Christoff Buch wrote: > Hi! > > So far I thought about turning off spooling, which speeded throughput up > from ca. 5MB/s to 7MB/s. > But the old backup - software did ca. 15MB/s. > So I'm still at only about 50%. >
Wow, count yourself lucky! Right now I am getting only 462k/s! Often it does closer to 320k/s. This is on an AMD64 4400 dual core with 1G of RAM and a SATA disk running FC5. I am backing up to a spool file on the same disk which will then be written to DVD. I have run hdparm and bonnie benchmarks on my HD and it produces very good results. I know that there is a performance hit to be taken when spooling to the same disk but I can't imagine that it should be nearly this big. I am running SQLite3 for the db. I set PRAGMA synchronous = OFF; on the db just for kicks but it seems to have had no effect. top says 50% of the cpu is in IO wait and the rest is idle with only a few percent being used for any real work. Typical iostat looks like: Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtn hda 0.00 0.00 0.00 0 0 sda 969.39 146.94 13069.39 144 12808 How on earth could it be writing so much more than it is reading? That is quite puzzling. If I strace the bacula-fd or bacula-sd processes they are just sitting in a select. I never see them doing anything else. But the spool file is growing so I know it is making progress. -- Tracy R Reed http://ultraviolet.org _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
