During disk I/O, I now get hangs about every 30 seconds. Running sync manually doesn't seem to cause this. I also get minor pauses periodically. Just after the system is hung, top shows that 80% of the time is spent in system. Overall, it seems like a lot of time is being spent in g_up/g_down but I think that's normal. USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND root 10 41.4 0.0 0 8 ?? RL 3:37PM 2:04.01 [idle] root 3 23.2 0.0 0 8 ?? DL 3:37PM 3:55.80 [g_up] root 4 15.8 0.0 0 8 ?? DL 3:37PM 2:41.11 [g_down] nate 1022 11.1 0.4 8864 3060 p1 RL+ 4:05PM 1:03.11 cvs up root 37 1.2 0.0 0 8 ?? DL 3:37PM 0:12.87 [syncer] Example just after hang: 2.3% user, 0.0% nice, 82.6% system, 0.8% interrupt, 14.3% idle Example during cvs up but no hang: 1.6% user, 0.0% nice, 52.7% system, 0.8% interrupt, 45.0% idle I noticed the clock interrupt and/or psm (irq 12) waiting on *Giant near when the hang occurs so perhaps someone is holding Giant too long. I'm just running a nearly stock GENERIC with ata: FreeBSD 7.0-CURRENT #4: Sun May 7 14:12:55 PDT 2006 -- NateReceived on Mon May 29 2006 - 21:36:58 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:56 UTC