Camelia, the Perl 6 bug

IRC log for #gluster, 2012-10-14

| Channels | #gluster index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:57 JoeJulian stigchristian, elyograg: Also works by using lvm.
02:33 badone joined #gluster
02:45 badone joined #gluster
02:46 hchiramm_ joined #gluster
02:47 badone_home joined #gluster
02:59 badone_home joined #gluster
03:02 badone_home joined #gluster
03:31 vpshastry joined #gluster
03:31 vpshastry left #gluster
04:01 crashmag joined #gluster
04:14 mdarade joined #gluster
04:43 badone_home joined #gluster
04:59 nightwalk joined #gluster
06:02 mdarade left #gluster
06:05 badone_home joined #gluster
06:34 syoyo_ joined #gluster
07:12 mdarade joined #gluster
07:18 mdarade left #gluster
07:21 glusterbot New news from newglusterbugs: [Bug 866126] gluster volume heal crashes glustershd <https://bugzilla.redhat.com/show_bug.cgi?id=866126>
08:32 badone_home joined #gluster
09:01 ctria joined #gluster
09:06 badone_home joined #gluster
09:48 sunus joined #gluster
10:04 DataBeaver How come glusterfsd manages to cause an extremely high load average and slow down I/O on the entire server when executing something seemingly simple like a find -delete on a single client?
11:02 oneiroi joined #gluster
11:23 hagarth joined #gluster
12:05 Nr18 joined #gluster
12:17 hagarth joined #gluster
12:33 Azrael808 joined #gluster
13:00 syoyo__ joined #gluster
14:11 kshlm joined #gluster
15:09 TheHaven joined #gluster
15:26 hagarth joined #gluster
15:28 edward1 joined #gluster
15:33 stigchristian__ left #gluster
15:45 lh joined #gluster
15:47 hagarth joined #gluster
16:24 hagarth joined #gluster
16:37 cjbehm joined #gluster
16:39 cjbehm hi all - i'm relatively new to gluster and had a question about replacing a server. 3 servers, replication and it would use the same servername. the question is - would the self-heal impact performance or impact availability? could i pre-load the replacement server with data?
16:40 cjbehm i've read the docs on replacing the server, which is fairly straightforward, just don't want to bring access to a crawl/halt while replacing a server
17:10 mohankumar joined #gluster
19:23 andreask joined #gluster
19:26 hagarth joined #gluster
19:32 cjbehm anyone online that's familiar with replacing a server in a replicated setup? have a question about the healing process on performance/availability
19:36 badone joined #gluster
19:41 badone_ joined #gluster
20:11 duerF joined #gluster
20:15 Azrael808 joined #gluster
21:09 rcattelan joined #gluster
22:10 dabeowulf joined #gluster
22:40 lkoranda joined #gluster
23:47 JoeJulian DataBeaver: Under powered server maybe? Need more cpu?
23:49 JoeJulian cjbehm: Yes, there would be a performance impact from self-heal. Preloading the replacement server could mitigate that somewhat if you didn't preload it across the network. If you use the network for that, you might as well just do the self-heal since you'll be creating the same impact.

| Channels | #gluster index | Today | | Search | Google Search | Plain-Text | summary