File size: 3,717 Bytes
4aa5fce
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
=== StoneTable is now known as aisrael
=== stokachu_ is now known as stokachu
=== \b is now known as benonsoftware
=== wgrant_ is now known as wgrant
=== Ursinha_ is now known as Ursinha
=== kadams54 is now known as kadams54-away
=== mwhudson_ is now known as mwhudson
=== mwhudson is now known as Guest80728
[08:33] <jam> fwereade: you would remember that we discussed writing something like that, but I don't think we ever actualyl did
[08:33] <fwereade> jam, ah right :)
[08:33] <jam> fwereade: what are you doing around now ? :)
[08:34] <fwereade> jam, taking the opportunity to code a little while (almost) nobody's around ;p
[08:34] <jam> understood
[08:36] <fwereade> jam, (although if you fancy reviewing either of my branches (or just conversing on the open issues in the first one) I'd be most grateful
=== mwhudson is now known as mwhudson__
=== mwhudson__ is now known as mwhudson
[09:38] <fwereade> jam, hey, I thought I needed something like this: http://paste.ubuntu.com/11749789/ ...but it turns out it's not quite as simple as that
[09:39] <fwereade> jam, but that general structure looks very familiar
[09:41] <fwereade> jam, do you think there's a place for a worker that just serializes ops onto a single goroutine and lets us focus on the op implementation?
[09:41] <fwereade> jam, possibly I should be calling them commands not ops?
=== kadams54 is now known as kadams54-away
=== Beret- is now known as Beret
=== Tribaal_ is now known as Tribaal
=== kadams54 is now known as kadams54-away
=== kadams54 is now known as kadams54-away
=== _thumper_ is now known as thumper
[22:29] <mup> Bug #1467331 opened: configstore lock should use flock where possible <juju-core:Triaged> <https://launchpad.net/bugs/1467331>
[22:38] <mup> Bug #1467331 changed: configstore lock should use flock where possible <juju-core:Triaged> <https://launchpad.net/bugs/1467331>
[22:41] <mup> Bug #1467331 opened: configstore lock should use flock where possible <juju-core:Triaged> <https://launchpad.net/bugs/1467331>
[23:20] <thumper> wallyworld: https://github.com/juju/juju/pull/2613
[23:20] <thumper> wallyworld: review board seems to be not picking up new pulls
[23:21] <wallyworld> thumper: will look after standup
[23:37] <wallyworld> thumper: lgtm
[23:37] <thumper> wallyworld: cheers
[23:38] <thumper> wallyworld: the other thought I had was the 1 second timeout
[23:38] <thumper> wallyworld: do you think they might be hitting that?
[23:39] <wallyworld> thumper: which timeout? is that mentioned in the bug report?
[23:40] <thumper> no... but it is the timeout that that fslock uses
[23:40] <thumper> the lock is only held during the duration of a read or write of the file
[23:40] <thumper> so I originally thought 1s should be more than enough
[23:40] <thumper> however if there are many ops running concurrently, they are serialized through this file
[23:41] <thumper> so perhas 2s?
[23:41] <thumper> I'm not sure
[23:41] <thumper> could just leave it as it is for now
[23:41] <thumper> I suppose
[23:41] <thumper> pretty sure we never hit this before
[23:42]  * thumper asks the bot to merge
[23:42] <thumper> will forward port to master
[23:42] <thumper> damn it is going to be cold at the gym
[23:42]  * thumper psychs himself up
[23:42] <wallyworld> thumper: yeah leave it for now if we haven't seen it before. but if the system has lots to do. may be a factor. i had to introduce reties to lxc ops due to i/o ontention recently
[23:50] <thumper> meh, just noticed bug 1464356
[23:50] <mup> Bug #1464356: TestCloudInit fails <blocker> <ci> <gccgo> <regression> <test-failure> <windows> <juju-core:Invalid> <juju-core 1.24:Triaged> <https://launchpad.net/bugs/1464356>
[23:50]  * thumper off to the gym