lvol has ??? in lvdisplay -v

[jerrym]

I see on a server lvol5 has three drives associated with it and it is mirrored but there are

stale extents and unknown drive listed as ???  .  What does this mean?

--- Logical volumes ---
LV Name                     /dev/vg00/lvol5
VG Name                     /dev/vg00
LV Permission               read/write
LV Status                   available/stale
Mirror copies               1
Consistency Recovery        MWC
Schedule                    parallel
LV Size (Mbytes)            6144
Current LE                  384
Allocated PE                768
Stripes                     0
Stripe Size (Kbytes)        0
Bad block                   on
Allocation                  PVG-strict
IO Timeout (Seconds)        default

   --- Distribution of logical volume ---
   PV Name            LE on PV  PE on PV
   /dev/dsk/c0t10d0   256       256
   /dev/dsk/c1t10d0   256       256
   /dev/dsk/c1t8d0    128       128

.

.

.

   00252 /dev/dsk/c0t10d0   01693 current  /dev/dsk/c1t10d0   01693 current
   00253 /dev/dsk/c0t10d0   01694 current  /dev/dsk/c1t10d0   01694 current
   00254 /dev/dsk/c0t10d0   01695 current  /dev/dsk/c1t10d0   01695 current
   00255 /dev/dsk/c0t10d0   01696 current  /dev/dsk/c1t10d0   01696 current
   00256 ???                00901 stale    /dev/dsk/c1t8d0    00901 current
   00257 ???                00902 stale    /dev/dsk/c1t8d0    00902 current
   00258 ???                00903 stale    /dev/dsk/c1t8d0    00903 current
   00259 ???                00904 stale    /dev/dsk/c1t8d0    00904 current
.

.

[Januhe]

Check with an ioscan if you can see the disk; it is possible that you lost one disk if you see a NO_HW the disk has gone bad. If thisis the case you will need to break the mirror and replace the disk and then set the mirror again on the new disk.

[jerrym]

Checked ioscan. everything ok.

Note that there are three disk in lvdisplay for the lvol.

--- Distribution of logical volume ---
   PV Name            LE on PV  PE on PV
   /dev/dsk/c0t10d0   256       256
   /dev/dsk/c1t10d0   256       256
   /dev/dsk/c1t8d0    128       128

c0t10d0 is mirrored to c1t1od0 if I read this right.

--- Logical extents ---
   LE    PV1                PE1   Status 1 PV2                PE2   Status 2
   00000 /dev/dsk/c0t10d0   01441 current  /dev/dsk/c1t10d0   01441 current

.

.

  00256 ???                00901 stale    /dev/dsk/c1t8d0    00901 current

But what is ??? suppose to be coming from that is mirrored to c1t8d0. ??

[Steven E. Protter · Accepted solution]

Shalom,

  /dev/dsk/c1t8d0 is most likely dead.

dmesg

ioscan -fnCdisk

ioscan -fnNCdisk # 11.31 legacy i/o disabled.

ioscan -m dsf

Lets get a full picture.

From What I see you have lost a mirror but still have the whole file system.

Mirroring worked.

SEP

:robottongue:

Using obnoctoid smilesy until someone takes them away.

You know I'm just joking right?

[kgill]

Hi ,

Check for /var/adm/syslog/syslog.log   & see in case you see any lbolt errors in this log file .

If you find any lbolt error , then match it with details in /dev/rdsk/

[jerrym]

Changed mail.log and found this. Which is sone of the boot drives.  I also cannot run ioscan -m dsf. This systems is 11.11.

Subject: encino: Event Monitor Notification

>------------ Event Monitoring Service Event Notification ------------<

Notification Time: Fri Jul  1 03:51:09 2011

encino sent Event Monitor notification information:

/storage/events/disks/default/0_0_6_1_0_4_0.8.0
 is >= 3.
Its current value is SERIOUS(4).

Event data from monitor:

Event Time..........: Fri Jul  1 03:51:09 2011
Severity............: SERIOUS
Monitor.............: disk_em
Event #.............: 100172
System..............: encino

Summary:
     Disk at hardware path 0/0/6/1/0/4/0.8.0 : Device connectivity or hardware
     failure

ioscan shows:

disk      0  0/0/6/1/0/4/0.8.0               sdisk    CLAIMED     DEVICE       HP 36.4GST336754LC
                           /dev/dsk/c0t8d0   /dev/rdsk/c0t8d0

[singh sanjeev]

vgdisplay -v vg00

#check if disk is avialble or not.

# try to check the disk with dd command.

#check the error in dmesg and syslog.

#log a case with HP

[Steven E. Protter]

Shalom,

The EMS message is merely confirmation of what you already know. A disk died.

when disk is safely replaced EMS will stop squawking about this.

SEP

[Steven E. Protter]

Shlalom again,

yes ioscan -m dsf will not work on 11.11

Having no indicator of your OS, I had to use my procedure which I scraped off an 11.31 support document.

The basics are still workable as in the other ioscan commands.

SEP