Date: Mon, 29 Mar 2004 17:02:53 -0500 From: Don Bowman <don@sandvine.com> To: "'current@freebsd.org'" <current@freebsd.org> Subject: question about dump performance on a snapshot Message-ID: <FE045D4D9F7AED4CBFF1B3B813C85337045D87A8@mail.sandvine.com>
next in thread | raw e-mail | index | archive | help
How long should it take to run 'dump' on a snapshot? I have a 6-disk raid 5 with 15K RPM scsi disk, so quite fast. It is 300GB in size, with 20% occupied. There are ~6000 files on this filesystem, all postgresql. I stopped postgresql, did a file system snapshot, and restarted postgresql. I then commenced a dump on the snap as below: dump -0 -a -C 32 -f $dumpfile -u $snapfile dumpfile and snapfile are both on the same filesystem. for the last 35 minutes, the size of the dump file has not gone up, and dump is taking 0% cpu: # dump -0 -a -C 32 -f $backupfile -u $outfile DUMP: Date of this level 0 dump: Mon Mar 29 16:32:25 2004 DUMP: Date of last level 0 dump: the epoch DUMP: Dumping /data/snapshot/snap.1 to /data/backup/dump.1 DUMP: mapping (Pass I) [regular files] DUMP: Cache 32 MB, blocksize = 65536 DUMP: mapping (Pass II) [directories] DUMP: estimated 59867851 tape blocks. DUMP: dumping (Pass III) [directories] DUMP: dumping (Pass IV) [regular files] load: 0.00 cmd: dump 39347 [runnable] 1.20u 5.70s 0% 48616k # ps -wwax |grep dump 39584 p1 L+ 0:00.00 grep dump 39336 d0 I+ 0:01.61 dump -0 -a -C 32 -f /data/backup/dump.1 -u /data/snapshot/snap.1 (dump) 39347 d0 I+ 0:06.90 dump: /data/snapshot/snap.1: pass 4: 0.94% done, finished in 1:34 (dump) 39348 d0 I+ 0:08.18 dump -0 -a -C 32 -f /data/backup/dump.1 -u /data/snapshot/snap.1 (dump) 39349 d0 I+ 0:08.18 dump -0 -a -C 32 -f /data/backup/dump.1 -u /data/snapshot/snap.1 (dump) 39350 d0 I+ 0:08.26 dump -0 -a -C 32 -f /data/backup/dump.1 -u /data/snapshot/snap.1 (dump) Although the dump always indicates it is 'runnable' when i hit ^T, it doesn't run, and doesn't consume any cpu. The machine is idle. machine is running current cvsup from march 21 on a dual xeon 2.8 with asr raid controller. # tunefs -p /dev/da0s1e tunefs: ACLs: (-a) disabled tunefs: MAC multilabel: (-l) disabled tunefs: soft updates: (-n) enabled tunefs: maximum blocks per file in a cylinder group: (-e) 2048 tunefs: average file size: (-f) 16384 tunefs: average number of files in a directory: (-s) 64 tunefs: minimum percentage of free space: (-m) 8% tunefs: optimization preference: (-o) time tunefs: volume label: (-L)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?FE045D4D9F7AED4CBFF1B3B813C85337045D87A8>