SDS going back in time?!

From: Tony Howat (t.howat@linst.ac.uk)
Date: Wed Mar 03 2004 - 11:03:32 EST


Managers,

I have a 280R with two mirrored bootable SDS disks installed.

OS: SunOS nms2 5.8 Generic_108528-23 sun4u sparc SUNW,Sun-Fire-280R

I've had a consistant issue with SDS having a penchant for resyncing every
time we reboot the box. The box runs Solaris 8. I have similar Solaris 9
based SDS installs which never do this.

Now, until today I've kind of ignored the problem, not noticing any data
loss and figuring a resync was hardly the worst thing that could happen.
Now it's jumped up and bitten me on the arse, to use the technical term.

It seems to have resynced a bizarrely old mirror of the partition, even
more strangely one that's in a consistent state. The metastat output when
we rebooted was :

SDS metadevice problem report for Tue Mar 2 2004

Metadevices are not Okay:

d30: Mirror
     Submirror 0: d10
       State: Okay
     Submirror 1: d20
       State: Okay
     Pass: 1
     Read option: roundrobin (default)
     Write option: parallel (default)
     Size: 529152 blocks

d10: Submirror of d30
     State: Okay
     Size: 529152 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t0d0s0 0 No Okay

d20: Submirror of d30
     State: Okay
     Size: 529152 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t1d0s0 0 No Okay

d31: Mirror
     Submirror 0: d11
       State: Okay
     Submirror 1: d21
       State: Okay
     Pass: 1
     Read option: roundrobin (default)
     Write option: parallel (default)
     Size: 8395200 blocks

d11: Submirror of d31
     State: Okay
     Size: 8395200 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t0d0s1 0 No Okay

d21: Submirror of d31
     State: Okay
     Size: 8395200 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t1d0s1 0 No Okay

d32: Mirror
     Submirror 0: d12
       State: Okay
     Submirror 1: d22
       State: Okay
     Pass: 1
     Read option: roundrobin (default)
     Write option: parallel (default)
     Size: 6298944 blocks

d12: Submirror of d32
     State: Okay
     Size: 6298944 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t0d0s3 0 No Okay

d22: Submirror of d32
     State: Okay
     Size: 6298944 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t1d0s3 0 No Okay

d33: Mirror
     Submirror 0: d13
       State: Okay
     Submirror 1: d23
       State: Okay
     Pass: 1
     Read option: roundrobin (default)
     Write option: parallel (default)
     Size: 4202688 blocks

d13: Submirror of d33
     State: Okay
     Size: 4202688 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t0d0s4 0 No Okay

d23: Submirror of d33
     State: Okay
     Size: 4202688 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t1d0s4 0 No Okay

d34: Mirror
     Submirror 0: d14
       State: Okay
     Pass: 1
     Read option: roundrobin (default)
     Write option: parallel (default)
     Size: 2106432 blocks

d14: Submirror of d34
     State: Okay
     Size: 2106432 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t0d0s5 0 No Okay

d24: Concat/Stripe
     Size: 2106432 blocks
     Stripe 0:
        Device Start Block Dbase
        c1t1d0s5 0 No

d101: Soft Partition
     Component: d35
     State: Okay
     Size: 1048576 blocks
        Extent Start Block Block count
             0 1 1048576

d102: Soft Partition
     Component: d35
     State: Okay
     Size: 20971520 blocks
        Extent Start Block Block count
             0 1048578 20971520

d103: Soft Partition
     Component: d35
     State: Okay
     Size: 41943040 blocks
        Extent Start Block Block count
             0 22020099 41943040

d104: Soft Partition
     Component: d35
     State: Okay
     Size: 4194304 blocks
        Extent Start Block Block count
             0 63963140 4194304

d105: Soft Partition
     Component: d35
     State: Okay
     Size: 2097152 blocks
        Extent Start Block Block count
             0 68157445 2097152

d35: Mirror
     Submirror 0: d15
       State: Resyncing
     Submirror 1: d25
       State: Resyncing
     Resync in progress: 11 % done
     Pass: 1
     Read option: roundrobin (default)
     Write option: parallel (default)
     Size: 121704960 blocks

d15: Submirror of d35
     State: Resyncing
     Size: 121704960 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t0d0s7 0 No Okay

d25: Submirror of d35
     State: Resyncing
     Size: 121704960 blocks
     Stripe 0:
        Device Start Block Dbase State Hot Spare
        c1t1d0s7 0 No Okay

....the resync completes and my data has reverted back to how it was three
weeks ago or so.

system SUNWmdg Solstice DiskSuite Tool
system SUNWmdja Solstice DiskSuite Japanese localization
system SUNWmdnr Solstice DiskSuite Log Daemon Configuration Files
system SUNWmdnu Solstice DiskSuite Log Daemon
system SUNWmdr Solstice DiskSuite Drivers
system SUNWmdu Solstice DiskSuite Commands
system SUNWmdx Solstice DiskSuite Drivers(64-bit)

    PKGINST: SUNWmdr
       NAME: Solstice DiskSuite Drivers
   CATEGORY: system
       ARCH: sparc
    VERSION: 4.2.1,REV=1999.12.03.10.00

I'm mystified as to how this could occur?

There are no automated restore systems on the machine which could cause the
effects.

Has anyone heard of similar?

--
Tony Howat
UNIX Network Administrator
The London Institute
_______________________________________________
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 - 23:28:11 EDT