Update:
with "local" storage backend on NFS server, speed of injection of files
into the s3ql is higher (factor 3) than with local and sshfs backend (was
1-1.5MB/s) - speed is not my concern, any component crashing and/or
rendering this backup space corrupt is!:
104-xxx-unit.mp3
100% 2974KB 5.9MB/s 00:00
112-xxx-unit.mp3
100% 8843KB 8.6MB/s 00:01
116-xxx-unit.mp3
100% 6614KB 6.5MB/s 00:01
Load is bit higher than with local and sshfs too:
top - 13:01:42 up 17:39, 4 users, load average: 4.99, 4.06, 2.10
But I'm suspecting something else going on, on the system:
Filesystem 1K-blocks Used Available Use%
Mounted on
*/dev/mapper/vg_strobiserver-lv_root 51475068 36425216 12412028 75%
/*
devtmpfs 1544224 0 1544224 0%
/dev
tmpfs 1550960 0 1550960 0%
/dev/shm
tmpfs 1550960 612 1550348 1%
/run
tmpfs 1550960 0 1550960 0%
/sys/fs/cgroup
tmpfs 1550960 10304 1540656 1%
/tmp
/dev/sda1 487652 182125 275831 40%
/boot
/dev/mapper/vg_strobiserver-lv_home 38766032 5762548 31011192 16%
/home
*192.168.1.195:/nfs/strobiserver 3841069376 1188181376 2574841920 32%
/mnt/local-nfslocal:///mnt/local-nfs/myfsdata-ttt 1073740784 854614
1072886171 1% /mnt/s3ql-ttt-local*
The lines in bold are changing: the last two are easy to understand
(storage backend & s3ql filesystem), but the first one has my /root/.s3ql
cache directory. If it's going to continu like that, that would explain why
there was nothing in the log file (disk full?). But that should not have
happened because, with sshfs backend I got 70GB of real data uploaded and
now with NFS also 60GB+.
Network speed did go down from the aforementioned 5MB/s towards 1MB/s, load
has not exceeded 5.
Going to an acceptance phase with real data now, will report back if the
symptoms happen again.
*Bottom-line:*Unclear what happened with my test setup earlier (it did
"lock" more than three times in a row), but having restarted from scratch
(and a clean reboot of all components) did proof that both NFS & sshfs
storage backends with s3ql do work, whereas the s3ql is exposed towards the
network by CIFS. Network bandwidth variance is high.
--
You received this message because you are subscribed to the Google Groups
"s3ql" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
