#ceph IRC Log

Index

IRC Log for 2010-08-20

Timestamps are in GMT/BST.

[0:29] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) has joined #ceph
[0:29] <MarkN> I thought I would give an update on my cluster, over the last 2 weeks I have not had any crashes on the latest stable, using ceph primarily as a data store that clients (less than 10) are mainly reading from. storing around 3Tb replicated twice with one mds.
[0:30] <gregaf> MarkN: yay! Can you remind me what the rest of your cluster looks like?
[0:31] <MarkN> it is very vanilla, 2Tb sata drives, journal on the same drive, 3 monitors 17Tb in total
[0:31] <MarkN> it is my test archive cluster where speed and performance do not matter.
[0:32] <MarkN> Just testing that i can have a large store and occasionally turn off a node and bring it up and not loose any data
[0:35] <gregaf> so like 5 OSDs with the monitors and MDS sharing with an OSD node?
[0:35] <johnl> hey MarkN, mind telling what kind of data profile? big files, small files? thousands? billions?
[0:37] <MarkN> gregaf: http://pastebin.org/615484 here is the pertinent parts of the ceph.conf
[0:38] <gregaf> nifty, thanks
[0:40] <MarkN> johnl: mainly large files of around 100Mb to 500Mb
[0:57] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) Quit (Quit: bye)
[1:45] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (Ping timeout: 480 seconds)
[1:45] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[2:02] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (Ping timeout: 480 seconds)
[2:02] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[2:10] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) Quit (Ping timeout: 480 seconds)
[2:11] * tjikkun (~tjikkun@195-240-122-237.ip.telfort.nl) has joined #ceph
[2:16] * Osso (osso@AMontsouris-755-1-31-136.w90-46.abo.wanadoo.fr) Quit (Quit: Osso)
[2:26] * s15y (~s15y@sac91-2-88-163-166-69.fbx.proxad.net) has joined #ceph
[5:35] * neale_ (~neale@pool-173-71-192-200.clppva.fios.verizon.net) Quit (Quit: neale_)
[6:41] * tjikkun (~tjikkun@195-240-122-237.ip.telfort.nl) Quit (Ping timeout: 480 seconds)
[6:49] * f4m8_ is now known as f4m8
[7:27] * Osso (osso@AMontsouris-755-1-31-136.w90-46.abo.wanadoo.fr) has joined #ceph
[7:29] * tjikkun (~tjikkun@2001:7b8:356:0:204:bff:fe80:8080) has joined #ceph
[8:01] <wido> sagewk: it was the pool "kvmimg"
[8:01] <wido> i removed it, hoped the OSD's would then start again
[9:45] * revstray (~rev@blue-labs.us) Quit (Ping timeout: 480 seconds)
[9:51] * deksai (~chris@71-13-57-82.dhcp.bycy.mi.charter.com) Quit (Read error: Operation timed out)
[9:53] * revstray (~rev@blue-labs.net) has joined #ceph
[10:00] * Yoric (~David@213.144.210.93) has joined #ceph
[10:15] * allsystemsarego (~allsystem@188.26.33.211) has joined #ceph
[12:40] * deksai (~chris@71-13-57-82.dhcp.bycy.mi.charter.com) has joined #ceph
[12:48] * deksai (~chris@71-13-57-82.dhcp.bycy.mi.charter.com) Quit (Ping timeout: 480 seconds)
[13:21] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) has joined #ceph
[14:04] * Osso (osso@AMontsouris-755-1-31-136.w90-46.abo.wanadoo.fr) Quit (Quit: Osso)
[14:13] * hijacker (~hijacker@213.91.163.5) has joined #ceph
[15:49] * f4m8 is now known as f4m8_
[16:40] * deksai (~chris@71-13-57-82.dhcp.bycy.mi.charter.com) has joined #ceph
[16:47] * conner (~conner@leo.tuc.noao.edu) Quit (Ping timeout: 480 seconds)
[16:52] * conner (~conner@leo.tuc.noao.edu) has joined #ceph
[17:18] * gregphone (~gregphone@cpe-76-90-74-194.socal.res.rr.com) has joined #ceph
[17:41] * gregphone_ (~gregphone@166.205.139.187) has joined #ceph
[17:42] * gregphone_ (~gregphone@166.205.139.187) Quit ()
[17:42] * gregphone (~gregphone@cpe-76-90-74-194.socal.res.rr.com) Quit (Read error: Connection reset by peer)
[17:42] * gregphone (~gregphone@cpe-76-90-74-194.socal.res.rr.com) has joined #ceph
[17:44] * gregphone_ (~gregphone@166.205.139.187) has joined #ceph
[17:44] * gregphone_ (~gregphone@166.205.139.187) has left #ceph
[17:50] * gregphone_ (~gregphone@166.205.139.187) has joined #ceph
[17:51] * gregphone (~gregphone@cpe-76-90-74-194.socal.res.rr.com) Quit (Ping timeout: 480 seconds)
[17:51] * gregphone_ is now known as gregphone
[18:00] * Yoric (~David@213.144.210.93) Quit (Quit: Yoric)
[18:03] * gregphone (~gregphone@166.205.139.187) has left #ceph
[19:22] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[19:25] <wido> sagewk: it was just a test with the osdmaps. I hoped that by injecting the new crushmap into it the "kvmimg" pool would be without a rule and would then be ignored
[19:25] <wido> i'll give your patch a try
[19:26] <sage> yeah i figured. unfortunately a few too many things depend on the map history for it to work
[20:03] <kblin> oh, btw, the "building from source" wiki link seems to miss the libgoogle-perftools-dev dependency. mind if I add that?
[20:03] <kblin> er, on the "debian" page, I mean
[20:04] <gregaf> it's a wiki, go for it!
[20:05] <kblin> man, if I had known I had to do maths to fix it... ;)
[20:46] <wido> sagewk: your fix seems to work, except for osd7, which crashed on "PG::read_log", so that seems to be another crash
[22:54] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) Quit (Ping timeout: 480 seconds)
[22:59] * allsystemsarego (~allsystem@188.26.33.211) Quit (Quit: Leaving)
[23:04] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit (Quit: Yoric)
[23:05] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) has joined #ceph
[23:13] * johnl (~johnl@cpc3-brad19-2-0-cust563.barn.cable.virginmedia.com) Quit (Ping timeout: 480 seconds)
[23:14] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) has joined #ceph
[23:15] * Yoric (~David@dau94-10-88-189-211-192.fbx.proxad.net) Quit ()
[23:32] * deksai (~chris@71-13-57-82.dhcp.bycy.mi.charter.com) Quit (Ping timeout: 480 seconds)
[23:39] * sagelap (~sage@166.134.249.3) has joined #ceph
[23:40] * sagelap (~sage@166.134.249.3) Quit ()

These logs were automatically created by CephLogBot on irc.oftc.net using the Java IRC LogBot.