Summary - SDS question.

Thanks to:
         Jed Dobson,
         Jeff Woolsley,
         Julie Peers,
            Ana Estuardo,
            Ralf Albrecht.
With regards to resyncing the submirror, 4 out 5 respondents suggested 
using metareplace
with -e option. One respondent suggested I just use metattach to resync the 
device. Being
excessively timid I was not brave enough to try the second method (which I 
have no doubt
would have worked) so I went with the majority and the first method which 
worked
fine.
I recreated the metadb replicas as below.

Thanks again to all respondents.


Original post:
 >Hi Sunmanagers,
 >       We experienced power fluctuations which caused a drive to 
momentarily spin
 >down. The drive has 5 slices 4 of which are submirrors of 4 different SDS 
mirrors.
 >The 5th slice contains metadb replicas.
 >I am fairly confident that there is nothing wrong with the disk.
 >Two of these submirrors are shown as needing maintenance and the other 2 
are OK.
 >Also one of the metadb replicas on this device are showing up as having 
device write
 >errors.
 >
 >        flags           first blk       block count
 >     a m  p  luo        16              1034            /dev/dsk/c0t0d0s7
 >     a    p  luo        1050            1034            /dev/dsk/c0t0d0s7
 >     a    p  luo        2084            1034            /dev/dsk/c0t0d0s7
 >     a    p  luo        16              1034            /dev/dsk/c2t0d0s7
 >     a    p  luo        1050            1034            /dev/dsk/c2t0d0s7
 >     a    p  luo        2084            1034            /dev/dsk/c2t0d0s7
 >      W   p  l          16              1034            /dev/dsk/c7t0d0s7
 >     a    p  luo        1050            1034            /dev/dsk/c7t0d0s7
 >     a    p  luo        2084            1034            /dev/dsk/c7t0d0s7
 >
 >Question.
 >       Should I use metareplace with the -e option to transition the 
submirrors needing
 >attention to the available state. The man page seems to be saying that 
you would use this option when
 >you have had a power disturbance. Has anyone had a similar experience and 
used this option?
 > I guess the alternative would be to do
 >metadetach->metaclear->metattach
 >
 >Also what steps should I take to fix the metadb replica?
 >metadb -d /dev/dsk/c7t0d0s7
 >followed by
 >metadb -a -c 3 /dev/dsk/c7t0d0s7 ?

Tim Hespe
System Admin.
University of New South Wales Library
t.hespe@unsw.edu.au 
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers