System: fedora 20
Name: s3ql
Arch: i686
Version: 2.9 (latest available through yum)
Release:: 1.fc20
"Local" storage backend, hosted through NFS share from server: 
mount: 192.168.1.195:/nfs/strobiserver on /mnt/local-nfs type nfs 
(rw,relatime,vers=3,rsize=32768,wsize=32768,namlen=255,hard,proto=tcp,timeo=600,retrans=2,sec=sys,mountaddr=192.168.1.195,mountvers=3,mountport=59909,mountproto=udp,local_lock=none,addr=192.168.1.195)

Here the steps I performed:

Created filesystem: mkfs.s3ql local:///mnt/local-nfs/myfsdata-ttt/  --> OK
Mounted filesystem: mount.s3ql local:///mnt/local-nfs/myfsdata-ttt 
/mnt/s3ql-ttt-local/ --allow-other --nfs --> OK (allow-other as 
s3ql-ttt-local gets exported by nfs/smb)
>> Using 4 upload threads.
>> Autodetected 4052 file descriptors available for cache entries

Filesystem shows up in both linux and through CIFS, copying files works.

Symptoms are coming up like an increasing load:

top - 14:39:08 up  5:00,  3 users,  load average: *5.05, 3.56, 1.76*
Tasks: 152 total,   2 running, 150 sleeping,   0 stopped,   0 zombie
%Cpu(s): 80.2 us, 14.5 sy,  0.0 ni,  0.0 id,  0.0 wa,  0.0 hi,  5.3 si,  
0.0 st
KiB Mem:   3101920 total,  3011764 used,    90156 free,    12516 buffers
KiB Swap:  5210108 total,     4416 used,  5205692 free,  2565888 cached

  PID USER      PR  NI    VIRT    RES    SHR S  %CPU %MEM     TIME+ COMMAND
 2977 root      20   0  589884 255564   4648 S **181.3**  8.2  13:32.46 
mount.s3ql

I got cpu0 and cpu1 of this kind:

cpu family      : 6
model           : 14
model name      : Intel(R) Core(TM) Duo CPU      T2600  @ 2.16GHz
stepping        : 12
microcode       : 0x54
cpu MHz         : 2167.000
cache size      : 2048 KB

While copy is still running, s3qlstat replies:
Directory entries:    2398
Inodes:               2400
Data blocks:          390
Total data size:      10680.65 MiB
After de-duplication: 3599.33 MiB (33.70% of total)
After compression:    3524.60 MiB (33.00% of total, 97.92% of de-duplicated)
Database size:        0.45 MiB (uncompressed)
(some values do not take into account not-yet-uploaded dirty blocks in 
cache)

At the time of writing, s3qlstat replies (which does not move anymore):

Directory entries:    303
Inodes:               305
Data blocks:          396
Total data size:      40149.03 MiB
After de-duplication: 3498.80 MiB (8.71% of total)
After compression:    3424.95 MiB (8.53% of total, 97.89% of de-duplicated)
Database size:        0.19 MiB (uncompressed)
(some values do not take into account not-yet-uploaded dirty blocks in 
cache)

du reports:
41112666        s3ql-ttt-local/ (as from s3ql server)
  3531860         myfsdata-ttt/ (as from NFS server)
so the numbers from s3qlstat seem to fit, but I doubt the de-duplication 
percentage.

Load is still increasing:
top - 19:10:05 up  3:20,  3 users,  load average: *22.00, 21.83, 20.47*
Tasks: 167 total,   1 running, 166 sleeping,   0 stopped,   0 zombie
libnuma: Warning: /sys not mounted or invalid. Assuming one node: No such 
file or directory
%Cpu(s): 16.4 us,  2.4 sy,  0.0 ni,  4.5 id, 76.0 wa,  0.0 hi,  0.7 si,  
0.0 st
KiB Mem:   3101920 total,  3018480 used,    83440 free,    17236 buffers

Logfiles in ~/.s3ql report only the s3ql.mount, no error/warning statement 
afterwards.

**In summary**: is this setup supported by S3QL: local storage backend by 
NFS share?

I'm retrying now with sshfs as local storage backend.

Please advice if I'm doing something stupid with s3ql :)
Thank you!

-- 
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.

Reply via email to