1 17:01 -!- pere
2 17:01 * pere = Petter Reinholdtsen
3 17:02 < pere> sorry for being late.
4 17:02 * h01ger = Holger Levsen
5 17:02 < h01ger> pere, you're the first and starting the meeting :)
6 17:02 < pere> oh
7 17:03 < h01ger> s/the first/in time/ - sepski joined 10min or so ago
8 17:03 < sepski> ?? isnt it 20:00 ??
9 17:03 < sepski> i just turned on the computer... :)
10 17:04 < pere> topic says 17:00 UTC, and that is 19:00 norwegian time, aka 4 minutes ago.
11 17:04 < h01ger> 17 utc, thats 19 cest. but on wiki.skolelinux.de also 18 utc was mentioned, i fixed it this noon to 17 utc..
12 17:04 < pere> there is a cipux meeting 18:00 UTC.
13 17:04 < pere> h01ger: ouch. did not know that.
14 17:04 * sepski = Ronny Aasen
15 17:05 < pere> so, should we hold the meeting with the three of us?
16 17:06 < pere> who moderates, who write the summary and who collect the log. no use continuing before that is set. I can moderate.
17 17:06 < h01ger> we could wait an hour and enter the cipux meeting :) /me looks at the agenda now to give a better answer :)
18 17:06 * h01ger can collect the log
19 17:06 < sepski> looking at the agenda there realy is not much change since yesterday i assume
20 17:06 < pere> well, we do know a bit more. :)
21 17:07 < pere> and the summary from the meetings are for those not present, for example those not present yesterday either.
22 17:07 < sepski> well i can do the summary then , and lets do it quickly, since i have kids that (don't) want to go to bed
23 17:08 * h01ger nods
24 17:08 -!- pere
25 17:08 < h01ger> http://wiki.debian.org/DebianEdu/Status/Etch is up2date?
26 17:08 < pere> h01ger: not completely, I believe.
27 17:08 * h01ger just updated it but saving takes ages
28 17:09 < sepski> pere, the purpose of the NMU was to add the comma support in the packages part.
29 17:09 < sepski> i tested that local and it seamed to work, strange our cd did not
30 17:09 < pere> for example, I belive the unstable cd build is "fixed", at least it isn't unstable any more. now it should fail more spectacularly.
31 17:09 * h01ger nods
32 17:10 < h01ger> we could also try to set SOURCE=1 to get a source-cd
33 17:10 < pere> h01ger: we could. who got time to check the content of the source CD?
34 17:11 < pere> as you probably noticed, I enabled the graphical installation as the default. It seem to break a few packages. I am unable to see the error message, as qemu do not allow me to switch away from tty1.
35 17:12 < h01ger> pere, i could. for i in "pkglist" ; do find /source-cd -name "*$i*" ; done - right?! :)
36 17:12 < h01ger> pere, i belief you can switch with alt-arrow-left+right
37 17:12 < h01ger> believe even
38 17:12 < pere> h01ger: only in text mode, not in graphical mode. :(
39 17:13 < h01ger> hu
40 17:13 < pere> h01ger: you would have to map the binary package names to source package names first.
41 17:13 < sepski> i can check those errors next time i install.
42 17:13 < h01ger> pere, ah, right. thats a bit harder. so i say "no" as i want to finish my taken tasks first...
43 17:13 < pere> as I see it, there are two errors blocking us from releasing test02. libnss-ldap, and the broken ltsp install.
44 17:14 < pere> are there others?
45 17:14 < h01ger> (why) can we take sysvinit from unstable to fix libnss-ldap?
46 17:15 < pere> h01ger: we could, but I am reluctant to divert from etch for such important package.
47 17:15 < pere> I am splitting the status list into three, fatal, important and annoying.
48 17:15 < sepski> the ltsp install breaks becouse my patch is not actualy included in the NMU.
49 17:16 < h01ger> so we'll have to wait for the fix in etch..
50 17:16 < sepski> bzr tricked me again
51 17:16 < sepski> ill upload again in a few houers
52 17:17 < pere> h01ger: well, we could NMU libnss-ldap.
53 17:18 < h01ger> pere, i thought the patch wont work without a newer sysvinit (though i havent looked at the patch)
54 17:18 < pere> I can have a look at that, but it will have to wait.
55 17:19 < pere> h01ger: that is correct, so we need to come up with a different patch.
56 17:19 < pere> I marked the two blockers fatal, and spread the rest into important and annoying. probably got the latters wrong.
57 17:20 < pere> any more to say on the etch status?
58 17:21 < h01ger> no. good idea to clean up / mark the bugs on the status page!
59 17:21 < pere> I proposed earlier that we should test rc1 now, to see if the problems we have noticed in beta3 is fixed there or not. anyone interested in doing that?
60 17:21 < h01ger> rc1 is not available yet. or do you mean the dailys?
61 17:22 < sepski> i am testing the dailys
62 17:22 < pere> h01ger: I mean the dailys, yes.
63 17:23 * h01ger will mostly have to work for money this week... :( i wont take more than i already have.
64 17:23 * pere will have to focus on the NUUG prize this week and next week, so I can't promise anything either.
65 17:24 -!- pere
66 17:24 < sepski> ill have to handle the kids now. ill write summary from the log afterwards. upload ltsp and check libnss again, since i got it working on my test with minor corrections
67 17:24 < pere> any news on the user admin stuff? as holger said this weekend, we need to make a decision soon, and start focusing the work.
68 17:25 < pere> the cipux meeting start in 35 minutes. it might be useful to talk to them about it. It is a sad that none of that team participate in this meeting.
69 17:26 < h01ger> pere, can you join #cipux?
70 17:26 -!- pere
71 17:26 < pere> h01ger: I can't. when this meeting is done, I have to run.
72 17:27 < h01ger> ok. so i will continue to play the bad guy^w^w^wbe the nice guy and help them :)
73 17:28 < pere> I believe our best strategy to fix the RC bug in debian-edu-config is to reduce the editing in debian-edu-config to a bare minimum, and then focus on solving the problem for the remaining few packages.
74 17:28 < pere> is it time to remove everything woody and sarge related from the source?
75 17:28 < pere> it could make it easier to see what remains
76 17:28 < h01ger> ack on the strategy and i would say so
77 17:29 < pere> and for every edit left, we evaluate and see if it is really needed, or if we can drop it.
78 17:29 * h01ger nods
79 17:30 < h01ger> winnie has done some work on it, maybe he can pick it up again? (hint hint highlight :)
80 17:30 < pere> this will be a lot of hard word. dropping the obsolete code is probably fairly easy, evaluating what is left will take more time
81 17:30 < pere> that would be great.
82 17:31 < pere> any more anyone want to say on that topic?
83 17:31 < pere> any other topics anyone want to cover during this meeting?
84 17:31 < h01ger> there has been activity on the munin-bug of that bug. (luk and the maintainer)
85 17:31 < h01ger> (yesterday)
86 17:32 < pere> any estimate for a fix?
87 17:32 < pere> should we just drop sysklogd and use a better syslogd? Are there any preseedable syslogd available?
88 17:33 * h01ger pings Joey again. i think its better to stay with debians syslogd. and for etch debian will stay with syslogd
89 17:33 < pere> any more comments before we close this meeting? When is the next meeting?
90 17:34 < sepski> *quick read of the backlog* lets not wait too long for the next meeting.
91 17:35 < pere> 2 weeks? 3 weeks?
92 17:35 < sepski> 2 is ok by me,
93 17:35 < sepski> but lets make sure we announce it properly
94 17:35 < pere> 2006-10-30?
95 17:35 < h01ger> for the #cipux-meeting: what is our general release plan? release ASAP after debian-etch? or?
96 17:35 < pere> h01ger: asap after etch, yes.
97 17:36 < pere> as soon as our stuff is working properly. :)
98 17:36 < sepski> should we test upgradeabillity, or dont we care about that ?
99 17:36 < h01ger> we should, cause we care?! :)
100 17:36 < pere> sepski: we should test upgradability, but I doubt we have enough people to do it.
101 17:37 < h01ger> ask on the user lists
102 17:37 < h01ger> we can...
103 17:37 < pere> yeah
104 17:37 < pere> there should at least be a documented migration plan from sarge- to etch-based debian-edu.
105 17:38 < pere> but as we want people to use online resizable ext3 file systems etc, I suspect reinstall is required.
106 17:38 < pere> it can't be safely added to existing file systems.
107 17:39 < pere> ok, enough monologues from me. I declear this meeting to be over. see you all in 2 weeks. :)
108 17:39 < h01ger> is that so important? i mean, it would ensure people test their backups (if they would need to reinstall)
109 17:39 < h01ger> cu :)
110 17:40 < pere> bring my greetings to the cipux people, and try to get them to understand our requirements.
111 17:40 -!- pere
112 17:43 < pere> sepski: btw, please post the text of the summary to debian-edu@, not the link. It should make more people read the summary.