Hi Rafal,
performance degradation is expected and normal while a system is in degraded mode or recovering from a failure. (Apart from that I think a latency of 0.4ms is not too bad...)
2 possible solutions to reduce impact:
* more spare disks * give reconstruct a lower priority: o options raid.reconstruct.perf_impact low
Hope that helps
Sebastian
On 05.02.2013 09:04, Rafał Radecki wrote:
Hi All.
I am new to this forum and quite new to NetApp. I work on FAS3210 with 8.0.2 7-Mode. We have 22 SAS disks in our appliance. The aggregate is:
aggr status
Aggr State Status Options aggrX online raid_dp, aggr root degraded redirect 32-bit because we have lost 2 disks 2 days ago (one spare and one data disk, NetApp has sent new disks to us). The problem is that the appliance is performing bad in this situation. NFS latency in file operations changed from for example 0,04ms to 0,4ms. I know that the aggregate is now using parity disks in place of one of the failed disks but is this performance drop common? Are there any known ways/howtos to cut down this problem?
Best regards, Rafal Radecki. _______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters