleanAndPrune method in juju/txn [08:44] jam: while would it time out, is my question.. [08:45] 'why would it time out' ? [08:45] ashipika: migrating from what to what? [08:45] ashipika: does this happen to be a mongo2.4 source using the CleanAndPrune that is in mgopurge 1.4? [08:46] as there are fixes in juju/txn to handle mongo2.4 better [09:33] uh juju-core autopkgtests fail in artful: https://objectstorage.prodstack4-5.canonical.com/v1/AUTH_77e2ada1e7a84929a74ba3b87153c0ac/autopkgtest-artful/artful/amd64/j/juju-core/20170426_074500_fee63@/log.gz [09:33] my read of that is that there is no jujud in the streams for artful? [09:34] but maybe not [11:24] a rename of some types that were confusing me in juju: https://github.com/juju/juju/pull/7283 [11:24] review appreciated [11:30] jam: fancy taking a look? ^ [11:31] rogpeppe: isn't there a separate 'cmd' package that isn't specific to Juju which is the reason we have CommandBase vs JujuCommandBase ? [11:32] jam: not really [11:32] jam: i mean, there *is* a separate cmd package [11:32] jam: but modelcmd already has CommandBase [11:32] jam: but the "Base" in that name implies something different from the "Base" in all the other type names [11:34] jam: i've been trying to sort the whole picture out in my head, and wrote this summary: http://paste.ubuntu.com/24459695/ [11:34] jam: where CommandBase is an evident anomaly [11:35] jam: this PR is proposing changing that so it looks like: http://paste.ubuntu.com/24459702/ [11:36] jam: FWIW I also think that sysCommandWrapper should probably be controllerCommandWrapper but that's less important [11:52] jam: are you planning to review it or should I try to find someone else? [11:52] sorry, I have to take my son to music for the next hour, it seems reasonably ok, as its just a mechanical change and i don't mind the names, but I haven't thought deeply about it yet [11:54] wallyworld: axw: I've updated https://github.com/juju/juju/pull/7265, PTAL :) would b nice to have it in for b3 \o/ [11:56] jam: ok, thanks [11:56] anastasiamac, wallyworld, axw: any chance of a review of this mechanical rename PR, by any chance? https://github.com/juju/juju/pull/7283 [11:57] rogpeppe: m about to eod :D can b done tomorrow? [11:57] rogpeppe: i mean, it's only 10pm here... but it's been a looong couple of days :) [11:57] anastasiamac: i'd prefer to get it in today, but np [11:58] anastasiamac: "only" 10pm! [11:58] anastasiamac: stop working already! :) [11:58] rogpeppe: :) if noone looks at it today, i'll do it tomorrow \o/ [11:58] anastasiamac: tyvm [11:59] anastasiamac: do we still have on-call-reviewers, BTW? [12:16] jam: thanks a lot [14:17] question concerning the `deploy` command [14:18] is this expected http://paste.ubuntu.com/24460316/ ? [14:20] bdx: I think that it is, if nothing is yet placed on that machine. [14:20] jrwren: ok, thanks [14:41] Q: is it possible to have two API functions with the same name with different arguments/return values for different API versions? If so, is there an example? [14:42] (and if not, where could I find an example of API version-dependent call?) === frankban is now known as frankban|afk [21:03] wpk: ping? [21:55] wallyworld: busy? [21:55] babbageclunk: yeah, in meeting, will ping soon [21:56] wallyworld: cool thanks [22:19] veebers: join release call now? [22:20] anastasiamac: ^^^^^ [22:20] wallyworld: omw [23:34] babbageclunk: sorry, meeting finished and i forgot about you [23:35] wallyworld: hey now! [23:36] did you want to talk? [23:36] yes, please - just plugging some stuff back in. [23:38] righto [23:38] in standup? [23:41] babbageclunk: wallyworldalso forgot about my nice, updated PR just for him :D [23:41] wallyworld also * [23:41] no he didn't [23:41] it's been reviewed [23:41] well, no he did not... :)