=== emgent_ is now known as emgent === ianmcorvidae|alt is now known as ianmcorvidae === asac_ is now known as asac === Alb3rts is now known as Alberts === Alberts is now known as Alb3rts === vuntz_ is now known as vuntz === ubottu changed the topic of #ubuntu-meeting to: Current meeting: Community Council | Calendar: http://fridge.ubuntu.com/event | Logs: https://wiki.ubuntu.com/MeetingLogs/ | 15 Jul 15:00 UTC: Server Team | 16 Jul 17:00 UTC: QA Team | 16 Jul 22:00 UTC: Platform Team | 17 Jul 13:00 UTC: Desktop Team [12:07] * persia peers about for a community council meeting [12:13] @schedule amsterdam [12:13] leoquant: Schedule for Europe/Amsterdam: Current meeting: Community Council | 15 Jul 17:00: Server Team | 16 Jul 19:00: QA Team | 17 Jul 00:00: Platform Team | 17 Jul 15:00: Desktop Team [12:13] leoquant: Should be CC meeting now: just quiet yet... [12:13] hmmm [12:14] https://wiki.ubuntu.com/CommunityCouncilAgenda doesn't appear to have been updated since the last meeting, which makes me wonder if the agenda is new or old. [12:15] * leoquant is reading [12:16] not been updated i think [12:18] Maybe. I don't see the discussion of the ratification of the Xubuntu strategy document in the logs from last meeting [12:18] http://irclogs.ubuntu.com/2008/07/01/%23ubuntu-meeting.html [12:18] That is supposed to be today [12:18] Ought be now, no? [12:19] persia i have got the cc meeting here in about 4 hours [12:19] As far as I know... [12:19] leoquant: Not at 1100 UTC today? [12:19] Four hours? Why does ubottu say it's now? I thought it was 11:00 UTC on the 3rd Tuesday of the months. [12:20] ubottu: amsterdam 17.00 its now 13.15 [12:20] leoquant: Error: I am only a bot, please don't think I'm intelligent :) [12:20] leoquant:Amsterdam 17:00 is the server team meeting [12:21] your just a couple of diodes right ubottu? [12:21] ah, soory persia [12:21] r [12:22] :/ [12:23] Hmm... None of the members seems to be in-channel. Any volunteers to contact them? [12:31] Right. I'll send IRC pings then, in the absence of any other volunteer. [12:31] persia, in #ubuntu-devel (I saw sabdfl) [12:37] I've sent /queries to everyone on https://launchpad.net/~communitycouncil/+members : maybe they'll join [12:42] philwyett what does this mean?: [quote]Hi,The team wiki is organised under a system of categories i.e.CategoryArchiveCategoryDocumentationCategoryCanonicalEmployeeetc.; and of course...CategoryHomepageAs part of a wiki clean up, as many pages as possible in a first edit [12:42] pass are being placed in their appropriate categories. Yours as auser page falls into 'CategoryHomepage' and was added by myselftoo it.The changes being made are for the good of the wiki organisation and [12:42] to facilitate future addition and maintenance.[/quote] [12:43] is this a private action [12:43] is it a team ¨wiki clean up¨ action? [12:46] leoquant: A more appropriate place would be to ask this by answering my email. However, I am working to clean up the wiki as a volunteer and as a private action. [12:48] please sign your mail with pgp, and you get an email response. ok? [12:54] leoquant: Your email to me never specified that and is not really a good reason not to reply after I took the time to respond to an email that only had a question mark for a subject line. [13:22] So the community council meeting got cancelled for today or something? [13:23] cody-somerville: looks like it i asked same thing a while ago [13:24] noone could tell me why [13:24] persia pinged all the members, but I haven't seen an answer === leonel_ is now known as leonel [14:15] i replied to persia but they had left [14:15] gnomefreak: i am at a all-day meeting for the wikimedia foundation and don't usually make this one for timezones anyway === ubottu changed the topic of #ubuntu-meeting to: Current meeting: Bugs for Hugs Day | Calendar: http://fridge.ubuntu.com/event | Logs: https://wiki.ubuntu.com/MeetingLogs/ | 15 Jul 15:00 UTC: Server Team | 16 Jul 17:00 UTC: QA Team | 16 Jul 22:00 UTC: Platform Team | 17 Jul 13:00 UTC: Desktop Team | 17 Jul 14:00 UTC: Java Team [14:55] mako: Sorry. I've been having network issues today. Thanks for the explanation. [15:08] hello all [15:09] #startmeeting [15:09] Meeting started at 09:11. The chair is mdz. [15:09] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [15:09] this is the bi-weekly Ubuntu Technical Board meeting [15:10] there is nothing on the agenda at present [15:10] so I'll open the floor for brief topics [15:10] [TOPIC] Other business [15:10] New Topic: Other business [15:12] nothing from me [15:12] Scott is on his way [15:13] there are a few things outstanding on the mailing list which we could discuss [15:13] cdrtools [15:13] DKMS [15:14] here now, EPIC Google Calendar FAIL again [15:14] that fsck on shutdown thing [15:14] AutoFsck [15:14] urgh [15:15] everything's an hour out this week :-/ [15:15] it must have got confused about me being in Istanbul [15:15] [TOPIC] cdrtools [15:15] New Topic: cdrtools [15:15] cdrtools: appears to need a lawyer at this point. My armchair-licence-laywer sense tells me that the licences are not compatible [15:15] sabdfl: anything to say on cdrtools? I think this is on your plate currently [15:16] i'm expecting to meet joerg in Portland next week [15:16] don't believe we can change our position [15:16] so, no further action [15:17] ok, so we seem to agree that there is in fact an issue, and need further discussion with joerg to work it out [15:17] [TOPIC] DKMS [15:17] New Topic: DKMS [15:18] upload privileges thereof? [15:18] so, DKMS is now in main, and Mario still wants to maintain it [15:18] Mark and I both said we were OK with limited upload privileges, and I added that I thought Mario ought to pursue core-dev as well [15:18] Keybuk: do you have a position? [15:18] my only thought is that we've previously granted core-dev with restriction in these cases [15:18] should we change those cases to now be upload privilege? [15:18] and how does this interact with future archive changes? [15:19] Keybuk: I think this can be a pilot test for restricted upload [15:19] the functionality exists, but requires some hand hacking in the Launchpad DB [15:19] if it works, we can then migrate the other special cases (tim and till, as I recall) [15:20] I have no objection to this [15:20] Keybuk: I think it dovetails nicely with ArchiveReorganisation [15:20] in that these are just more lists of packages [15:20] with specific permissions and other attributes [15:21] ok, so I'll confirm to Celso that we can experiment on Mario [15:21] [TOPIC] AutoFsck [15:21] New Topic: AutoFsck [15:21] this first came up some months ago, and was re-raised just recently on the list by sabdfl [15:21] I think we're in agreement on two points: [15:21] 1) we could do better at checking filesystems in a user-friendly way [15:22] 2) AutoFsck is not a solution we could adopt as-is [15:22] the author says that he is unable to develop it further, and it could probably be designed more robustly, but the concepts seem worth exploring [15:22] which concepts? [15:22] like running fsck on shutdown [15:23] it doesn't run fsck on shutdown though [15:23] rather than on startup, when the user clearly wants to use the computer [15:23] Keybuk: quoting you: [15:23] (1) run the filesystem check on shutdown when close to the maximum count [15:23] (2) get rid of the mount count check altogether [15:23] AutoFsck attempts a variant of (1). [15:24] I haven't read the code, only your technical review [15:24] have I misunderstood it? [15:24] ish [15:25] but the underlying "I've finished with my machine for a bit, check the filesystem and power it off" idea is a sound one [15:25] of course, an actual user shutdown is often a bad time since the machine is often about to be moved or something [15:25] I suggest that we have a more open discussion about the problem and possible solutions [15:25] and see what comes of it [15:25] I agree [15:26] then channel that into planning discussions for post-8.10 [15:26] Keybuk: would you post something to -devel or -devel-discuss and start things off? your review of AutoFsck might be a good starting point [15:26] indeed, I'll clean that up and bounce it across [15:26] [ACTION] Keybuk to start discussion about filesystem checking on -devel/-devel-discuss [15:26] ACTION received: Keybuk to start discussion about filesystem checking on -devel/-devel-discuss [15:26] [TOPIC] Other business [15:26] New Topic: Other business [15:26] anything else? 30 seconds [15:27] TB membership? [15:27] [ACTION] mdz to confirm upload arrangement for Mario with Launchpad [15:27] ACTION received: mdz to confirm upload arrangement for Mario with Launchpad [15:27] as in, the expanding thereof [15:27] [TOPIC] technical board membership [15:27] New Topic: technical board membership [15:28] "Appointments to the board are made by Mark Shuttleworth subject to confirmation by a vote amongst the maintainers." [15:28] [LINK] http://www.ubuntu.com/community/processes/techboard [15:28] LINK received: http://www.ubuntu.com/community/processes/techboard === thekorn_ is now known as thekorn [15:28] that page also needs updating to reflect mjg59's resignation [15:29] that too [15:29] sabdfl: perhaps we should send something out to core developers inviting participation in the technical board? [15:29] [ACTION] mdz to get techboard web page updated to reflect current membership [15:29] ACTION received: mdz to get techboard web page updated to reflect current membership [15:29] sounds reasonable [15:29] do we want any participation from other communities? [15:30] sabdfl: I think it's appropriate for tech board members to have a strong working knowledge of Ubuntu practices [15:30] given that it's responsible for setting policy [15:31] in general, I think we should encourage participation and input from other communities, but I'm a bit more conservative about the decision-making process [15:32] agreed, and that's the primary function of the TB, but I wondered recently if it would be good for us and them if GNOME, KDE and others had more direct representation in our governance, and where to reflect that if so [15:33] I think the best representation for those two specifically is via the Ubuntu core developers who package their work [15:34] at a technical level, yes [15:34] [LINK] https://bugs.edge.launchpad.net/ubuntu-website/+bug/248729 [15:34] LINK received: https://bugs.edge.launchpad.net/ubuntu-website/+bug/248729 [15:34] Launchpad bug 248729 in ubuntu-website "/processes/techboard out of date" [Undecided,New] [15:35] perhaps the CC would be a better place for more general representation [15:35] i don't want to do the old "advisory board" because those have no teeth [15:35] not something we need to blog on [15:35] perhaps the TB could make itself more known to upstream communities, and inform them of good points of contact for working with Ubuntu [15:35] err s/blog/block/ [15:36] and we could certainly provide more upstream-oriented resources, like we do for Debian developers [15:37] I'll talk with Jono about accumulating a FAQ in the course of his upstream outreach [15:37] jono and jorge are making great progress [15:37] i'll keep thinking about more direct representation [15:38] [ACTION] sabdfl to solicit tech board nominees from the community [15:38] ACTION received: sabdfl to solicit tech board nominees from the community [15:39] [TOPIC] Other business [15:39] New Topic: Other business [15:39] last call [15:39] nothing from me [15:39] thanks [15:39] #endmeeting [15:39] Meeting finished at 09:41. [15:41] thanks all, cheers [15:59] _/\_ [15:59] hey all [15:59] hi [15:59] howdy [16:00] o/ === ubottu changed the topic of #ubuntu-meeting to: Current meeting: Server Team | Calendar: http://fridge.ubuntu.com/event | Logs: https://wiki.ubuntu.com/MeetingLogs/ | 16 Jul 17:00 UTC: QA Team | 16 Jul 22:00 UTC: Platform Team | 17 Jul 13:00 UTC: Desktop Team | 17 Jul 14:00 UTC: Java Team [16:00] hia guys ! [16:00] o/ [16:01] let's get this started [16:01] #startmeeting [16:01] Meeting started at 10:03. The chair is mathiaz. [16:01] Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] [16:01] \o/ [16:01] _/\_ [16:01] ___ ? [16:01] dendrobates: scoliosis? [16:01] hello [16:02] https://wiki.ubuntu.com/ServerTeam/Meeting [16:02] \o/ [16:02] _/\_ [16:02] /o\ [16:02] nealmcb: thanks for the meeting agenda [16:02] :) [16:02] last meeting minutes: https://wiki.ubuntu.com/MeetingLogs/Server/20080708 [16:02] kirkland: updated the lsb section [16:03] mathiaz: i certainly did, big progress there [16:03] mathiaz: actually, broke it out to its own wiki page [16:04] mathiaz: just waiting on one more change to the library function to make it upstream [16:04] mathiaz: could i perhaps call for volunteers from the community to help with other init scripts? [16:04] the new page to track all of this: https://wiki.ubuntu.com/InitScriptStatusActions [16:04] mathiaz: the changes are pretty simple [16:04] hello [16:04] mathiaz: good patch practice [16:04] mathiaz: while still being very useful functionality [16:05] all: there's a recipe on the page mathiaz mentioned [16:05] cool [16:05] kirkland: do you have a list of other services that need to be updated ? [16:05] mathiaz: I'll grab that by next week. owh created that list for hardy some time ago [16:06] kirkland: I see only references to bugs that have already been fixed [16:06] kirkland: i will, i'm a little busy this days so i have just a little time at nights for contributing, and init scripts take me just little time [16:06] :D [16:06] mathiaz: see the link to Onno Benshop's page [16:06] but i thing a list of pendients will be really usefull [16:06] https://wiki.ubuntu.com/OnnoBenschop/ubuntu-server/init.d-status [16:07] on a quick look [16:07] i don't understand that page [16:07] i'll nail down a good list [16:07] by the next meeting [16:08] kirkland: awesome [16:08] [ACTION] kirkland to update the init script wiki page with a list of packages that should be fixed. [16:08] ACTION received: kirkland to update the init script wiki page with a list of packages that should be fixed. [16:08] nxvl: any news on the augeas front ? [16:08] and i'm going on holidays the week after the next one so i will have more time [16:08] mathiaz: a lot [16:09] ok [16:09] augeas is already on the archives [16:09] it has been acepted an included already [16:09] also i got it accepted on debian [16:09] thee is a session about augeas at OLS that I will go to [16:09] and it reched the archives on sunday IIRC [16:09] or yesterday maybe [16:10] raphink has been working on some lenses [16:10] nxvl: great - so the next step is to write more lenses [16:10] any more feedback from ebox on augeas? [16:10] and he's reviewing them with lutter (the upstream PL) [16:10] mathiaz: yes it is [16:10] debian too - great! [16:10] nealmcb: there are still to few lenses [16:10] nxvl: great ! [16:11] nealmcb: i think it will better to write more of them, and then ping the eBox team again [16:11] also [16:11] i have been mailed about Model:Config [16:11] nxvl: could you add a point to the roadmap about augeas integration ? [16:11] http://config-model.wiki.sourceforge.net/ [16:11] LINK received: http://config-model.wiki.sourceforge.net/ [16:11] https://sourceforge.net/project/screenshots.php?group_id=155650 [16:11] and they said it will support augeas soon [16:12] so i'm waiting for it to start palying [16:12] if they made it soon, i think we can have UCSA for intrepid+1 [16:12] at least the first version [16:12] mathiaz: doing it right now [16:12] nxvl: could config-model be integrated with augeas ? [16:12] nxvl: it seems that both would fit well [16:13] nxvl: the blue section could use augeas and its lenses [16:13] nxvl: I'm refering to the picture at http://config-model.wiki.sourceforge.net/ [16:14] hm - nm - I've just noticed that augeas is being integrated in Config::Model [16:14] great minds thinking alike :) [16:15] nxvl: do you think about packaging config-model ? [16:16] sorry, needed to minimize the window [16:16] * nxvl read the questions [16:17] mathiaz: yes, they said they are planning on supporting augeas [16:17] for me that's really important since augeas will let us manage the config files AND let the sysadmins edit them by hand [16:17] without breaking anything === gnomefre1k is now known as gnomefreak [16:18] i'm in contact with config-model upstream, so i'm waiting for them to support augeas and keeping an eye on it [16:18] and yes [16:18] i have planned on packaging it BUT after having a good amount on lenses [16:18] and after it supports augeas [16:18] nxvl: waiting for support augeas may be a good thing [16:19] nxvl: I wouldn't wait for a lot of lenses [16:19] yes, that's true since we can start playing with just few lenses [16:19] but [16:19] nxvl: having config available would show case what can be done with augeas [16:19] also i have an exchange of mails with them [16:19] nxvl: and thus trigger more interest in writing lenses [16:20] and lenses will not be the only think needed, it will also need a model on config-model, which keeps the logic behind the config files managment [16:20] which seems pretty fair to me [16:20] since just config files managment isn't enought [16:20] nxvl: sure [16:21] oh! ok [16:21] i understand what you mean [16:21] yes, it sounds awesome for me [16:21] but still we need to wait until it support augeas [16:21] which i hope will be soon [16:22] nxvl: yes - I'd suggest to wait for augeas support and then package for ubuntu [16:22] ep [16:22] will do [16:22] :D [16:22] i wil keep track of it next week [16:22] since this week i'm in final exams [16:22] nxvl: great - thanks [16:22] let's move on [16:23] so next week with one think lees to care about i will give ucsa the time i was giving to the university [16:23] [TOPIC] Encrypted ~/Private Directory in Each User's Home [16:23] New Topic: Encrypted ~/Private Directory in Each User's Home [16:23] kirkland: ^ ? [16:23] mathiaz: in good shape [16:23] mathiaz: MIRs nearly done [16:23] mathiaz: ie, nearly approved [16:23] kirkland: testing instructions are up-to-date ? [16:23] kirkland: did you resolve the pam stack issue ? [16:23] mathiaz: need to clean up some sprintf's in one of the the libraries, other than that, all approved [16:24] mathiaz: we're using jdstrand's auth-client-config as a temporary work around for now [16:24] mathiaz: slangasek has a comprehensive pam stack configurator in his head, he's trying to put together for intrepid [16:24] mathiaz: wiki testing instructions are most definitely up to date [16:24] mathiaz: i would very much appreciate any intrepid server users out there using/testing it!!! [16:25] kirkland: ok - so thre is a workaround even if the pam integration doesn't make it for intrepid [16:25] kirkland: is there a link for the testing? [16:25] sommer: https://wiki.ubuntu.com/EncryptedPrivateDirectory#head-4a2aa7460fdca18bfe78bb1283becff406bbc13c [16:25] kirkland: thx [16:25] mathiaz: hmm, there's a one liner that the sysadmin has to run, specifically: [16:25] kirkland: I plan to write a blog post on ubuntuserver asking for testing [16:25] mathiaz: sudo auth-client-config -p ecryptfs_standard -t pam-auth,pam-session [16:25] mathiaz: it's a one time deal [16:26] mathiaz: i think we're running into debian policy problems, with one package needing to modify another package's config files [16:26] kirkland: can't you call that from the posting ? [16:26] kirkland: *postinst* [16:26] mathiaz: i'm under the impression that Debian Policy says no [16:27] mathiaz: libecryptfs0 package provides pam_ecryptfs.so [16:27] kirkland: it is a command line, not a change to the conf [16:27] kirkland: IIRC, since it's a command you could use it [16:27] mathiaz: it needs to make two modifications, to /etc/pam.d/common-auth, and common-session to make the unwrap passphrase work correctly [16:27] kirkland: you may wanna ask slangasek about it though [16:27] only if you could revert the change from pre/post-rm, right? [16:27] mathiaz: slangasek would not like it done that way [16:27] kirkland: ok [16:27] mathiaz: it will be our fall back for intrepid [16:28] right - let's move on [16:28] [TOPIC] Migrate new installs and upgrades of client and server packages to use SSL v3 or TLS [16:28] New Topic: Migrate new installs and upgrades of client and server packages to use SSL v3 or TLS [16:28] ivoks: you made the list [16:28] yay [16:29] so, basicaly, this can be done per package or in openssl [16:29] ivoks: there are only a couple of packages [16:29] ivoks: what would be required in openssl ? [16:29] if we would drop sslv2 from openssl, we would solve all problems [16:29] surely except for upgrades? [16:29] openssl can be compiled without SSLv2 [16:29] and this is the problem [16:30] we cant go with openssl compile changes cause of upgrade [16:30] but we can do per package configuration change on fresh install [16:31] mathiaz: right, only couple of them; if you think of any other that provides SSL, please add it to the list :) [16:31] sslv3 would have to conflict with sslv2? [16:31] no [16:31] what if I have a client that only speaks v2? [16:31] most of the services provide sslv2 and sslv3 [16:31] I would not like to be locked out at openssl level [16:31] in most of the cases, clients asks for sslv2 [16:31] ivoks: right - dropping sslv2 from openssl should definitely be discussed on ubuntu-devel [16:32] nijaba: then that client is very broken [16:32] ivoks: I think the path you suggest is safer and more reasonable for intrepid [16:32] sslv3 is here for a decade [16:32] ivoks: right, but there are a lot of very broken thing in the enterprise [16:32] ivoks: ie do it on a per-package basis [16:32] I kind of like the idea of just dropping v2 from openssl. It does make me cringe a little about breaking people, but it's a sure way to make sure it's off. :P [16:33] dropping sslv2 may be worth discussing at next uds [16:33] mathiaz: and i would go only with service providing packages; ie, not with clients [16:33] for the intrepid timeframe, we'd better focus the per-package approach [16:34] right... in most cases, changes are trivial... [16:34] ivoks: excellent - the list of package is there. [16:34] some packages will require code changes; uw-imapd [16:34] What about providing two openssl packages? One with and one without v2. [16:34] it seems like per-package would be more accepted by Debian upstreams too [16:34] Brazen: that seems too complicated [16:34] ok [16:35] i'll provide patches for all packages on the list by the end of the week [16:35] ivoks: do you know which packages are easier to do ? like the one that don't require src code changes ? [16:35] and then i'll examine what else we have in universe :/ [16:35] ivoks: great - thanks for this work [16:35] mathiaz: all packages listed on wiki need 1-2 lines in config [16:36] let's move on [16:36] :) [16:36] [TOPIC] Integration of Dovecot SASL and Postfix [16:36] New Topic: Integration of Dovecot SASL and Postfix [16:36] ivoks: have you looked into cyrus sasl integration ? [16:36] i tought we decided to replace that with Cyrus SASL [16:36] mathiaz: i have couple of cyrus sasl production enviroments [16:36] and i think everybody who played with email servers know how to set it up [16:37] ivoks: correct - I've renamed the task to :Integration of SASL and Postfix [16:37] i need to run, read you all guys! [16:37] * nxvl HUGS everyone [16:37] only 'issue' is to package it right [16:37] nxvl: bye; good work ;) [16:37] since our postifx is jailed, we'll have to bind mount cyrus socket [16:38] and that brings us back to the core of dovecot's sasl 'problem' :D [16:38] with one exception; cyrus sasl is configured for sasl out of the box [16:38] ivoks: right - could you discuss this issue with lamont ? [16:38] so, we should just bind it's socket to postfix [16:38] sure [16:39] ivoks: great - thanks [16:39] it was my pleasure ;) [16:39] [ACTION] ivoks to discuss cyrus socket integration with lamont [16:39] ACTION received: ivoks to discuss cyrus socket integration with lamont [16:40] that's all there is on the Last meeting minutes [16:40] let's move on to review progress made on the specification listed on the Roadmap. [16:40] https://wiki.ubuntu.com/ServerTeam/Roadmap [16:41] [TOPIC] Track pages on help.ubuntu.com that need to be updated [16:41] New Topic: Track pages on help.ubuntu.com that need to be updated [16:41] sommer: ? [16:41] err, not much progress with the wiki [16:42] the samba sections of the serverguide are updated though :) [16:42] except for integrating with AD, but that's coming soon [16:42] sommer: awesome - I discussed with LaserJock about bzr branch [16:42] sommer: I haven't done more work on that front [16:42] sommer: but we don't need to keep the .po files in the bzr branch [16:43] mathiaz: that's cool, I briefly looked at it and didn't get too far either [16:43] sommer: we'd just had to pull them from lp when releasing a new package [16:43] sommer: I'd put that in the release process rather then working on the package itself [16:43] mathiaz: gotcha, seems pretty straight forward [16:44] sommer: the difference between an upstream write (just using the bzr branch to update the server guide content) [16:44] sommer: and the package maintainer that is responsible for pulling all the things together (with the translateion) [16:45] sommer: dropping the po files would make the bzr branch a few 100k [16:45] sommer: making branching super-fast [16:45] super fast is good [16:46]