=== allenap_ is now known as allenap === ming is now known as Guest41611 [03:22] smoser: When I get my filesystem all set up, where do I write the fstab file? === Guest11359 is now known as basicer [03:23] ie where does state['fstab'] point ? === vladk|offline is now known as vladk === CyberJacob|Away is now known as CyberJacob === basicer is now known as Guest21557 === CyberJacob is now known as CyberJacob|Away [11:11] gmb: as mentioned on the call yesterday, I've been continuing the cleanup on the import code. [11:11] Or did I mention it the day before? One forgets. [11:11] Anyway, I have a branch in the works that may clarify a few things that you've been figuring out. [11:24] hmm, with trusty, i try to perform the same sort of install i used to do before... I can bootstrap all right, but deployn a simple charm such as mysql fail. no machine is being picked up [11:24] http://pastebin.ubuntu.com/7329138/ [11:30] melmoth: looks like it's more of a juju problem than a maas problem. It _may_ turn out to be a maas problem, but you may be best off asking in #juju about finding the errors' root cause. [11:30] hmmm, i wonder if it s a good idea to say "local:precise/mysql" when i m using default trusty [11:31] i ll try to deploy stuff drom local:trusty/ and ask on #juju if it still fail. [11:31] Ah, I hadn't noticed the "precise." :) [11:36] allenap, could you help me debug a problem building in the daily PPA? [11:36] I requested a daily/trusty build, and it failed to upload because a package with the same version but different contents was already in the PPA. [11:37] This has happened before, but I don't recall why. The revision should contain the current revision of the packaging branch, right? [11:37] Correction: The *version* should contain the current revision of the packaging branch, right? [11:39] Hmm... it built from packaging r272, but my packaging branch had landed as r273. Is this an 1.5/1.6 problem? [11:53] allenap: you might want to have a look at the comment I added on bug 1305102 https://bugs.launchpad.net/maas/+bug/1305102/comments/5 [11:53] Launchpad bug 1305102 in MAAS "celeryd 100% cpu when large dhcpd lease file" [Critical,Triaged] [11:56] rvba: maybe you have any ideas about ^ ? [11:57] I got an upload error from the daily trusty PPA. [11:57] looking… [11:57] Maybe I should just try again, but I don't want to exhaust my daily attempts. :) [11:57] jtv: did you use a recipe to build the package? [11:58] jtv: maas-maintainers/+recipe/maas-daily-trusty build the *1.5* branch? [11:58] (i.e. not trunk) [11:58] s/branch?/branch!/ [11:58] Ah! That's the recipe I used. [11:58] And yet, not the branch I've been working on. [11:58] That's you problem then. [11:58] your* [11:59] I suggest you create a recipe to build a package from trunk and put it in the experimental PPA. [11:59] Yes, that sounds like a good idea. [11:59] Also, shouldn't the daily/trusty recipe build from a trusty packaging branch? [12:00] jtv: that's what it does [12:00] Then why the different package contents..? [12:01] It said a package with the same version but different contents was already in the PPA. [12:02] jtv: I don't know. Looks like it should be exactly the same content. [12:03] Could be a transient clash as version numbers sync up again after the fork, I suppose. [12:04] jtv: I’m glad that rvba’s here, because I don’t have a clue. DDs went in only as far as needed, and were promptly ejected from braincore at the end of that project. [12:05] \o allenap [12:05] o/ rvba [12:06] rvba: I guess we should check the permissions on the parent directory. [12:06] Hi allenap [12:06] re. dhcpd lease rotating. [12:06] allenap: yes [12:06] Eyup jtv. [12:07] allenap: paste.ubuntu.com/7329355/ [12:07] allenap: I /think/ the dhcp.leases~ is the old, "rotated", file. [12:08] rvba: Not sure how that’s being rotated by a process running as dhcpd. [12:08] rvba: ls -d /var/lib/maas/dhcp [12:08] rvba: ls -ld /var/lib/maas/dhcp [12:08] allenap: sudo ls -ld /var/lib/maas/dhcp [12:08] drwxr-xr-x 2 root root 4096 Apr 25 09:08 /var/lib/maas/dhcp [12:09] allenap, bigjools, gmb, rvba: we have a new recipe: "maas-daily-trunk," against the experimental-builds PPA. [12:12] rvba: It looks like /var/lib/maas/dhcp should be owned by dhcpd. [12:12] allenap: I seem to remember there is a catch somewhere about that… [12:13] I can't remember what it was though. [12:17] Hush :) [12:17] allenap, rvba: IIRC we had a changeover, in Precise, from "it has to be owned by root" to "it has to be owned by dhcpd," or vice versa. [12:18] jtv: Yes, that was it. [12:21] Oh by the way, the import code has a bunch of check_call()s... should those be call_and_check()s? [12:24] And speaking of which, large imports cleanup up for review: https://code.launchpad.net/~jtv/maas/split-out-repowriter/+merge/217209 === vladk is now known as vladk|offline [14:00] allenap: are you still working on that speedup for the leases parser? I was quite looking forward to that. [14:01] It's probably too much to hope that we could just call into ISC's compiled parser... [14:05] jtv: https://code.launchpad.net/~allenap/maas/dhcp-leases-parsing/+merge/215112 is the code. I’m not sure what I’ve got left to do. There are tests, it looks cleanish, but something probably held my hand. === vladk|offline is now known as vladk === vladk is now known as vladk|offline === matsubara is now known as matsubara-afk === vladk|offline is now known as vladk === roadmr is now known as roadmr_afk === vladk is now known as vladk|offline === matsubara-afk is now known as matsubara === roadmr_afk is now known as roadmr === roadmr is now known as roadmr_afk === roadmr_afk is now known as roadmr