infrastructure July 2008

infrastructure@lists.fedoraproject.org

52 participants
46 discussions

Meeting Log - 2008-07-24
by Ricky Zhou 24 Jul '08

24 Jul '08

16:00 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Who's here? 16:00 * ricky 16:00 < ivazquez> Pong. 16:01 * abadger1999 here 16:01 < fchiulli> * fchiulli lurking about 16:01 < mmcgrath> fchiulli: howdy! 16:01 < mmcgrath> Ok, lets get started. This could be a short meeting. 16:02 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Tickets 16:02 < mmcgrath> .tiny https://fedorahosted.org/fedora-infrastructure/query?status=new&status=assi… 16:02 < zodbot> mmcgrath: http://tinyurl.com/2hyyz6 16:02 < mmcgrath> .ticket 671 16:02 < zodbot> mmcgrath: #671 (in.fedoraproject.org) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/671 16:02 < mmcgrath> anyone representing in.fedoraproject.org around? 16:02 -!- kschreyack [n=kschreya(a)mail.wirelesscapital.com] has joined #fedora-meeting 16:02 < ricky> mether: ping 16:03 < alex[slx]> Just going to watch this one (first time). 16:03 * skvidal is here, but late 16:04 < mmcgrath> Ok, moving on 16:04 < mmcgrath> .ticket 395 16:04 < zodbot> mmcgrath: #395 (Audio Streaming of Fedora Board Conference Calls) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/395 16:04 < mmcgrath> jcollie: you about? 16:05 < mmcgrath> k, moving on :) 16:06 < mmcgrath> .ticket 446 16:06 < zodbot> mmcgrath: #446 (Possibility to add external links on spins page) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/446 16:06 < mmcgrath> dgilmore: you around? 16:06 * mmcgrath suspects OLS 16:06 < mmcgrath> next ticket 16:06 < mmcgrath> .ticket 576 16:06 < zodbot> mmcgrath: #576 (Infrastructure Contact Information) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/576 16:06 < mmcgrath> I don't think there is anything new there 16:07 < mmcgrath> ricky: you heard anything? 16:07 * mmcgrath has been using sshfs and encfs pretty successfully. 16:07 < mmcgrath> scripted its pretty easy to setup 16:07 < ricky> G: I know you've been busy this week - any new thoughts? 16:09 < mmcgrath> heh, I suspect he's gone too 16:09 < mmcgrath> so moving on! 16:09 < mmcgrath> see, I told you this would be short :) 16:09 < ricky> Heh, yeah 16:09 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- DB outage 16:09 < mmcgrath> So the db outage was this week. 16:10 < mmcgrath> uber thanks to ricky! 16:10 < mmcgrath> It went pretty well, bit longer then we'd expected because of a bad first dump. 16:10 < mmcgrath> other then that though. All of the database servers have been putting along quite well 16:10 < mmcgrath> thats really all I have on that 16:10 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- network issues 16:11 < mmcgrath> We continue to see network issues between most of our hosts in the main fedora rack in PHX. 16:11 -!- MadBus [n=scott(a)c-68-51-66-81.hsd1.in.comcast.net] has left #Fedora-meeting ["Leaving"] 16:11 < mmcgrath> some changes went in today that seem to have helpped some hosts, but not others. 16:11 < mmcgrath> nothing really new there. 16:11 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- mmcgrath gone 16:11 < mmcgrath> I'll be gone starting tomorrow till Sunday and I'll be camping. 16:11 < mmcgrath> so don't expect for me to be around much 16:11 < ricky> Cool :-) 16:12 < fchiulli> mmcgrath: May the weather gods shine down on you. 16:12 < mmcgrath> :) 16:12 < mmcgrath> paintball. 16:12 * abadger1999 gets the orbital laser primed to make the batman symbol in case of emergencies :-) 16:12 < mmcgrath> rock climbing 16:12 < mmcgrath> and paintball rock climbing. 16:12 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Open Floor 16:13 < mmcgrath> Who's got stuff to discuss? 16:13 < ricky> Anybody that has stuff on lockbox should start thinking about moving it :-) 16:14 < mmcgrath> ricky: yeah, might want to send an email to the list about that too. I'm pretty sure we're good there though 16:14 < ricky> Once all of the logs are in place, I'll probably talk to spevack and stickster about moving their stats stuff over 16:14 < ricky> Cool 16:14 < mmcgrath> ricky: one thing I'd like to do is look at moving the logging to tcp. 16:14 < mmcgrath> looks like we can setup spooling so that even when log1 goes down, it won't lose logs. 16:14 < mmcgrath> needs a closer look. 16:14 < ricky> Ah, cool 16:14 < mmcgrath> Anyone else have anything to discuss? 16:15 * stickster blinks 16:15 < abadger1999> python-fedora-0.3 has been deployed to fas and app servers. 16:15 < abadger1999> Looks good for deploying to the rest of infrastructure after we have the new fas package 16:15 < abadger1999> (fasClient needed tweaks) 16:15 < ricky> Ah, yes - hopefully, we'll have a FAS update with that soon. 16:16 < mmcgrath> yeah 16:16 < mmcgrath> stickster: do you know when the new privacy policy will be a go? 16:16 < stickster> mmcgrath: I don't -- since spot is at OLS and I'll be in BOS next week, I'd say we can give you an update by then 16:17 < abadger1999> Can we update the fas server before the privacy policy is live or do they have to be concurrent? 16:17 < ricky> abadger1999: At worst, moving over early is making things slightly better 16:17 < ricky> (As in, it's breaking our current policy less now) 16:17 < abadger1999> That was my thought too. 16:17 < stickster> Right, there's no problem with tightening up the actual services early. 16:18 < stickster> As long as when our new policy goes into action, it doesn't conflict with the actual service :-) 16:18 < mmcgrath> <nod> 16:18 < abadger1999> heh. Cool. 16:18 < stickster> But I am certain that you guys have once again outdone yourselves in that regard 16:18 -!- fchiulli_ [i=824c4010(a)gateway/web/ajax/mibbit.com/x-35e769ac5445486c] has joined #fedora-meeting 16:18 * stickster washes Infrastructure car a little 16:18 < mmcgrath> what link do we use to refer to the new policy, same link? 16:19 < stickster> *wax on* *wax off* 16:19 < stickster> mmcgrath: Yes, the link won't change. 16:19 -!- fchiulli [i=824c4010(a)gateway/web/ajax/mibbit.com/x-7817b8260bce18db] has quit "http://www.mibbit.com ajax IRC Client" 16:19 < stickster> We'll get the content slotted into the right place. 16:20 -!- fchiulli_ is now known as fchiulli 16:21 < mmcgrath> k 16:21 < mmcgrath> so thats that then. 16:21 < mmcgrath> anyone have anything else to discuss? 16:22 < mmcgrath> we'll close in 30 16:22 < mmcgrath> 15 16:22 < mmcgrath> 5 16:22 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Meeting end!

1 0

bash $TMOUT
by Mike McGrath 23 Jul '08

23 Jul '08

Sooooo, I'd like to set a $TMOUT for all of our bash sessions. I see a lot of shells just needlessly open. This is going to piss people off though, I haven't even done it yet and its pissing me off :) Are there any very vocal oppositions to this? Any alternatives? I'd like to at a minimum install it on fedorapeople. -Mike

7 14

Spam headers @fp.o addresses
by Mike McGrath 23 Jul '08

23 Jul '08

I've started adding spam headers to all @fp.o emails. Please keep your eyes and ears out for any strangeness. We'll likely have to further configure things. I've also been considering changing the standard spam headers to fedora specific headers so people understand where they are coming from and can use them, in conjunction with their local spam scanners. I'm really not sure what happens to our spam headers if another spamassassin server scans them, I'd assume ours would just vanish. -Mike

2 1

Database outage puppet updates
by Ricky Zhou 22 Jul '08

22 Jul '08

During tonight's outage, I've accumulated a good number of unpushed commits in my puppet clone, specifically on the db node manifests, the db service, db backup scripts, and just about all of the noc1 nagios hosts and hostgroups (and the pgsql service). I plan to push all these changes in one shot tomorrow if the db migration works out, so I'd appreciate it if everybody could avoid pushing puppet changes in these areas until this has happened in order to prevent ugly/complex merges. Thanks a lot, Ricky

1 1

[Fwd: Cron <root@publictest10> /bin/sleep $(($RANDOM/20)); /usr/bin/fasClient -i]
by Toshio Kuratomi 22 Jul '08

22 Jul '08

I've updated fasClient with what we have in fas's git repository on fas[12], app[123456], publictest10, publictest9, fedorapeople.org,and releng2. We shouldn't get more DeprecationWarnings from those hosts. We'll have a new FAS package in the next few days that will incorporate these changes. -Toshio -------- Original Message -------- Subject: Cron <root@publictest10> /bin/sleep $(($RANDOM/20)); /usr/bin/fasClient -i Date: Tue, 22 Jul 2008 06:29:37 -0700 From: root(a)publictest10.fedoraproject.org (Cron Daemon) To: root(a)publictest10.fedoraproject.org /usr/bin/fasClient:377: DeprecationWarning: send_request(input) is deprecated. Use send_request(req_params) instead request = self.send_request('group/list', auth=True, input=params) /usr/bin/fasClient:424: DeprecationWarning: send_request(input) is deprecated. Use send_request(req_params) instead data = self.send_request('user/list', auth=True, input=params)

1 0

doing FAS2 with sqlalchemy-migrate
by Yaakov Nemoy 20 Jul '08

20 Jul '08

Hey List, There are alot of tickets for FAS2 outstanding that require some changes to the DB. Toshio is working on getting sqlalchemy-migrate into Fedora now, so I felt it would be fitting to get FAS2 to be an example of how to do it right. (This means I am probably doing something wrong.) http://git.fedorahosted.org/git/fas.git?p=fas.git;a=commitdiff;h=21643256e4… This git commit sets up migration for us. The instructions how to use it are there. In short, the overlying design is to assume the DB has already been configured and deployed using the old method. This will manage changes we do on top. This will make it easier for people with working development trees to simply migrate their dev systems over without having to start their DB fresh. Please poke holes in this plan, so we can have something more solid that can be used as a gold standard for Fedora Infrastructure development. -Yaakov

2 4

Database Upgrades
by Mike McGrath 18 Jul '08

18 Jul '08

So next week will be an interesting week regarding our databases. It just so happened this way but db1 and db3 will be ready at about the same time. The new db1 will replace the old db1 exactly. We'd already migrated to it a couple of weeks ago but had to revert after some memory issues came up. I believe that has been resolved. db3 is a little different. db2 will remain as the primary database for all databases except for koji. Koji will get moved to db3 which will also have Postgresql 8.3 on it. I'd like to leave db2 at 8.1 for the immediate future, after we've determined db3 is a success we'll schedule a time to update db2 to 8.3 as well. Initial, very basic tests (db import) showed that with koji at least, postgresql 8.3 was significantly faster. 8.5 hours instead of 14.4 hours to import the koji database. -Mike

2 2

Outage Notification - 2008-07-22 04:59 UTC (MAJOR)
by Mike McGrath 17 Jul '08

17 Jul '08

There will be an outage starting at 2008-07-22 04:59 UTC, which will last approximately 10 hours. To convert UTC to your local time, take a look at http://fedoraproject.org/wiki/Infrastructure/UTCHowto or run: date -d '2008-07-22 04:59 UTC' Affected Services: Websites Buildsystem Database Unaffected Services: CVS / Source Control DNS Mail Torrent Fedora Hosted Talk Ticket Link: https://fedorahosted.org/fedora-infrastructure/ticket/705 Reason for Outage: Re-migrating db1 to its new hardware (memory issues have hopefully been resolved). Moving koji database from db2 to a dedicated new server, db3. The db1 outage will cause the wiki and smolt to be offline for no longer then an hour. The koji move will cause the buildsystem (and whatever depends on the build system) to be down for between 8 and 10 hours. We haven't padded the outage so if something goes wrong, we'll have to abort and re-schedule it for another time. Likely the next night so be prepared. Also please make sure any builds you complete any builds prior to the time of the outage. Contact Information: Please join #fedora-admin in irc.freenode.net or respond to this email to track the status of this outage.

1 0

Meeting Log - 2008-07-17
by Ricky Zhou 17 Jul '08

17 Jul '08

16:00 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Who's Here? 16:00 * ricky 16:00 * ianweller 16:00 * SmootherFrOgZ 16:00 < G> baaa 16:01 * abadger1999 16:01 < mmcgrath> Ok, lets get cracking 16:02 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Tickets 16:02 < mmcgrath> .tiny https://fedorahosted.org/fedora-infrastructure/query?status=new&status=assi… 16:02 < zodbot> mmcgrath: http://tinyurl.com/2hyyz6 16:02 * skvidal is 16:02 * dgilmore 16:02 < mmcgrath> .ticket 395 16:02 < zodbot> mmcgrath: #395 (Audio Streaming of Fedora Board Conference Calls) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/395 16:02 < mmcgrath> jcollie: around? 16:03 < mmcgrath> ok, we'll skip that for now 16:03 < mmcgrath> .ticket 446 16:03 < zodbot> mmcgrath: #446 (Possibility to add external links on spins page) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/446 16:03 < mmcgrath> dgilmore: ping ^^ 16:04 -!- balor [n=balor(a)gimili.plus.com] has joined #fedora-meeting 16:04 < mmcgrath> wow, going to be a short meeting I suspect. 16:04 < ricky> Heh. 16:04 < mmcgrath> .ticket 547 16:04 < SmootherFrOgZ> ^^ 16:04 < zodbot> mmcgrath: #547 (Koji DB Server as postgres 8.3) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/547 16:04 < mmcgrath> I've been working on testing the imports and things. 16:05 < SmootherFrOgZ> do you want help on this part ? 16:05 < mmcgrath> So far so good. When I first started testing I had a ext3 abort on me but haven't seen that happen since. And have done many more restores with it 16:05 < mmcgrath> SmootherFrOgZ: thanks but I think we're all set. I'm just waiting to hear back from f13 about the exact outage times. 16:05 < mmcgrath> It will take 8-10 hours to import 16:05 < SmootherFrOgZ> k 16:06 < mmcgrath> It took 5 fewer hours to import to 8.3 then to 8.1 of the same hardware configuration. 16:06 < mmcgrath> so we may see significant performance gain with koji. 16:06 < mmcgrath> .ticket 576 16:06 < zodbot> mmcgrath: #576 (Infrastructure Contact Information) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/576 16:06 < G> no update here 16:07 < mmcgrath> G: you given any further thought to this? I've been testing some sshfs and encfs combo's with success. its an option worth looking at 16:07 < mmcgrath> <nod> 16:07 < mmcgrath> next ticket 16:07 < G> yeah, it's been playing at the back of my mind :) 16:07 < mmcgrath> .ticket 448 16:07 < zodbot> mmcgrath: #448 (Hosting request for fedora-quotes) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/448 16:07 * mmcgrath swore this was all fixed 16:07 < mmcgrath> I'll comment and close it. 16:07 < ricky> Thanks 16:07 * ricky should have done that last week 16:08 < mmcgrath> Ok, thats that. 16:08 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Torrent1 16:08 < mmcgrath> skvidal: ^^ 16:08 < mmcgrath> whats the torrent1 plan? its been up and running for a while. 16:09 < skvidal> it has been 16:09 < skvidal> yay 16:09 < mmcgrath> yeah 16:09 < skvidal> we can give it a shot for alpha 16:09 < skvidal> we'll see how much whatever-the-hell-xen-does 16:09 < skvidal> makes it blow up 16:09 < skvidal> :) 16:09 < mmcgrath> Thats till in a few weeks right? 16:09 < mmcgrath> f13: ping 16:10 < mmcgrath> notting: ping 16:10 < G> yeah, begining of August iirc 16:10 < notting> mmcgrath: yes? 16:10 < mmcgrath> do either of you know how much room is needed for the alpha release? 16:10 < skvidal> mmcgrath: there's 244G free on torrent1 16:10 < skvidal> we should be fine :) 16:10 < notting> if you budget the amount that was needed for f9, you can't go wrong 16:10 -!- sdziallas [n=sebastia(a)p57A2F945.dip.t-dialin.net] has joined #fedora-meeting 16:10 * nirik notes that the Xfce spin should be done... but it's not up anywhere yet... I think f13 was going to see about trying to put it on torrent1. 16:10 < dgilmore> notting: +10% 16:10 < mmcgrath> skvidal: there's also about that much on the primary mirror. 16:11 < dgilmore> mmcgrath: can i put sparc there 16:11 < mmcgrath> no sparc for you! 16:11 < skvidal> dgilmore: hush 16:11 < skvidal> mmcgrath: nod 16:11 < mmcgrath> I'll make a ticket and check it out 16:11 < skvidal> mmcgrath: only sad thing 16:11 < skvidal> archive.fp.o has 7.7G free :( 16:11 < dgilmore> ouch 16:11 < skvidal> yah 16:12 < skvidal> not sure what we should do there 16:12 < G> yeah, that came up on nagios when I added it 16:12 < mmcgrath> skvidal: hopefully the HP box will be fixed by then. 16:13 < mmcgrath> I've purchased the warranty, its not gone through yet though. 16:13 < skvidal> nod 16:13 < mmcgrath> it has multiple T of storage available. 16:13 < skvidal> okay 16:13 < skvidal> yay for multiple T 16:13 < mmcgrath> skvidal: any other concerns with torrent1? 16:14 < skvidal> other than the ongoing disaster that is sb1? 16:14 < skvidal> no 16:14 < ricky> *cough*better than sb3*cough* 16:14 < mmcgrath> heh 16:14 < mmcgrath> ok, moving on 16:14 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Outage next week 16:15 < G> heh 16:15 < dgilmore> mmcgrath: power it all off :) 16:15 < ricky> Innage 16:15 < mmcgrath> I would have liked to do this tonight or tomorrow but I actually won't be around much this weekend so that seemed bad. 16:15 < mmcgrath> heh 16:15 < dgilmore> i assume this is to migrate to db3? 16:15 < mmcgrath> So here's whats going to happen. Hopefully monday night. 16:15 < ricky> And db1 at the same time 16:15 < mmcgrath> dgilmore: yeah, and the db1 migration. 16:15 < G> just as long as the voting app doesn't go down :) 16:15 < dgilmore> mmcgrath: lots of people will be absent next week 16:16 < ricky> It shouldn't 16:16 < dgilmore> fedora has a big presence at OLS 16:16 < G> (esp as we are in the middle of the FESCo elections) 16:16 < mmcgrath> .ticket 705 16:16 < zodbot> mmcgrath: #705 (DB upgrade (db1 db2)) - Fedora Infrastructure - Trac - https://fedorahosted.org/projects/fedora-infrastructure/ticket/705 16:16 < mmcgrath> G: it should be ok. 16:16 < mmcgrath> the longest outage will be koji, for around 10 hours. 16:16 < mmcgrath> That doesn't include much buffer for error. If something happens after hour 1, we're calling off the outage and doing it another time. 16:16 < dgilmore> mmcgrath: do we have postgresql-8.3 built? 16:16 < mmcgrath> The shorter outages are in smolt and mediawiki. 16:16 < G> dgilmore: yep 16:17 < mmcgrath> I'm hoping less then an hour. I'll be testing though. 16:17 < mmcgrath> dgilmore: yeah, its already on db3 and I've imported to it twice with success. 16:17 < mmcgrath> Just shy of 9 hours. 16:17 < dgilmore> then you need to analyze it 16:17 < dgilmore> but that can be done under load 16:17 < abadger1999> dgilmore: Yep. I've been tracking the F-9 builds and rebuilding for EL-5. 16:17 < dgilmore> abadger1999: :) 16:17 < mmcgrath> dgilmore: we need to analyze _after_ a fresh import? 16:17 < mmcgrath> do the pgsql admins drink a lot or just hate operations people and... logic? 16:18 < abadger1999> I think so. 16:18 < dgilmore> mmcgrath: its recommended 16:18 < abadger1999> Not a vacuum but analyse yes. 16:18 < ricky> Hehe 16:18 < mmcgrath> what a useless piece of crap. 16:18 < SmootherFrOgZ> haha 16:18 < mmcgrath> seriously, how could someone think thats ok? 16:18 < abadger1999> (analyze runs over your data as it sits on the disks and decides what query plans will be better) 16:18 < dgilmore> supposed to get the planers right 16:18 < mmcgrath> we don't have to deal with any of that bs in mysql, and never have. 16:18 < dgilmore> indeed 16:19 < mmcgrath> anyway, ok. I'll be analyzing the tables as well. 16:19 * mmcgrath needs to not think about that right now as it just makes him angry at the world. 16:19 < mmcgrath> skvidal: did you just take sb1 down? 16:19 < skvidal> no 16:19 < mmcgrath> .ping serverbeach1.fedoraproject.org 16:19 < zodbot> pong 16:19 < skvidal> but I've noticed it has issues everytime we talk about it 16:19 < mmcgrath> hrm 16:19 < abadger1999> It's shy. 16:19 < G> VPN most likely 16:19 < mmcgrath> yeah, that pong is a lie. I can't get to it 16:20 < skvidal> G: not from where I am 16:20 < skvidal> the pong is a lie 16:20 < mmcgrath> zodbot's pinging system is pretty weak. 16:20 < G> .ping 16:20 < zodbot> pong 16:20 < skvidal> hahah 16:20 < ricky> Blech. 16:20 < ricky> .ping aw243hy2a3i420r 16:20 < zodbot> pong 16:20 < mmcgrath> oh, its that kind of ping :) 16:20 < mmcgrath> heheheh 16:20 < dgilmore> .slap serverbeach1.fedoraproject.org 16:20 < mmcgrath> .list 16:20 < zodbot> mmcgrath: Admin, Channel, ChannelLogger, ChannelStats, Config, Fedora, Internet, Koji, Later, Misc, Owner, Plugin, RSS, Seen, ShrinkUrl, String, Time, URL, Unix, UrbanDict, User, and Web 16:20 < mmcgrath> .list Internet 16:20 < zodbot> mmcgrath: dns, hexip, and whois 16:20 < ricky> .list Web 16:20 < zodbot> ricky: doctype, fetch, headers, netcraft, size, title, urlquote, and urlunquote 16:20 < ianweller> .help pi9nt 16:20 < zodbot> ianweller: Error: There is no command "pi9nt". 16:20 < ianweller> .help ping 16:20 < zodbot> ianweller: (ping takes no arguments) -- Checks to see if the bot is alive. 16:20 < mmcgrath> .list misc 16:20 < zodbot> mmcgrath: apropos, help, last, list, more, ping, source, tell, and version 16:20 < skvidal> so 16:20 < skvidal> sb1 is down 16:20 < ianweller> hahahahhahhahhhaahahaha 16:20 < ricky> .misc ping 16:20 < zodbot> pong 16:20 < mmcgrath> ahwell. 16:20 < ianweller> we've been trusting on that forever?! 16:20 < G> oh SPAM! 16:20 < skvidal> .spam 16:21 < ricky> Woooohoo, "Rapid" Reboot time! 16:21 < mmcgrath> Ok, so thats really all I have right now so I'll open the floor. 16:21 < skvidal> ricky: "rapid" is right 16:21 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Open Floor 16:21 < mmcgrath> who has things they'd like to discuss? 16:21 * skvidal suggests we poke sb1 in the eye 16:21 * ianweller suggests the spleen 16:21 < mmcgrath> skvidal: how is that possible? How did sb1 _just_ go down as we were talking about it? 16:21 < mmcgrath> did anyone even ssh to it or anything? 16:21 < skvidal> yep 16:21 * ricky couldn't get it 16:21 < skvidal> I ssh'd in and run 'df -h' 16:21 < ricky> **in 16:21 < mmcgrath> skvidal: and that killed it? 16:21 < skvidal> and then I logged out 16:21 < G> heh 16:21 < skvidal> apparently it did 16:21 < ianweller> skvidal: well don't do that then. ;) 16:22 < mmcgrath> thats amazing 16:22 < skvidal> yah 16:22 < skvidal> so 16:22 * dgilmore notes he will be more distracted than usual for the next 3 weeks 16:22 < ricky> :-) 16:22 < G> I've got nothing in particular, the proxy4->proxy1 rename went well 16:22 -!- SMParrish_mobile [i=20800ca5(a)gateway/web/ajax/mibbit.com/x-4c55ac3e294ff48b] has left #fedora-meeting [] 16:22 < skvidal> do we have any hosts with enough ram and enough disk to take a torrent1 image? 16:22 < mmcgrath> G: yes it did, thank you for that. 16:22 * ricky will probably start obsessing over sb3 again.. soonish 16:22 < skvidal> ricky: are you scheduling the sb1 reboot now or should I? 16:22 < G> ummm lets see what else 16:22 < mmcgrath> skvidal: let me think about that. 16:23 < skvidal> mmcgrath: oh and they need to be not behind a firewall 16:23 < poelcat> is there an RFE ticket about email diff notifications from the wiki? 16:23 < ricky> skvidal: Sure, I'll get it 16:23 < mmcgrath> thats the real trick. 16:23 * poelcat couldn't find one 16:23 < skvidal> mmcgrath: option2 is what I suggested before 16:23 < mmcgrath> poelcat: I've not seen one. 16:23 < mmcgrath> skvidal: give option 2 a try. 16:23 * poelcat will create one 16:23 < skvidal> mmcgrath: we nuke sb1 from orbit and reinstall the fucker as SOLELY torrent 16:23 < skvidal> mmcgrath: okay, I'll see if I can do it tomorrow. 16:23 < G> poelcat: I'm sure I saw one 16:23 < mmcgrath> ianweller: do you know? 16:23 < skvidal> ricky: please do hit the reboot 16:24 < G> poelcat: the other things to note: 16:24 < ianweller> mmcgrath: no i do not know 16:24 < ricky> (Reboot initiated) 16:24 < poelcat> the more and more page watches I add the more painful clicking on the URL is getting :) 16:24 < ricky> skvidal: Ooh, let me know what luck you have with bare metal kickstarting 16:24 < ricky> (I've not had any luck with sb3 :-() 16:24 < poelcat> particuarly with all the feature page changes 16:24 < ianweller> G: weren't you looking into that? 16:24 < G> 1. Mediawiki only sends 1 notification per page per gap/whatever - i.e. you have to actually goto the page or view the diff to start getting new notifications 16:25 < skvidal> ricky: I shall, we did them before they can't be _that_ hard ;) 16:25 * dgilmore notes mediawiki sucks 16:25 < G> 2. As a result, the link to the diff is a dynamic one that'll always be from notifiededit-1 to current 16:25 < dgilmore> its only saving grace is a database backend 16:26 < poelcat> dgilmore: i was going to recommend moin, but they don't have a db ;-) 16:26 * nirik is happy with the rss feed of changes. 16:26 < G> poelcat: thats definately not how to get on my side :) 16:26 < ricky> skvidal: All I know is that the SOP (and a few other tries) didn't work - feel free to give it a kick or two :-) 16:27 < G> poelcat: so the problem is, Mediawiki's notification system is already fundamentally flawed in 1.11 :) 16:28 < mmcgrath> G: poelcat: mind taking that to the ticket? We can probably close the meeting quicker :) 16:28 < poelcat> mmcgrath: you got it 16:28 < mmcgrath> sweet 16:28 < ricky> I really wonder what they were thinking when they decided to put a link to the diff in the email 16:28 < mmcgrath> Ok, anyone have anything else they'd like to discuss? If not we can close the meeting in 30 16:28 < G> ricky: it saves on spam 16:29 < mmcgrath> 20 16:29 < ricky> Actually, I wonder how crazy wikipedia editors keep track of stuff like they do 16:29 < mmcgrath> 10 16:29 -!- mmcgrath changed the topic of #fedora-meeting to: Infrastructure -- Meeting End

1 0

[Fwd: Re: Announcing the Fedora EDU Spin Preview]
by Paul W. Frields 11 Jul '08

11 Jul '08

Anyone have insight into this? Paul -------- Forwarded Message -------- > From: Jeroen van Meeuwen <kanarip(a)kanarip.com> > Reply-To: Development discussions related to Fedora > <fedora-devel-list(a)redhat.com> > To: Development discussions related to Fedora > <fedora-devel-list(a)redhat.com> > Subject: Re: Announcing the Fedora EDU Spin Preview > Date: Fri, 11 Jul 2008 01:16:35 +0200 > > Luke Macken wrote: > > On Thu, Jul 10, 2008 at 11:49:44AM -0400, Christopher Aillon wrote: > >> Also, you need to get approval from the Spin SIG and Board before you > >> can call it a spin. > >> > >> https://fedoraproject.org/wiki/SIGs/Spins/SpinSubmissionProcess > > > > Sadly, this is more of a roadblock than a process at the moment, seeing > > as how it has been blocking on the creation of a fedora-spins mailing > > list for the past 3 months. Does someone with postfix/mailman knowledge > > want to help get this ball rolling ? > > > > I don't know what the problem is exactly... We have multiple mailing > lists running in all kinds of locations... @redhat.com, > @fedorahosted.org and @lists.fedoraproject.org... Not sure what's > holding us back. > > Kind regards, > > Jeroen van Meeuwen > -kanarip > -- Paul W. Frields gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717 http://paul.frields.org/ - - http://pfrields.fedorapeople.org/ irc.freenode.net: stickster @ #fedora-docs, #fedora-devel, #fredlug

4 4

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

infrastructure July 2008