SUMMARY: Urgent: Mirror Disk Needs Maintenance

From: Slamet Ahmad Fadilah (SLAMET.FADILAH@SIEMENS.COM.PK)
Date: Fri May 14 2004 - 08:45:08 EDT


Please find the original problem at the bottom of this mail.
 
Problem:
One of sub mirror disks was under state "Needs Maintenance"
 
Summary:
1. metareplace or metareplace -e as suggested by the metastat was useless.
2. metadetach-metattach were also failed.
3. init s, umount the respective filesystem, fsck it, also brought nothing
 
Solution:
The disk was not replaced by a new one. But "resyncing by force" the whole
slices of the faulty disk.
 
 
Anyway I am still doubtful why the command 'metareplace -e' gave me a
confusing error message "No left space on device"
 
Many thanks for sharing infos, particularly to Andrew Cogswell, Ed Rolison ,
Kulkarni Golvind, Rossman and Trevor Mc Lachlan.
 
 
Have a nice weekend,
Slamet Fadilah
 
-----Original Message-----
 
Dear All,
 
One disk of my disk array (D1000) is under status "Needs Maintenance".
 
Here is its metastat of the faulty disk:
---------------
d81: Mirror
    Submirror 0: d16
      State: Needs maintenance Wed 12 May 2004 05:02:54 PM GMT
    Submirror 1: d33
      State: Okay Tue 11 May 2004 05:09:27 PM GMT
    Pass: 1
    Read option: roundrobin (default)
    Write option: parallel (default)
    Size: 35956494 blocks
 
d16: Submirror of d81
    State: Needs maintenance Wed 12 May 2004 05:02:54 PM GMT
    Invoke: metareplace d81 c1t2d0s0 <new device>
    Size: 35956494 blocks
    Stripe 0:
      Device Start Dbase State Hot Spare Time
      c1t2d0s0 0 No Maintenance Wed 12 May 2004
05:02:54 PM GMT
 
 
d33: Submirror of d81
    State: Okay Tue 11 May 2004 05:09:27 PM GMT
    Size: 35956494 blocks
    Stripe 0:
      Device Start Dbase State Hot Spare Time
      c3t8d0s0 0 No Okay Tue 11 May 2004
05:09:11 PM GMT
--------------
 
 
I tried to metareplace -e, but it gives error message:
 
#metastat -e d81 c1t2d0s0
Metareplace: isbrcomp1: c1t2d0s0: No left space on device.
 
FYI, from command format, I can confirm that the size of slice 0 of disk
c1t2d0 is zero (0).
 
Any idea why?
 
Is disk replacement is the only solution?
 
 
I got a plan as follow, please help me to verify it :
 
#init s
 
#ufsdump 0ucf /dev/rmt/0n /commonArea #to backup first the /commonArea
 
#umount /commonArea
 
#fsck /dev/rdsk/c1t2d0s0
 
#mount /dev/md/dsk/d81 /commonArea
 
#/usr/opt/SUNWmd/sbin/metareplace -e d81 /dev/dsk/c1t2d0s0
 
#/usr/opt/SUNWmd/sbin/metastat
 
#/usr/opt/SUNWmd/sbin/metareplace -e d81 /dev/dsk/c3t8d0s0
 
#/usr/opt/SUNWmd/sbin/metastat
 
#init 6
 
 
 
Thanks a lot for your time.
I'll summarize.
 
 
Best Regards,
Slamet Fadilah
TAC2 RC, Siemens Pakistan, Ufone Project
+92 333 5243669
 <mailto:slamet.fadilah@siemens.com.pk> slamet.fadilah@siemens.com.pk
_______________________________________________
sunmanagers mailing list
sunmanagers@sunmanagers.org
http://www.sunmanagers.org/mailman/listinfo/sunmanagers



This archive was generated by hypermail 2.1.7 : Wed Apr 09 2008 - 23:28:39 EDT