marcel partap
2010-Aug-04 16:37 UTC
flush-btrfs-1 doing 1MiB/s (avg) writing on idle system??
Dear BTRFS devs, still trying to find the cause for https://bugzilla.kernel.org/show_bug.cgi?id=12309 i stumbled over the strange fact that the flush-btrfs-1 thread is writing about 1MiBps constantly onto my root fs, even after turning off io-prone BOINC and NTOP.. there are other threads doing minor IO but that seems unrelated. My kernel is> Linux handtuch 2.6.35 #1 SMP PREEMPT Mon Aug 2 13:55:56 CEST 2010 x86_64 AMD Phenom(tm) II X4 920 Processor AuthenticAMD GNU/Linuxwith patch #488 from mentioned bug report applied. The partition is mounted>/dev/root on / type btrfs (rw,noatime,noacl)and btrfs filesystem show /dev/sda2 gives> Label: none uuid: 50719157-1813-4552-bcd5-65f85649e503 > Total devices 1 FS bytes used 230.12GB > devid 1 size 465.58GB used 335.29GB path /dev/sda2 > Btrfs v0.19-16-g075587c-dirtyI have uploaded an archived blktrace for about 10 minutes at http://hfopi.org/files/temp/btrfs-blktrace-idle.tar.bz2 (18M) and the resulting diagram is http://hfopi.org/files/temp/btrfs-blktrace-idle-trace.png As i couldn''t find an easy way to see what actually is located at these blocknumbers i am now kindly asking for assistance here - can anyone make any sense out of this? It surely makes me feel a lil'' uncomfortable about my system''s kosherness.. best regards, marcel ;) -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html