20080305
⇤ ← Revision 1 as of 2008-03-06 04:52:07
Size: 18360
Comment:
|
Size: 18991
Comment:
|
Deletions are marked like this. | Additions are marked like this. |
Line 1: | Line 1: |
== Agenda == Items we will be discussing: * Review ACTION points from previous meeting. * Server survey - too long. Should we cut some questions ? * Review ["ServerTeam/Roadmap"] and ["ServerTeam/ReportingPage"]. * Agree on next meeting date and time. == Minutes == === Forum Reporting === faulkes- created a project on Launchpad [1] for his forum analyser [1]. There is a README file in the src directory outlining how to install the software. [1]: https://launchpad.net/forumstats/ [2]: http://ubuntu.oss-mgmt.com/ === Server survey === owh drafted an announcement for the survey [3]. He also added a list of resources to which the announcement should be sent. [3]: https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch nijaba reported that testers found the survey too long to complete. ACTION: faulkes-, owh and nijaba to work on tuning the questions to make the survey shorter. === ServerTestingTeam === mathiaz revamped the ServerTestingTeam wiki pages [4]. The process to run tests and report results has been streamlined. He will send a separate announcement to ubuntu-server about the ServerTestingTeam. [4]: https://wiki.ubuntu.com/ServerTestingTeam ACTION: mathiaz to send an email about the ServerTestingTeam === iSCSI Support === soren has come across a few use cases he hadn't taken into account: Booting off of the network and having the root filesystem on iscsi. He'll discuss supporting this scenario with dendrobates and slangasek. ACTION: soren to talk with dendrobates and slangasek about iscsi support for root fs. === Fix bugs marked by the QATeam === The QA team maintains a list of bugs that should be fixed for hardy [5]. kirkland fixed one bug related to nfs. owh and kirkland will take a look at the ntp bug. [5]: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=qa-hardy-server. === Virtualization === soren uploaded kvm 62. He is also looking into adding parvirt clock to the kernel which should improve timing issues in virtual machines. === Windows Integration === dendrobates is waiting for a new release from upstream. === Ebox === zul filed a FeatureFreezeException request, which has been granted. A new version of ebox is available in universe - testers are welcomed ! === Agree on next meeting date and time === Next meeting will be on Wednesday, March 12th at 21:00 UTC in #ubuntu-meeting. |
|
Line 2: | Line 68: |
TZ UTC-5 | |
Line 4: | Line 69: |
(04:00:29 PM) mathiaz: so - let's get started ! (04:00:34 PM) mathiaz: #startmeeting (04:00:35 PM) MootBot: Meeting started at 21:00. The chair is mathiaz. (04:00:35 PM) MootBot: Commands Available: [TOPIC], [IDEA], [ACTION], [AGREED], [LINK], [VOTE] (04:01:09 PM) faulkes-: lead away mathiaz (04:01:19 PM) mathiaz: Today's agenda can be found online: https://wiki.ubuntu.com/ServerTeam/Meeting (04:02:29 PM) mathiaz: [TOPIC] Review ACTION points from previous meeting. (04:02:29 PM) MootBot: New Topic: Review ACTION points from previous meeting. (04:02:46 PM) mathiaz: Previous meeting: https://wiki.ubuntu.com/MeetingLogs/Server/20080227 (04:03:16 PM) mathiaz: [TOPIC] forum reporting (04:03:16 PM) MootBot: New Topic: forum reporting (04:03:33 PM) faulkes-: ok then (04:03:40 PM) mathiaz: anything to add or disscuss on the forum front faulkes- ? (04:03:47 PM) faulkes-: yes (04:03:56 PM) mathiaz: I think you've published the code on LP (04:04:07 PM) faulkes-: I did a complete rewrite of the code and published it to LP (04:04:17 PM) faulkes-: https://launchpad.net/forumstats/ (04:04:38 PM) mathiaz: faulkes-: great - do you have some documentation somewhere ? (04:04:48 PM) faulkes-: it is much more exact in the nature of how it selects and adds stuff (04:05:00 PM) faulkes-: there is a README file supplied with the source (04:05:18 PM) faulkes-: and I have documented the source as well, it's only about 120 lines long (04:05:30 PM) mathiaz: faulkes-: great. So what's the next step ? (04:05:50 PM) mathiaz: faulkes-: do you want to add more features to it ? (04:06:02 PM) faulkes-: the next step is linking the individual graphs and categories, to being able to see the individual posts (04:06:07 PM) mathiaz: faulkes-: or should we think about what we can do with the outcome ? (04:06:43 PM) faulkes-: which should be fairly easy to do, I have been very busy the last week with pm course exams and taking on a new FT position (04:07:14 PM) faulkes-: the outcome, is more a time based situation, I will also add in the ability to select by date (04:07:44 PM) mathiaz: faulkes-: ok. Seems great. (04:07:50 PM) faulkes-: because the posts affecting something from time X to time Y, may be unrelated due to version from time Z to time L (04:08:01 PM) owh: faulkes-: Is there a way to host the current branch so it can display results? (04:08:37 PM) owh: As in, be "live" (04:08:38 PM) faulkes-: that may be possible but it would require access to my database (04:08:57 PM) faulkes-: if you want to host one yourself, you can run the code completely independent of me (04:09:05 PM) neversfelde_ is now known as neversfelde (04:09:10 PM) faulkes-: and allow it to gather stats (04:09:20 PM) faulkes-: the requirements are listed in the README for running it (04:09:23 PM) owh: Does the code have a means to generate a database as well? (04:09:47 PM) faulkes-: there is sql files included in the LP branch (04:09:53 PM) owh: Cool. (04:10:07 PM) faulkes-: the README has instructions regarding them (04:10:19 PM) owh: Excellent, documentation that actually helps :) (04:10:32 PM) mathiaz: How is the classification of post done ? (04:10:45 PM) mathiaz: faulkes-: I think this is the trickiest part (04:11:01 PM) faulkes-: classification is done via a dictionary (04:11:11 PM) faulkes-: unfortunately, that means building the dictionary (04:11:20 PM) faulkes-: I have included the one which is currently used (04:11:35 PM) mathiaz: faulkes-: ok - is there a way to reassign posts to a different category if they've been misclassifed ? (04:11:49 PM) faulkes-: so, if you wanted to use the code to monitor say, Desktop forum, you would need to build your own (04:12:11 PM) faulkes-: mathiaz: currently no, that would be an admin feature I guess, but wouldn't be hard to build in (04:12:24 PM) faulkes-: as well as add additional categories, terms, etc.. (04:12:56 PM) owh: Going that route would then require a sort of "meta" individual, that seems like a lot of work. (04:12:56 PM) faulkes-: right now, it is by hand, mostly for lack of time on my part to build that in but it would be a fairly straight forward excersize (04:13:03 PM) mathiaz: faulkes-: These are more features. Let's try to get something done first and see how can extract information from that. (04:13:32 PM) mathiaz: Let's move on. (04:13:37 PM) mathiaz: [TOPIC] Server survey (04:13:37 PM) MootBot: New Topic: Server survey (04:13:56 PM) owh: That would be me. (04:13:58 PM) faulkes-: me again I guess as nijaba is away or do you want to handle it mathiaz? (04:14:00 PM) mathiaz: nijaba is not around. (04:14:15 PM) faulkes-: I can go from his action email, that you updated (04:14:22 PM) mathiaz: I've updated the ReportingPage with a status he sent me. (04:14:39 PM) mathiaz: Anything else to add ? (04:14:39 PM) mathiaz: https://wiki.ubuntu.com/ServerTeam/ReportingPage (04:15:03 PM) faulkes-: sec (04:15:40 PM) owh: While faulkes- is looking... (04:15:42 PM) owh: I created https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch which has as many contacts in it as possible, but suggestions are welcome. The intent is to use the body text as the email text. (04:15:46 PM) owh: I tried coming up with actual press contacts where relevant, but I don't have any actual Canonical press contact. (04:15:50 PM) owh: I'm not sure how we best ask Server Team Members to blog about the survey so it shows up on Planet Ubuntu. (04:15:51 PM) mathiaz: It seems that the survey is too long to take. (04:16:05 PM) owh: Yes, I'd agree with that. (04:16:17 PM) faulkes-: well, we've only had a small sampling so far (04:16:25 PM) faulkes-: however (04:16:50 PM) faulkes-: we should be always reporting feature/bug requests to LP, I don't generally check the page itself (04:16:58 PM) faulkes-: so there is some stuff there that I can work on (04:17:14 PM) faulkes-: as for shortening the survey, I think we can do that effectively (04:17:23 PM) kirkland: mathiaz: I've communicated with Jonathan Corbet of LWN before, I can send him a note asking if it would be possible to publicize on LWN.net (04:17:44 PM) kirkland: no promises--he only sometimes responds to my email :-) (04:17:46 PM) faulkes-: I will look at tuning the questions so it is faster (04:17:56 PM) owh: kirkland: Add it to the wiki page, you don't need to give the email address if you're not comfortable with it. (04:18:01 PM) mathiaz: kirkland: seems great - could you add that to https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch ? (04:18:12 PM) owh: kirkland: That way we don't hit LWN multiple times. (04:18:52 PM) owh: faulkes-: I wonder if the survey length is a function of a poorly defined audience/spec. I'm not saying it is, I'm just wondering. (04:19:02 PM) mathiaz: [ACTION] faulkes- to work on tuning the questions to make the survey shorter (04:19:02 PM) MootBot: ACTION received: faulkes- to work on tuning the questions to make the survey shorter (04:19:07 PM) owh: faulkes-: It seems to be sprawling all over the place. (04:19:28 PM) mathiaz: owh: faulkes-: could you coordinate with nijaba ? (04:19:40 PM) owh: Yup (04:19:40 PM) faulkes-: there is some logical order which needs to be done on the questions (04:19:52 PM) faulkes-: mathiaz: yes, I will speak to nijaba (04:19:57 PM) kirkland: mathiaz: owh: done. (04:20:03 PM) mathiaz: owh: faulkes-: thanks - (04:20:06 PM) mathiaz: Let's move on. (04:20:11 PM) owh: Perhaps we can catch up afterwards faulkes- and have a chat about it. (04:20:19 PM) faulkes-: certainly (04:20:49 PM) mathiaz: [TOPIC] ServerTestingTeam (04:20:49 PM) MootBot: New Topic: ServerTestingTeam (04:21:11 PM) mathiaz: So I spent some time to revamp the ServerTestingTeam wiki pages. (04:21:27 PM) mathiaz: https://wiki.ubuntu.com/ServerTestingTeam (04:21:58 PM) mathiaz: The idea is to increase hardware testing for hardy. (04:22:12 PM) mathiaz: Reporting the results is done in the wiki. (04:22:38 PM) faulkes-: interesting, ok, I may be able to add some to that (04:22:53 PM) mathiaz: I'll also send an email to the previous people involved in the team and ask if they can still contribute. (04:23:10 PM) mathiaz: Anyone that has some server hardware is welcome to participate. (04:23:30 PM) faulkes-: for those unaware of my current status, I've moved from being a consultant to a FTE job, I will be converting the current environment to -server (04:23:34 PM) mathiaz: The idea is to download an iso and make sure the installation is successfll (04:24:05 PM) mathiaz: Yet another way to contribute to the Server Team :) (04:24:06 PM) faulkes-: mathiaz: I should have extra hardware resources now available to put to it (04:24:32 PM) mathiaz: faulkes-: great ! I've updated the wiki page to streamline the reporting process (04:25:12 PM) owh: mathiaz: That's looking great. Do you think it would be useful to add that link to the Survey Announcement to encourage more helpers, or might that be counter productive? (04:25:29 PM) mathiaz: owh: I'll make a separate announcement (04:25:33 PM) faulkes-: mathiaz: will ServerTestingTeam be a subset of the ServerTeam for membership purposes or? (04:25:33 PM) owh: Given that we're targeting -server users. (04:25:51 PM) mathiaz: faulkes-: for now, there isn't any LP team or membership (04:26:00 PM) mathiaz: It may be usefull at some point. (04:26:07 PM) faulkes-: owh: well, I think the survey is geared more to established environments rather than test ones (04:26:27 PM) mathiaz: faulkes-: correct - this is targeted at development versions. (04:26:32 PM) owh: Sure, but they're the same audience I would have thought. (04:26:43 PM) faulkes-: however, I see the reverse of that, in which we could bring in more people who have to support -server (04:26:55 PM) ***faulkes- nods (04:27:11 PM) owh: Hence my query :) (04:27:28 PM) mathiaz: I'll send an email to -server to bring more people on board for the ServerTestingTeam (04:27:40 PM) faulkes-: I think for now, we should leave it seperate and let mathiaz start (04:28:01 PM) mathiaz: [ACTION] mathiaz to send an email about the ServerTestingTeam (04:28:01 PM) MootBot: ACTION received: mathiaz to send an email about the ServerTestingTeam (04:28:32 PM) mathiaz: Let's move on and Review Roadmap and ReportingPage. (04:28:41 PM) mathiaz: https://wiki.ubuntu.com/ServerTeam/Roadmap (04:28:47 PM) mathiaz: https://wiki.ubuntu.com/ServerTeam/ReportingPage (04:29:12 PM) mathiaz: So I'd like to go through the items that are on the Roadmap but haven't any status on the ReportingPage (04:30:04 PM) owh: Is there a way to combine the two pages so it becomes more obvious? (04:30:16 PM) owh: Or is that counter productive? (04:30:32 PM) mathiaz: owh: well - I'm still trying to figure out how we can do that correctly (04:30:47 PM) mathiaz: so I'm still experimenting with this (04:31:00 PM) mathiaz: The ReportingPage is something new (04:31:04 PM) owh: We could make a template that "encourages" correct editing. (04:31:34 PM) mathiaz: [TOPIC] iscsi support (04:31:35 PM) MootBot: New Topic: iscsi support (04:31:43 PM) mathiaz: soren: any news on that ? (04:32:06 PM) soren: YEah. (04:32:25 PM) soren: Um... It seems that there are a few use cases that I had not taken into account when I implemented it. (04:32:46 PM) soren: Booting off of the network and having your root filesystem on iscsi for instance. (04:33:14 PM) soren: So.. That's still work that probably needs to be done. (04:33:27 PM) mathiaz: soren: would this be for hardy ? (04:33:34 PM) mathiaz: soren: isn't that a new feature ? (04:33:42 PM) soren: mathiaz: Yes. Yes, it is. (04:34:00 PM) soren: mathiaz: WEll, from certain users' perspective it's a bug. (04:34:26 PM) soren: I'll have to take it up with dendrobates and slangasek. (04:35:04 PM) owh: mathiaz: Isn't it only a bug if someone looses functionality, aren't we talking about *new* iscsi support? (04:35:20 PM) mathiaz: [ACTION] soren to talk with dendrobates and slangasek about iscsi support for root fs (04:35:21 PM) MootBot: ACTION received: soren to talk with dendrobates and slangasek about iscsi support for root fs (04:35:30 PM) owh: I mean, at present it's just "not yet supported". (04:35:50 PM) mathiaz: owh: correct - this is why it needs to be discussed with the release team. (04:36:03 PM) soren: owh: It's a grey area. (04:36:08 PM) owh: Ah. (04:36:13 PM) ***owh shuts up. (04:36:14 PM) mathiaz: owh: we could try to get a FFexception. (04:36:17 PM) soren: Just because something has never worked doesn't mean it's a feature. (04:36:39 PM) mathiaz: [TOPIC] Fix bugs marked by the QATeam (04:36:40 PM) MootBot: New Topic: Fix bugs marked by the QATeam (04:36:42 PM) soren: Er... crap, that came out wrong. (04:36:49 PM) owh: I understand, I just figured we're better off making sure that what there is is working. (04:37:06 PM) mathiaz: The qa team has put up a list of bug for hardy (04:37:07 PM) mathiaz: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=qa-hardy-server (04:37:22 PM) mathiaz: kirkland: did you fix the nfs bug ? (04:37:27 PM) faulkes-: at least from my experience, booting from san for root fs has been around quite awhile, especially via pxe (04:37:46 PM) kirkland: mathiaz: affirmative, with the help of kees, and inifinity (04:38:11 PM) kirkland: mathiaz: no longer in that list ;-) woohoo (04:38:33 PM) mathiaz: kirkland: great ! (04:38:38 PM) mathiaz: there is an ntp bug left now (04:39:05 PM) owh: Isn't that just a case of changing the init.d order for ntp? (04:39:13 PM) kirkland: mathiaz: I was thinking that it could be solved by similar locking (04:39:28 PM) mathiaz: owh: I don't think so. (04:40:09 PM) kirkland: mathiaz: i think it has to do with NetworkManager (04:40:12 PM) mathiaz: kirkland: are the two bugs caused by the same problem (04:40:22 PM) mathiaz: kirkland: yes - that's what I think too. (04:40:25 PM) kirkland: mathiaz: only distantly... (04:40:32 PM) kirkland: mathiaz: both have upstart implications (04:40:40 PM) mathiaz: so it's an ifup/down script problem may be. (04:40:58 PM) mathiaz: ntp hasn't been designed to run on a laptop/desktop. (04:41:20 PM) owh: It seems to be related to DHCP rather than NetworkManager per se, that is, the ifup/down script. (04:42:10 PM) mathiaz: anyone wants to have a look at the bug ? (04:42:29 PM) zul: not me (04:42:51 PM) faulkes-: I wont likely be able to touch it either (04:42:55 PM) ***soren looks up in the air (04:43:04 PM) owh: Fine. (04:43:08 PM) ***faulkes- whistles innocently, hands behind his back (04:43:10 PM) owh: I'll have a squiz. (04:43:16 PM) owh: Sigh. (04:43:18 PM) owh: :) (04:43:18 PM) kirkland: mathiaz: I don't think it's ifup/ifdown so much (04:43:18 PM) kirkland: but perhaps... (04:43:18 PM) kirkland: mathiaz: should I tackle that bug? (04:43:18 PM) kirkland: (I hadn't been concentrating on it) (04:43:46 PM) mathiaz: [ACTION] owh to look into the ntp bug on the qa-server tag list (04:43:47 PM) MootBot: ACTION received: owh to look into the ntp bug on the qa-server tag list (04:44:01 PM) owh: Damn, too slow :) (04:44:01 PM) mathiaz: kirkland: too late ;) - but you can still help out owh (04:44:16 PM) ***owh accepts any help:) (04:44:28 PM) mathiaz: [TOPIC] Virtualization (04:44:29 PM) MootBot: New Topic: Virtualization (04:44:30 PM) kirkland: ;-) ping me owh (04:44:51 PM) mathiaz: soren: any news ? (04:45:09 PM) soren: Yes. (04:45:19 PM) soren: kvm 62 was uploaded. (04:45:54 PM) soren: I've just now (within the last 47 seconds) discovered an issie with virtio_net in it which I'm working on. (04:46:07 PM) soren: I've sent the corresponding updates to the kernel to the kernel team. (04:46:19 PM) soren: It'll be pulled after the alpha releases. (04:46:42 PM) soren: I'm also looking into some last minute virtio fun to include in the guest kernel. (04:46:49 PM) soren: The paravirt clock, to be exact. (04:46:59 PM) zul: sounds scarey (04:47:06 PM) soren: Other htan that, I'm going into mad bug fixing mode. (04:47:34 PM) soren: zul: I pretty much hooks up your guest clock to the host's ditto. And that's the end of sucky timing issues in virtual machines. (04:47:47 PM) mathiaz: soren: great (04:47:56 PM) mathiaz: [TOPIC] Windows authentication integration (04:47:57 PM) MootBot: New Topic: Windows authentication integration (04:48:12 PM) zul: soren: how well tested is it? does it report clock drifts and the like? (04:48:33 PM) mathiaz: dendrobates: is there a new version likewise-open ready to be tested ? (04:48:42 PM) soren: zul: That's the sort of stuff I need to check up on first. (04:48:53 PM) zul: soren: yah :) (04:48:53 PM) mathiaz: people started to test what's in hardy and reported bugs. (04:49:00 PM) dendrobates: mathiaz: not yet. waiting on upstream. (04:49:15 PM) dendrobates: mathiaz: was supposed to be today. (04:50:04 PM) mathiaz: Ok - I think that's all for the Roadmap. (04:50:14 PM) mathiaz: [TOPIC] Any other business (04:50:14 PM) MootBot: New Topic: Any other business (04:50:23 PM) mathiaz: anyone wants to add something ? (04:50:31 PM) zul: ebox had a ffe filed today (04:50:54 PM) zul: but nothing from me (04:51:09 PM) mathiaz: nealmcb: could you add an factoid for servergui ? (04:51:19 PM) mathiaz: nealmcb: I've added an item about it on the Roadmap (04:51:36 PM) sommer: ivoks mentioned that he's been sick for the last 10 days, but should be back soon (04:52:22 PM) sommer: to work on bacula, etc (04:53:19 PM) owh: Does this mean we have a meeting that actually finishes in less than an hour? (04:53:30 PM) mathiaz: [TOPIC] Agree on next meeting date and time. (04:53:31 PM) MootBot: New Topic: Agree on next meeting date and time. (04:53:42 PM) mathiaz: same time, same place, next week ? (04:53:47 PM) owh: Yup (04:53:53 PM) kirkland: +1 (04:53:59 PM) sommer: o// (04:54:07 PM) owh: These early mornings are going to kill me one day :) (04:54:56 PM) mathiaz: alright then. See ya next week (04:55:02 PM) mathiaz: thanks all for being here. (04:55:11 PM) mathiaz: And happy alpha6 testing ! :) (04:55:23 PM) sommer: thanks mathiaz, later all (04:55:32 PM) owh: Thanks mathiaz (04:55:43 PM) mathiaz: #endmeeting (04:55:44 PM) MootBot: Meeting finished at 21:55. |
Started logging meeting in #ubuntu-meeting [21:01:08] <faulkes-> lead away mathiaz [21:01:18] <mathiaz> Today's agenda can be found online: https://wiki.ubuntu.com/ServerTeam/Meeting [21:02:28] <mathiaz> [TOPIC] Review ACTION points from previous meeting. [21:02:46] <mathiaz> Previous meeting: https://wiki.ubuntu.com/MeetingLogs/Server/20080227 [21:03:15] <mathiaz> [TOPIC] forum reporting [21:03:33] <faulkes-> ok then [21:03:39] <mathiaz> anything to add or disscuss on the forum front faulkes- ? [21:03:46] <faulkes-> yes [21:03:56] <mathiaz> I think you've published the code on LP [21:04:06] <faulkes-> I did a complete rewrite of the code and published it to LP [21:04:16] <faulkes-> https://launchpad.net/forumstats/ [21:04:37] <mathiaz> faulkes-: great - do you have some documentation somewhere ? [21:04:47] <faulkes-> it is much more exact in the nature of how it selects and adds stuff [21:05:00] <faulkes-> there is a README file supplied with the source [21:05:17] <faulkes-> and I have documented the source as well, it's only about 120 lines long [21:05:29] <mathiaz> faulkes-: great. So what's the next step ? [21:05:50] <mathiaz> faulkes-: do you want to add more features to it ? [21:06:01] <faulkes-> the next step is linking the individual graphs and categories, to being able to see the individual posts [21:06:06] <mathiaz> faulkes-: or should we think about what we can do with the outcome ? [21:06:42] <faulkes-> which should be fairly easy to do, I have been very busy the last week with pm course exams and taking on a new FT position [21:07:14] <faulkes-> the outcome, is more a time based situation, I will also add in the ability to select by date [21:07:44] <mathiaz> faulkes-: ok. Seems great. [21:07:49] <faulkes-> because the posts affecting something from time X to time Y, may be unrelated due to version from time Z to time L [21:08:00] <owh> faulkes-: Is there a way to host the current branch so it can display results? [21:08:36] <owh> As in, be "live" [21:08:38] <faulkes-> that may be possible but it would require access to my database [21:08:57] <faulkes-> if you want to host one yourself, you can run the code completely independent of me [21:09:10] <faulkes-> and allow it to gather stats [21:09:20] <faulkes-> the requirements are listed in the README for running it [21:09:22] <owh> Does the code have a means to generate a database as well? [21:09:47] <faulkes-> there is sql files included in the LP branch [21:09:52] <owh> Cool. [21:10:06] <faulkes-> the README has instructions regarding them [21:10:19] <owh> Excellent, documentation that actually helps :) [21:10:32] <mathiaz> How is the classification of post done ? [21:10:44] <mathiaz> faulkes-: I think this is the trickiest part [21:11:01] <faulkes-> classification is done via a dictionary [21:11:10] <faulkes-> unfortunately, that means building the dictionary [21:11:19] <faulkes-> I have included the one which is currently used [21:11:35] <mathiaz> faulkes-: ok - is there a way to reassign posts to a different category if they've been misclassifed ? [21:11:49] <faulkes-> so, if you wanted to use the code to monitor say, Desktop forum, you would need to build your own [21:12:10] <faulkes-> mathiaz: currently no, that would be an admin feature I guess, but wouldn't be hard to build in [21:12:24] <faulkes-> as well as add additional categories, terms, etc.. [21:12:55] <owh> Going that route would then require a sort of "meta" individual, that seems like a lot of work. [21:12:56] <faulkes-> right now, it is by hand, mostly for lack of time on my part to build that in but it would be a fairly straight forward excersize [21:13:02] <mathiaz> faulkes-: These are more features. Let's try to get something done first and see how can extract information from that. [21:13:32] <mathiaz> Let's move on. [21:13:37] <mathiaz> [TOPIC] Server survey [21:13:56] <owh> That would be me. [21:13:58] <faulkes-> me again I guess as nijaba is away or do you want to handle it mathiaz? [21:14:00] <mathiaz> nijaba is not around. [21:14:14] <faulkes-> I can go from his action email, that you updated [21:14:22] <mathiaz> I've updated the ReportingPage with a status he sent me. [21:14:38] <mathiaz> Anything else to add ? [21:14:39] <mathiaz> https://wiki.ubuntu.com/ServerTeam/ReportingPage [21:15:03] <faulkes-> sec [21:15:40] <owh> While faulkes- is looking... [21:15:42] <owh> I created https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch which has as many contacts in it as possible, but suggestions are welcome. The intent is to use the body text as the email text. [21:15:46] <owh> I tried coming up with actual press contacts where relevant, but I don't have any actual Canonical press contact. [21:15:50] <owh> I'm not sure how we best ask Server Team Members to blog about the survey so it shows up on Planet Ubuntu. [21:15:51] <mathiaz> It seems that the survey is too long to take. [21:16:04] <owh> Yes, I'd agree with that. [21:16:17] <faulkes-> well, we've only had a small sampling so far [21:16:25] <faulkes-> however [21:16:50] <faulkes-> we should be always reporting feature/bug requests to LP, I don't generally check the page itself [21:16:58] <faulkes-> so there is some stuff there that I can work on [21:17:14] <faulkes-> as for shortening the survey, I think we can do that effectively [21:17:23] <kirkland> mathiaz: I've communicated with Jonathan Corbet of LWN before, I can send him a note asking if it would be possible to publicize on LWN.net [21:17:44] <kirkland> no promises--he only sometimes responds to my email :-) [21:17:46] <faulkes-> I will look at tuning the questions so it is faster [21:17:56] <owh> kirkland: Add it to the wiki page, you don't need to give the email address if you're not comfortable with it. [21:18:01] <mathiaz> kirkland: seems great - could you add that to https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch ? [21:18:12] <owh> kirkland: That way we don't hit LWN multiple times. [21:18:52] <owh> faulkes-: I wonder if the survey length is a function of a poorly defined audience/spec. I'm not saying it is, I'm just wondering. [21:19:02] <mathiaz> [ACTION] faulkes- to work on tuning the questions to make the survey shorter [21:19:07] <owh> faulkes-: It seems to be sprawling all over the place. [21:19:28] <mathiaz> owh: faulkes-: could you coordinate with nijaba ? [21:19:40] <owh> Yup [21:19:40] <faulkes-> there is some logical order which needs to be done on the questions [21:19:53] <faulkes-> mathiaz: yes, I will speak to nijaba [21:19:57] <kirkland> mathiaz: owh: done. [21:20:03] <mathiaz> owh: faulkes-: thanks - [21:20:06] <mathiaz> Let's move on. [21:20:12] <owh> Perhaps we can catch up afterwards faulkes- and have a chat about it. [21:20:19] <faulkes-> certainly [21:20:49] <mathiaz> [TOPIC] ServerTestingTeam [21:21:11] <mathiaz> So I spent some time to revamp the ServerTestingTeam wiki pages. [21:21:28] <mathiaz> https://wiki.ubuntu.com/ServerTestingTeam [21:21:58] <mathiaz> The idea is to increase hardware testing for hardy. [21:22:12] <mathiaz> Reporting the results is done in the wiki. [21:22:39] <faulkes-> interesting, ok, I may be able to add some to that [21:22:53] <mathiaz> I'll also send an email to the previous people involved in the team and ask if they can still contribute. [21:23:10] <mathiaz> Anyone that has some server hardware is welcome to participate. [21:23:30] <faulkes-> for those unaware of my current status, I've moved from being a consultant to a FTE job, I will be converting the current environment to -server [21:23:34] <mathiaz> The idea is to download an iso and make sure the installation is successfll [21:24:05] <mathiaz> Yet another way to contribute to the Server Team :) [21:24:06] <faulkes-> mathiaz: I should have extra hardware resources now available to put to it [21:24:32] <mathiaz> faulkes-: great ! I've updated the wiki page to streamline the reporting process [21:25:12] <owh> mathiaz: That's looking great. Do you think it would be useful to add that link to the Survey Announcement to encourage more helpers, or might that be counter productive? [21:25:29] <mathiaz> owh: I'll make a separate announcement [21:25:34] <owh> Given that we're targeting -server users. [21:25:34] <faulkes-> mathiaz: will ServerTestingTeam be a subset of the ServerTeam for membership purposes or? [21:25:51] <mathiaz> faulkes-: for now, there isn't any LP team or membership [21:26:00] <mathiaz> It may be usefull at some point. [21:26:07] <faulkes-> owh: well, I think the survey is geared more to established environments rather than test ones [21:26:27] <mathiaz> faulkes-: correct - this is targeted at development versions. [21:26:32] <owh> Sure, but they're the same audience I would have thought. [21:26:43] <faulkes-> however, I see the reverse of that, in which we could bring in more people who have to support -server [21:26:56] * faulkes- nods [21:27:11] <owh> Hence my query :) [21:27:28] <mathiaz> I'll send an email to -server to bring more people on board for the ServerTestingTeam [21:27:41] <faulkes-> I think for now, we should leave it seperate and let mathiaz start [21:28:01] <mathiaz> [ACTION] mathiaz to send an email about the ServerTestingTeam [21:28:32] <mathiaz> Let's move on and Review Roadmap and ReportingPage. [21:28:42] <mathiaz> https://wiki.ubuntu.com/ServerTeam/Roadmap [21:28:47] <mathiaz> https://wiki.ubuntu.com/ServerTeam/ReportingPage [21:29:12] <mathiaz> So I'd like to go through the items that are on the Roadmap but haven't any status on the ReportingPage [21:30:04] <owh> Is there a way to combine the two pages so it becomes more obvious? [21:30:16] <owh> Or is that counter productive? [21:30:32] <mathiaz> owh: well - I'm still trying to figure out how we can do that correctly [21:30:47] <mathiaz> so I'm still experimenting with this [21:31:00] <mathiaz> The ReportingPage is something new [21:31:04] <owh> We could make a template that "encourages" correct editing. [21:31:35] <mathiaz> [TOPIC] iscsi support [21:31:43] <mathiaz> soren: any news on that ? [21:32:06] <soren> YEah. [21:32:25] <soren> Um... It seems that there are a few use cases that I had not taken into account when I implemented it. [21:32:47] <soren> Booting off of the network and having your root filesystem on iscsi for instance. [21:33:14] <soren> So.. That's still work that probably needs to be done. [21:33:27] <mathiaz> soren: would this be for hardy ? [21:33:34] <mathiaz> soren: isn't that a new feature ? [21:33:42] <soren> mathiaz: Yes. Yes, it is. [21:34:00] <soren> mathiaz: WEll, from certain users' perspective it's a bug. [21:34:26] <soren> I'll have to take it up with dendrobates and slangasek. [21:35:04] <owh> mathiaz: Isn't it only a bug if someone looses functionality, aren't we talking about *new* iscsi support? [21:35:20] <mathiaz> [ACTION] soren to talk with dendrobates and slangasek about iscsi support for root fs [21:35:30] <owh> I mean, at present it's just "not yet supported". [21:35:51] <mathiaz> owh: correct - this is why it needs to be discussed with the release team. [21:36:03] <soren> owh: It's a grey area. [21:36:08] <owh> Ah. [21:36:14] * owh shuts up. [21:36:14] <mathiaz> owh: we could try to get a FFexception. [21:36:17] <soren> Just because something has never worked doesn't mean it's a feature. [21:36:39] <mathiaz> [TOPIC] Fix bugs marked by the QATeam [21:36:42] <soren> Er... crap, that came out wrong. [21:36:49] <owh> I understand, I just figured we're better off making sure that what there is is working. [21:37:06] <mathiaz> The qa team has put up a list of bug for hardy [21:37:08] <mathiaz> https://bugs.launchpad.net/ubuntu/+bugs?field.tag=qa-hardy-server [21:37:22] <mathiaz> kirkland: did you fix the nfs bug ? [21:37:28] <faulkes-> at least from my experience, booting from san for root fs has been around quite awhile, especially via pxe [21:37:46] <kirkland> mathiaz: affirmative, with the help of kees, and inifinity [21:38:11] <kirkland> mathiaz: no longer in that list ;-) woohoo [21:38:33] <mathiaz> kirkland: great ! [21:38:38] <mathiaz> there is an ntp bug left now [21:39:05] <owh> Isn't that just a case of changing the init.d order for ntp? [21:39:13] <kirkland> mathiaz: I was thinking that it could be solved by similar locking [21:39:28] <mathiaz> owh: I don't think so. [21:40:09] <kirkland> mathiaz: i think it has to do with NetworkManager [21:40:12] <mathiaz> kirkland: are the two bugs caused by the same problem [21:40:23] <mathiaz> kirkland: yes - that's what I think too. [21:40:25] <kirkland> mathiaz: only distantly... [21:40:32] <kirkland> mathiaz: both have upstart implications [21:40:40] <mathiaz> so it's an ifup/down script problem may be. [21:40:59] <mathiaz> ntp hasn't been designed to run on a laptop/desktop. [21:41:20] <owh> It seems to be related to DHCP rather than NetworkManager per se, that is, the ifup/down script. [21:42:10] <mathiaz> anyone wants to have a look at the bug ? [21:42:29] <zul> not me [21:42:51] <faulkes-> I wont likely be able to touch it either [21:42:55] * soren looks up in the air [21:43:04] <owh> Fine. [21:43:08] * faulkes- whistles innocently, hands behind his back [21:43:10] <owh> I'll have a squiz. [21:43:16] <owh> Sigh. [21:43:18] <owh> :) [21:43:18] <kirkland> mathiaz: I don't think it's ifup/ifdown so much [21:43:18] <kirkland> but perhaps... [21:43:18] <kirkland> mathiaz: should I tackle that bug? [21:43:18] <kirkland> (I hadn't been concentrating on it) [21:43:46] <mathiaz> [ACTION] owh to look into the ntp bug on the qa-server tag list [21:43:55] <owh> Damn, too slow :) [21:44:01] <mathiaz> kirkland: too late ;) - but you can still help out owh [21:44:16] * owh accepts any help:) [21:44:28] <mathiaz> [TOPIC] Virtualization [21:44:30] <kirkland> ;-) ping me owh [21:44:51] <mathiaz> soren: any news ? [21:45:09] <soren> Yes. [21:45:19] <soren> kvm 62 was uploaded. [21:45:54] <soren> I've just now (within the last 47 seconds) discovered an issie with virtio_net in it which I'm working on. [21:46:07] <soren> I've sent the corresponding updates to the kernel to the kernel team. [21:46:19] <soren> It'll be pulled after the alpha releases. [21:46:42] <soren> I'm also looking into some last minute virtio fun to include in the guest kernel. [21:46:49] <soren> The paravirt clock, to be exact. [21:46:59] <zul> sounds scarey [21:47:06] <soren> Other htan that, I'm going into mad bug fixing mode. [21:47:34] <soren> zul: I pretty much hooks up your guest clock to the host's ditto. And that's the end of sucky timing issues in virtual machines. [21:47:47] <mathiaz> soren: great [21:47:56] <mathiaz> [TOPIC] Windows authentication integration [21:48:12] <zul> soren: how well tested is it? does it report clock drifts and the like? [21:48:33] <mathiaz> dendrobates: is there a new version likewise-open ready to be tested ? [21:48:42] <soren> zul: That's the sort of stuff I need to check up on first. [21:48:53] <zul> soren: yah :) [21:48:53] <mathiaz> people started to test what's in hardy and reported bugs. [21:49:00] <dendrobates> mathiaz: not yet. waiting on upstream. [21:49:15] <dendrobates> mathiaz: was supposed to be today. [21:50:04] <mathiaz> Ok - I think that's all for the Roadmap. [21:50:14] <mathiaz> [TOPIC] Any other business [21:50:23] <mathiaz> anyone wants to add something ? [21:50:31] <zul> ebox had a ffe filed today [21:50:54] <zul> but nothing from me [21:51:09] <mathiaz> nealmcb: could you add an factoid for servergui ? [21:51:19] <mathiaz> nealmcb: I've added an item about it on the Roadmap [21:51:37] <sommer> ivoks mentioned that he's been sick for the last 10 days, but should be back soon [21:52:22] <sommer> to work on bacula, etc [21:53:19] <owh> Does this mean we have a meeting that actually finishes in less than an hour? [21:53:30] <mathiaz> [TOPIC] Agree on next meeting date and time. [21:53:42] <mathiaz> same time, same place, next week ? [21:53:47] <owh> Yup [21:53:53] <kirkland> +1 [21:53:59] <sommer> o// [21:54:07] <owh> These early mornings are going to kill me one day :) [21:54:56] <mathiaz> alright then. See ya next week [21:55:02] <mathiaz> thanks all for being here. [21:55:11] <mathiaz> And happy alpha6 testing ! :) [21:55:23] <sommer> thanks mathiaz, later all [21:55:32] <owh> Thanks mathiaz [21:55:43] <mathiaz> #endmeeting Meeting ended. |
Agenda
Items we will be discussing:
- Review ACTION points from previous meeting.
- Server survey - too long. Should we cut some questions ?
Review ["ServerTeam/Roadmap"] and ["ServerTeam/ReportingPage"].
- Agree on next meeting date and time.
Minutes
Forum Reporting
faulkes- created a project on Launchpad [1] for his forum analyser [1]. There is a README file in the src directory outlining how to install the software.
[1]: https://launchpad.net/forumstats/ [2]: http://ubuntu.oss-mgmt.com/
Server survey
owh drafted an announcement for the survey [3]. He also added a list of resources to which the announcement should be sent.
[3]: https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch
nijaba reported that testers found the survey too long to complete.
ACTION: faulkes-, owh and nijaba to work on tuning the questions to make the survey shorter.
ServerTestingTeam
mathiaz revamped the ServerTestingTeam wiki pages [4]. The process to run tests and report results has been streamlined. He will send a separate announcement to ubuntu-server about the ServerTestingTeam.
[4]: https://wiki.ubuntu.com/ServerTestingTeam
ACTION: mathiaz to send an email about the ServerTestingTeam
iSCSI Support
soren has come across a few use cases he hadn't taken into account: Booting off of the network and having the root filesystem on iscsi. He'll discuss supporting this scenario with dendrobates and slangasek.
ACTION: soren to talk with dendrobates and slangasek about iscsi support for root fs.
Fix bugs marked by the QATeam
The QA team maintains a list of bugs that should be fixed for hardy [5]. kirkland fixed one bug related to nfs. owh and kirkland will take a look at the ntp bug.
[5]: https://bugs.launchpad.net/ubuntu/+bugs?field.tag=qa-hardy-server.
Virtualization
soren uploaded kvm 62. He is also looking into adding parvirt clock to the kernel which should improve timing issues in virtual machines.
Windows Integration
dendrobates is waiting for a new release from upstream.
Ebox
zul filed a FeatureFreezeException request, which has been granted. A new version of ebox is available in universe - testers are welcomed !
Agree on next meeting date and time
Next meeting will be on Wednesday, March 12th at 21:00 UTC in #ubuntu-meeting.
Log
Started logging meeting in #ubuntu-meeting [21:01:08] <faulkes-> lead away mathiaz [21:01:18] <mathiaz> Today's agenda can be found online: https://wiki.ubuntu.com/ServerTeam/Meeting [21:02:28] <mathiaz> [TOPIC] Review ACTION points from previous meeting. [21:02:46] <mathiaz> Previous meeting: https://wiki.ubuntu.com/MeetingLogs/Server/20080227 [21:03:15] <mathiaz> [TOPIC] forum reporting [21:03:33] <faulkes-> ok then [21:03:39] <mathiaz> anything to add or disscuss on the forum front faulkes- ? [21:03:46] <faulkes-> yes [21:03:56] <mathiaz> I think you've published the code on LP [21:04:06] <faulkes-> I did a complete rewrite of the code and published it to LP [21:04:16] <faulkes-> https://launchpad.net/forumstats/ [21:04:37] <mathiaz> faulkes-: great - do you have some documentation somewhere ? [21:04:47] <faulkes-> it is much more exact in the nature of how it selects and adds stuff [21:05:00] <faulkes-> there is a README file supplied with the source [21:05:17] <faulkes-> and I have documented the source as well, it's only about 120 lines long [21:05:29] <mathiaz> faulkes-: great. So what's the next step ? [21:05:50] <mathiaz> faulkes-: do you want to add more features to it ? [21:06:01] <faulkes-> the next step is linking the individual graphs and categories, to being able to see the individual posts [21:06:06] <mathiaz> faulkes-: or should we think about what we can do with the outcome ? [21:06:42] <faulkes-> which should be fairly easy to do, I have been very busy the last week with pm course exams and taking on a new FT position [21:07:14] <faulkes-> the outcome, is more a time based situation, I will also add in the ability to select by date [21:07:44] <mathiaz> faulkes-: ok. Seems great. [21:07:49] <faulkes-> because the posts affecting something from time X to time Y, may be unrelated due to version from time Z to time L [21:08:00] <owh> faulkes-: Is there a way to host the current branch so it can display results? [21:08:36] <owh> As in, be "live" [21:08:38] <faulkes-> that may be possible but it would require access to my database [21:08:57] <faulkes-> if you want to host one yourself, you can run the code completely independent of me [21:09:10] <faulkes-> and allow it to gather stats [21:09:20] <faulkes-> the requirements are listed in the README for running it [21:09:22] <owh> Does the code have a means to generate a database as well? [21:09:47] <faulkes-> there is sql files included in the LP branch [21:09:52] <owh> Cool. [21:10:06] <faulkes-> the README has instructions regarding them [21:10:19] <owh> Excellent, documentation that actually helps :) [21:10:32] <mathiaz> How is the classification of post done ? [21:10:44] <mathiaz> faulkes-: I think this is the trickiest part [21:11:01] <faulkes-> classification is done via a dictionary [21:11:10] <faulkes-> unfortunately, that means building the dictionary [21:11:19] <faulkes-> I have included the one which is currently used [21:11:35] <mathiaz> faulkes-: ok - is there a way to reassign posts to a different category if they've been misclassifed ? [21:11:49] <faulkes-> so, if you wanted to use the code to monitor say, Desktop forum, you would need to build your own [21:12:10] <faulkes-> mathiaz: currently no, that would be an admin feature I guess, but wouldn't be hard to build in [21:12:24] <faulkes-> as well as add additional categories, terms, etc.. [21:12:55] <owh> Going that route would then require a sort of "meta" individual, that seems like a lot of work. [21:12:56] <faulkes-> right now, it is by hand, mostly for lack of time on my part to build that in but it would be a fairly straight forward excersize [21:13:02] <mathiaz> faulkes-: These are more features. Let's try to get something done first and see how can extract information from that. [21:13:32] <mathiaz> Let's move on. [21:13:37] <mathiaz> [TOPIC] Server survey [21:13:56] <owh> That would be me. [21:13:58] <faulkes-> me again I guess as nijaba is away or do you want to handle it mathiaz? [21:14:00] <mathiaz> nijaba is not around. [21:14:14] <faulkes-> I can go from his action email, that you updated [21:14:22] <mathiaz> I've updated the ReportingPage with a status he sent me. [21:14:38] <mathiaz> Anything else to add ? [21:14:39] <mathiaz> https://wiki.ubuntu.com/ServerTeam/ReportingPage [21:15:03] <faulkes-> sec [21:15:40] <owh> While faulkes- is looking... [21:15:42] <owh> I created https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch which has as many contacts in it as possible, but suggestions are welcome. The intent is to use the body text as the email text. [21:15:46] <owh> I tried coming up with actual press contacts where relevant, but I don't have any actual Canonical press contact. [21:15:50] <owh> I'm not sure how we best ask Server Team Members to blog about the survey so it shows up on Planet Ubuntu. [21:15:51] <mathiaz> It seems that the survey is too long to take. [21:16:04] <owh> Yes, I'd agree with that. [21:16:17] <faulkes-> well, we've only had a small sampling so far [21:16:25] <faulkes-> however [21:16:50] <faulkes-> we should be always reporting feature/bug requests to LP, I don't generally check the page itself [21:16:58] <faulkes-> so there is some stuff there that I can work on [21:17:14] <faulkes-> as for shortening the survey, I think we can do that effectively [21:17:23] <kirkland> mathiaz: I've communicated with Jonathan Corbet of LWN before, I can send him a note asking if it would be possible to publicize on LWN.net [21:17:44] <kirkland> no promises--he only sometimes responds to my email :-) [21:17:46] <faulkes-> I will look at tuning the questions so it is faster [21:17:56] <owh> kirkland: Add it to the wiki page, you don't need to give the email address if you're not comfortable with it. [21:18:01] <mathiaz> kirkland: seems great - could you add that to https://wiki.ubuntu.com/ServerTeam/ServerSurveyLaunch ? [21:18:12] <owh> kirkland: That way we don't hit LWN multiple times. [21:18:52] <owh> faulkes-: I wonder if the survey length is a function of a poorly defined audience/spec. I'm not saying it is, I'm just wondering. [21:19:02] <mathiaz> [ACTION] faulkes- to work on tuning the questions to make the survey shorter [21:19:07] <owh> faulkes-: It seems to be sprawling all over the place. [21:19:28] <mathiaz> owh: faulkes-: could you coordinate with nijaba ? [21:19:40] <owh> Yup [21:19:40] <faulkes-> there is some logical order which needs to be done on the questions [21:19:53] <faulkes-> mathiaz: yes, I will speak to nijaba [21:19:57] <kirkland> mathiaz: owh: done. [21:20:03] <mathiaz> owh: faulkes-: thanks - [21:20:06] <mathiaz> Let's move on. [21:20:12] <owh> Perhaps we can catch up afterwards faulkes- and have a chat about it. [21:20:19] <faulkes-> certainly [21:20:49] <mathiaz> [TOPIC] ServerTestingTeam [21:21:11] <mathiaz> So I spent some time to revamp the ServerTestingTeam wiki pages. [21:21:28] <mathiaz> https://wiki.ubuntu.com/ServerTestingTeam [21:21:58] <mathiaz> The idea is to increase hardware testing for hardy. [21:22:12] <mathiaz> Reporting the results is done in the wiki. [21:22:39] <faulkes-> interesting, ok, I may be able to add some to that [21:22:53] <mathiaz> I'll also send an email to the previous people involved in the team and ask if they can still contribute. [21:23:10] <mathiaz> Anyone that has some server hardware is welcome to participate. [21:23:30] <faulkes-> for those unaware of my current status, I've moved from being a consultant to a FTE job, I will be converting the current environment to -server [21:23:34] <mathiaz> The idea is to download an iso and make sure the installation is successfll [21:24:05] <mathiaz> Yet another way to contribute to the Server Team :) [21:24:06] <faulkes-> mathiaz: I should have extra hardware resources now available to put to it [21:24:32] <mathiaz> faulkes-: great ! I've updated the wiki page to streamline the reporting process [21:25:12] <owh> mathiaz: That's looking great. Do you think it would be useful to add that link to the Survey Announcement to encourage more helpers, or might that be counter productive? [21:25:29] <mathiaz> owh: I'll make a separate announcement [21:25:34] <owh> Given that we're targeting -server users. [21:25:34] <faulkes-> mathiaz: will ServerTestingTeam be a subset of the ServerTeam for membership purposes or? [21:25:51] <mathiaz> faulkes-: for now, there isn't any LP team or membership [21:26:00] <mathiaz> It may be usefull at some point. [21:26:07] <faulkes-> owh: well, I think the survey is geared more to established environments rather than test ones [21:26:27] <mathiaz> faulkes-: correct - this is targeted at development versions. [21:26:32] <owh> Sure, but they're the same audience I would have thought. [21:26:43] <faulkes-> however, I see the reverse of that, in which we could bring in more people who have to support -server [21:26:56] * faulkes- nods [21:27:11] <owh> Hence my query :) [21:27:28] <mathiaz> I'll send an email to -server to bring more people on board for the ServerTestingTeam [21:27:41] <faulkes-> I think for now, we should leave it seperate and let mathiaz start [21:28:01] <mathiaz> [ACTION] mathiaz to send an email about the ServerTestingTeam [21:28:32] <mathiaz> Let's move on and Review Roadmap and ReportingPage. [21:28:42] <mathiaz> https://wiki.ubuntu.com/ServerTeam/Roadmap [21:28:47] <mathiaz> https://wiki.ubuntu.com/ServerTeam/ReportingPage [21:29:12] <mathiaz> So I'd like to go through the items that are on the Roadmap but haven't any status on the ReportingPage [21:30:04] <owh> Is there a way to combine the two pages so it becomes more obvious? [21:30:16] <owh> Or is that counter productive? [21:30:32] <mathiaz> owh: well - I'm still trying to figure out how we can do that correctly [21:30:47] <mathiaz> so I'm still experimenting with this [21:31:00] <mathiaz> The ReportingPage is something new [21:31:04] <owh> We could make a template that "encourages" correct editing. [21:31:35] <mathiaz> [TOPIC] iscsi support [21:31:43] <mathiaz> soren: any news on that ? [21:32:06] <soren> YEah. [21:32:25] <soren> Um... It seems that there are a few use cases that I had not taken into account when I implemented it. [21:32:47] <soren> Booting off of the network and having your root filesystem on iscsi for instance. [21:33:14] <soren> So.. That's still work that probably needs to be done. [21:33:27] <mathiaz> soren: would this be for hardy ? [21:33:34] <mathiaz> soren: isn't that a new feature ? [21:33:42] <soren> mathiaz: Yes. Yes, it is. [21:34:00] <soren> mathiaz: WEll, from certain users' perspective it's a bug. [21:34:26] <soren> I'll have to take it up with dendrobates and slangasek. [21:35:04] <owh> mathiaz: Isn't it only a bug if someone looses functionality, aren't we talking about *new* iscsi support? [21:35:20] <mathiaz> [ACTION] soren to talk with dendrobates and slangasek about iscsi support for root fs [21:35:30] <owh> I mean, at present it's just "not yet supported". [21:35:51] <mathiaz> owh: correct - this is why it needs to be discussed with the release team. [21:36:03] <soren> owh: It's a grey area. [21:36:08] <owh> Ah. [21:36:14] * owh shuts up. [21:36:14] <mathiaz> owh: we could try to get a FFexception. [21:36:17] <soren> Just because something has never worked doesn't mean it's a feature. [21:36:39] <mathiaz> [TOPIC] Fix bugs marked by the QATeam [21:36:42] <soren> Er... crap, that came out wrong. [21:36:49] <owh> I understand, I just figured we're better off making sure that what there is is working. [21:37:06] <mathiaz> The qa team has put up a list of bug for hardy [21:37:08] <mathiaz> https://bugs.launchpad.net/ubuntu/+bugs?field.tag=qa-hardy-server [21:37:22] <mathiaz> kirkland: did you fix the nfs bug ? [21:37:28] <faulkes-> at least from my experience, booting from san for root fs has been around quite awhile, especially via pxe [21:37:46] <kirkland> mathiaz: affirmative, with the help of kees, and inifinity [21:38:11] <kirkland> mathiaz: no longer in that list ;-) woohoo [21:38:33] <mathiaz> kirkland: great ! [21:38:38] <mathiaz> there is an ntp bug left now [21:39:05] <owh> Isn't that just a case of changing the init.d order for ntp? [21:39:13] <kirkland> mathiaz: I was thinking that it could be solved by similar locking [21:39:28] <mathiaz> owh: I don't think so. [21:40:09] <kirkland> mathiaz: i think it has to do with NetworkManager [21:40:12] <mathiaz> kirkland: are the two bugs caused by the same problem [21:40:23] <mathiaz> kirkland: yes - that's what I think too. [21:40:25] <kirkland> mathiaz: only distantly... [21:40:32] <kirkland> mathiaz: both have upstart implications [21:40:40] <mathiaz> so it's an ifup/down script problem may be. [21:40:59] <mathiaz> ntp hasn't been designed to run on a laptop/desktop. [21:41:20] <owh> It seems to be related to DHCP rather than NetworkManager per se, that is, the ifup/down script. [21:42:10] <mathiaz> anyone wants to have a look at the bug ? [21:42:29] <zul> not me [21:42:51] <faulkes-> I wont likely be able to touch it either [21:42:55] * soren looks up in the air [21:43:04] <owh> Fine. [21:43:08] * faulkes- whistles innocently, hands behind his back [21:43:10] <owh> I'll have a squiz. [21:43:16] <owh> Sigh. [21:43:18] <owh> :) [21:43:18] <kirkland> mathiaz: I don't think it's ifup/ifdown so much [21:43:18] <kirkland> but perhaps... [21:43:18] <kirkland> mathiaz: should I tackle that bug? [21:43:18] <kirkland> (I hadn't been concentrating on it) [21:43:46] <mathiaz> [ACTION] owh to look into the ntp bug on the qa-server tag list [21:43:55] <owh> Damn, too slow :) [21:44:01] <mathiaz> kirkland: too late ;) - but you can still help out owh [21:44:16] * owh accepts any help:) [21:44:28] <mathiaz> [TOPIC] Virtualization [21:44:30] <kirkland> ;-) ping me owh [21:44:51] <mathiaz> soren: any news ? [21:45:09] <soren> Yes. [21:45:19] <soren> kvm 62 was uploaded. [21:45:54] <soren> I've just now (within the last 47 seconds) discovered an issie with virtio_net in it which I'm working on. [21:46:07] <soren> I've sent the corresponding updates to the kernel to the kernel team. [21:46:19] <soren> It'll be pulled after the alpha releases. [21:46:42] <soren> I'm also looking into some last minute virtio fun to include in the guest kernel. [21:46:49] <soren> The paravirt clock, to be exact. [21:46:59] <zul> sounds scarey [21:47:06] <soren> Other htan that, I'm going into mad bug fixing mode. [21:47:34] <soren> zul: I pretty much hooks up your guest clock to the host's ditto. And that's the end of sucky timing issues in virtual machines. [21:47:47] <mathiaz> soren: great [21:47:56] <mathiaz> [TOPIC] Windows authentication integration [21:48:12] <zul> soren: how well tested is it? does it report clock drifts and the like? [21:48:33] <mathiaz> dendrobates: is there a new version likewise-open ready to be tested ? [21:48:42] <soren> zul: That's the sort of stuff I need to check up on first. [21:48:53] <zul> soren: yah :) [21:48:53] <mathiaz> people started to test what's in hardy and reported bugs. [21:49:00] <dendrobates> mathiaz: not yet. waiting on upstream. [21:49:15] <dendrobates> mathiaz: was supposed to be today. [21:50:04] <mathiaz> Ok - I think that's all for the Roadmap. [21:50:14] <mathiaz> [TOPIC] Any other business [21:50:23] <mathiaz> anyone wants to add something ? [21:50:31] <zul> ebox had a ffe filed today [21:50:54] <zul> but nothing from me [21:51:09] <mathiaz> nealmcb: could you add an factoid for servergui ? [21:51:19] <mathiaz> nealmcb: I've added an item about it on the Roadmap [21:51:37] <sommer> ivoks mentioned that he's been sick for the last 10 days, but should be back soon [21:52:22] <sommer> to work on bacula, etc [21:53:19] <owh> Does this mean we have a meeting that actually finishes in less than an hour? [21:53:30] <mathiaz> [TOPIC] Agree on next meeting date and time. [21:53:42] <mathiaz> same time, same place, next week ? [21:53:47] <owh> Yup [21:53:53] <kirkland> +1 [21:53:59] <sommer> o// [21:54:07] <owh> These early mornings are going to kill me one day :) [21:54:56] <mathiaz> alright then. See ya next week [21:55:02] <mathiaz> thanks all for being here. [21:55:11] <mathiaz> And happy alpha6 testing ! :) [21:55:23] <sommer> thanks mathiaz, later all [21:55:32] <owh> Thanks mathiaz [21:55:43] <mathiaz> #endmeeting Meeting ended.
MeetingLogs/Server/20080305 (last edited 2008-08-06 16:23:56 by localhost)