20:02:02 #startmeeting Infrastructure 20:02:02 Meeting started Thu Jan 28 20:02:02 2010 UTC. The chair is mmcgrath. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:02:02 Useful Commands: #action #agreed #halp #info #idea #link #topic. 20:02:07 #topic Role Call 20:02:08 who's here? 20:02:12 * a-k is 20:02:40 Here 20:03:01 here 20:05:02 well, this is going to be a short meeting :) 20:05:06 #topic Infrastructure -- Tickets. 20:05:13 * Oxf13 here 20:05:33 Looks like no tickets 20:05:39 #topic PHX Network outage this morning 20:05:56 the outage seems to be over. I've not heard of the root cause yet. We were seeing a 25% packet loss. 20:06:13 #topic Search Engine 20:06:18 a-k: want to talk about the latest? 20:06:25 Sure 20:06:38 The additional memory allocated for publictest3 helped Xapian a lot 20:06:47 #link http://publictest3.fedoraproject.org 20:06:58 ... now has about 40k documents indexed to depth 5, instead of 15k documents to depth 4 20:07:07 I haven't tested the new limits for how deeply I can crawl before things break again 20:07:08 excellent. 20:07:18 I'm in the process of installing Nutch 20:07:30 I really do expect Nutch to be disk bound, because it makes a copy of everything it indexes 20:07:41 I'll see how fast it eats space probably later today 20:07:43 a raw copy or does it have some sort of compression it does? 20:07:52 Do we know how much disk the wiki uses? 20:08:00 I don't know about compression 20:08:13 well 20:08:17 I can find out though ... 20:08:19 I can give you an estimate, one sec. 20:10:02 a-k: k, I'm running a job to see how big an uncompressed version of the db dump is 20:10:09 OK 20:10:10 looks like it's about 920M 20:10:26 M, not G ? 20:10:30 correct 20:10:36 that excludes images and things 20:10:41 that's just the text from the wiki 20:10:46 OK 20:10:50 Oops 20:11:05 #link http://publictest3.fedoraproject.org/cgi-bin/omega 20:11:18 left off the important part in the last #link 20:11:29 I think that's about it for now 20:11:39 a-k: excellent thanks. 20:11:50 I think that's really it for the moment. 20:11:59 pretty fast 20:12:01 it's been a slow week. 20:12:05 #topic Open Floor 20:12:10 Anyone have anything they'd like to discuss? 20:12:41 mmcgrath: i'd like talk to you about how I can help out. 20:12:49 mmcgrath: maybe after the meeting 20:12:51 I'd like to test doing simultaneous composes on /mnt/koji 20:12:55 yawns1: sure, we can chat in #fedora-admin after the meeting. 20:13:00 to see if we can get no frozen rawhide activated for F13 20:13:08 mcgrath: great 20:13:12 what will that entail? 20:13:34 mmcgrath: we'll need to use a second releng box that has write access to /mnt/koji 20:13:48 20:13:50 we may already have that, lmacken was testing some bodhi stuff 20:13:57 releng1 should be up and ready for it. 20:14:17 Oxf13: is everything that should be on alt already there for the work you were doing today? 20:14:17 Oxf13: hopefully it will go soothly with the new storage 20:14:17 ah goot. 20:14:23 mmcgrath: yes 20:14:38 reminds me i need to follow up on getting a tracking number 20:14:45 dgilmore: yeah, but the timing is kinda bad. need to have NFR ready in like 2 weeks 20:14:54 Oxf13: right 20:15:02 Oxf13: has it already made it to serverbeach1.fedoraproject.org ? 20:15:06 if it works with current storage, we can expect it just gets faster 20:15:28 mmcgrath: does not appear to have 20:15:37 mmcgrath: it was on alt. within a few minutes of when I pinged you about it 20:15:53 Oxf13: k, its schedule dto run every hour at 15 past 20:15:56 which is in like 15 seconds :) 20:16:02 lets see if it starts pickign it up and how long it takes. 20:16:07 k 20:16:16 it's a fairly small amount of content this time 20:16:24 bingo, there it goes. lets see how it goes. 20:16:28 862M 20:16:50 alrighty 20:17:08 Oxf13: anyone have anything else they'd like to discuss? if not we can close the meeting early today. 20:17:17 "releng1" has a hostname of "koji01" 20:17:33 mmcgrath: i don't think I have anything else, we can keep talking about releng1 in #fedora-admin 20:17:45 Oxf13: heh, quality. yeah I'll get that all figured out today. 20:18:06 k 20:18:19 Alrighty, I'll close the meeting in 30 if no one has anything else. 20:25:29 Sorry guys, forgot to actually end the meeting 20:25:30 #endmeeting