happy new year all! my backups of network appliance nfs mounts has gotten intolerable. i have 3 FAS250's i'm working with: masspec, birch, aspen. all are connected to the same switch via gigE network, with single hops to the bacula server (2.2.6 patched). first full i did was this one:
Job: mspec.2007-12-18_21.29.07 Backup Level: Full Client: "molbio-fd" 2.2.6 (10Nov07) x86_64-unknown-linux-gnu,redhat, FileSet: "Mspec Set" 2007-12-18 20:45:51 Pool: "Full" (From Job resource) Storage: "LTO4" (From Job resource) Scheduled time: 18-Dec-2007 21:29:24 Start time: 18-Dec-2007 21:53:40 End time: 19-Dec-2007 15:55:36 Elapsed time: 18 hours 1 min 56 secs Priority: 10 FD Files Written: 863,458 SD Files Written: 863,458 FD Bytes Written: 1,825,660,355,131 (1.825 TB) SD Bytes Written: 1,825,879,267,061 (1.825 TB) Rate: 28123.4 KB/s Software Compression: None VSS: no Encryption: no Volume name(s): 002045L4|002042L4 Volume Session Id: 3 Volume Session Time: 1198028560 Last Volume Bytes: 960,677,286,912 (960.6 GB) Non-fatal FD errors: 0 SD Errors: 0 FD termination status: OK SD termination status: OK Termination: Backup OK adequate backup rate of 28MB/s. the next filer that got a full was aspen: Job: aspen.2007-12-20_23.25.27 Backup Level: Full (upgraded from Incremental) Client: "molbio-fd" 2.2.6 (10Nov07) x86_64-unknown-linux-gnu,redhat, FileSet: "Aspen Set" 2007-12-20 23:25:00 Pool: "Inc" (From Run pool override) Storage: "LTO4" (From Job resource) Scheduled time: 20-Dec-2007 23:25:00 Start time: 20-Dec-2007 23:25:02 End time: 21-Dec-2007 23:02:33 Elapsed time: 23 hours 37 mins 31 secs Priority: 10 FD Files Written: 8,069,999 SD Files Written: 8,069,999 FD Bytes Written: 990,743,048,680 (990.7 GB) SD Bytes Written: 992,311,396,359 (992.3 GB) Rate: 11648.8 KB/s Software Compression: None VSS: no Encryption: no Volume name(s): 002040L4|002049L4 Volume Session Id: 16 Volume Session Time: 1198028560 Last Volume Bytes: 15,757,378,560 (15.75 GB) Non-fatal FD errors: 0 SD Errors: 0 FD termination status: OK SD termination status: OK Termination: Backup OK slower at 12MB/s but still tolerable. the last started on christmas day, birch: Job: birch.2007-12-25_16.47.08 Backup Level: Full Client: "molbio-fd" 2.2.6 (10Nov07) x86_64-unknown-linux-gnu,redhat, FileSet: "Birch Set" 2007-12-22 09:56:56 Pool: "Full" (From Job resource) Storage: "LTO4" (From Job resource) Scheduled time: 25-Dec-2007 16:47:11 Start time: 25-Dec-2007 16:47:25 End time: 31-Dec-2007 23:09:01 Elapsed time: 6 days 6 hours 21 mins 36 secs Priority: 10 FD Files Written: 16,679,881 SD Files Written: 16,679,881 FD Bytes Written: 1,105,891,427,122 (1.105 TB) SD Bytes Written: 1,108,951,797,447 (1.108 TB) Rate: 2043.0 KB/s Software Compression: None VSS: no Encryption: no Volume name(s): 000299L4 Volume Session Id: 5 Volume Session Time: 1198587778 Last Volume Bytes: 1,504,677,113,856 (1.504 TB) Non-fatal FD errors: 0 SD Errors: 0 FD termination status: OK SD termination status: OK Termination: Backup OK not acceptable at 2MB/s. i cannot find any real difference in the network config or nfs mount config on these filesystms. i suspect it has to do with the nature of the filesystems. masspec has less than a millon files, aspen has around 8 million files and birch has nearly 17 million. has anyone had similar experience working with nfs backups of this nature? anything i can do to improve performance to get the filer backed up in a reasonable time window? -- michael ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
