TSM database restore

[menonk]

Hi,
I am performing a tsm database restore.
./dsmserv restore db volumenames=<volume name> devclass=LTO commit=yes

However it throws an error and core dumps.

I have checked the dsmserv.opt file. The dsmconfig & volhist have been defined properly
devconfig file
/* Device Configuration */
DEFINE DEVCLASS LTO3 DEVTYPE=LTO FORMAT=ULTRIUM3 MOUNTLIMIT=1 MOUNTWAIT=60 MOUNT
RETENTION=60 LIBRARY=SL500A WORM=NO
SET SERVERNAME TAECADSM
DEFINE LIBRARY SL500A LIBTYPE=MANUAL
DEFINE DRIVE SL500A ST0 ONLINE=Yes
DEFINE PATH TAECADSM ST0 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=SL500A DEVICE=/de
v/rmt/2mt ONLINE=YES

volhist file contains the database volume name containing the database.
The tape drive has been configured properly and I have performed backup/restore using OS utility( tar)

Here is the activity that happens

./dsmserv restore db volumenames=07L072 devclass=LTO commit=yes
ANR0900I Processing options file dsmserv.opt.

Tivoli Storage Manager for Sun Solaris 8/9
Version 5, Release 3, Level 3.2

Licensed Materials - Property of IBM

(C) Copyright IBM Corporation 1990, 2006.
All rights reserved.
U.S. Government Users Restricted Rights - Use, duplication or disclosure
restricted by GSA ADP Schedule Contract with IBM Corporation.

ANR7800I DSMSERV generated at 10:32:12 on Jun 13 2006.
ANR7801I Subsystem process ID is 27446.
ANR4726I The ICC support module has been loaded.
ANR8200I TCP/IP driver ready for connection with clients on port 1500.
ANR0200I Recovery log assigned capacity is 7164 megabytes.
ANR0201I Database assigned capacity is 250000 megabytes.
ANR4621I Database backup device class LTO.
ANR4622I Volume 1: 07L072.
ANR4630I Starting point-in-time database restore (commit).
ANR0300I Recovery log format started; assigned capacity 7164 megabytes.
ANR8326I 001: Mount LTO volume 07L072 R/O in drive LTOMANUAL01 (/dev/rmt/1mt)
of library LTOMANUALMOUNT within 60 minutes.
ANR0301I Recovery log format in progress; 4 megabytes of 7164.
ANR0301I Recovery log format in progress; 8 megabytes of 7164.
ANR0301I Recovery log format in progress; 12 megabytes of 7164.
ANR0301I Recovery log format in progress; 16 megabytes of 7164.
ANR0301I Recovery log format in progress; 20 megabytes of 7164.
:
:
:
ANR8326I 001: Mount LTO volume 07L072 R/O in drive LTOMANUAL01 (/dev/rmt/1mt)
of library LTOMANUALMOUNT within 59 minutes.
ANR0301I Recovery log format in progress; 756 megabytes of 7164.
NR8302E I/O error on drive LTOMANUAL01 (/dev/rmt/1mt) with volume (OP=GET_ME-
DIUM_INFO, Error Number=22, CC=207, KEY=05, ASC=24, ASCQ=00,
SENSE=70.00.05.00-
.00.00.00.1E.00.00.00.00.24.00.00.C0.00.02.00.00.00.00.,
Description=Device is
not in a state capable of performing request). Refer to Appendix D in the
'Messages' manual for recommended action.
ANR9999D mmsman.c(1304): ThreadId<11> Couldn't determine the mediaType of media
07L072 in drive LTOMANUAL01 (/dev/rmt/1mt).
ANR9999D ThreadId<11> issued message 9999 from: <- 0x000000010160A924
outDiagf <- 0x00000001017F84F0 HandleManualMount <- 0x00000001017F4690
ManMountVolume <- 0x0000000100DB547C MmsMountVolume <- 0x0000000101872418
LtoOpen <- 0x0000000100D1B5B8 AgentThread <- 0x000000010008A4CC StartThread
<- 0xFFFFFFFF7EB1ECD8 *UNKNOWN* <- 0x000000010008A280 StartThread
ANR0301I Recovery log format in progress; 848 megabytes of 7164.
ANR0301I Recovery log format in progress; 852 megabytes of 7164.
ANR0301I Recovery log format in progress; 856 megabytes of 7164.
ANR0301I Recovery log format in progress; 860 megabytes of 7164.
ANR0301I Recovery log format in progress; 864 megabytes of 7164.
ANR0301I Recovery log format in progress; 868 megabytes of 7164.
:
:
:
:
ANR1401W Mount request denied for volume 07L072 - mount failed.
ANR7821S Thread 0 (tid 1) terminating on signal 11 (Segmentation violation).
ANR9999D Additional signal information:
ANR9999D Address 0x0000000000000000 is not mapped into the server's address space
ANR9999D
ANR9999D ===========================================
ANR9999D REGISTERS Information
ANR9999D ===========================================
ANR9999D PC:0x000000010008121C nPC:0x0000000100081220 CCR:0x0000000000000000
ANR9999D Y:0x0000000000000000 SP:0xFFFFFFFF7FFFC101 FP:0xFFFFFFFF7FFFC1D1
ANR9999D
ANR9999D G0:0x0000000000000000 G1:0xFFFFFFFF7EB1EF20 G2:0xFFFFFFFF7FFFC9C8
ANR9999D G3:0xFFFFFFFF7FFFC900 G4:0x0000000000000000 G5:0x0000000000000000
ANR9999D G6:0x0000000000000000 G7:0x0000000102090260
ANR9999D
ANR9999D O0:0x0000000000000000 O1:0x0000000000000000 O2:0x0000000000000000
ANR9999D O3:0x0000000000000000 O4:0x0000000000000000 O5:0x0000000000000000
ANR9999D O6:0xFFFFFFFF7FFFC101 O7:0x00000001000811E8
ANR9999D
ANR9999D L0:0x0000000103123688 L1:0x0000000000000000 L2:0x0000000001B44800
ANR9999D L3:0x0000000000000001 L4:0x0000000103124118 L5:0x0000000103124118
ANR9999D L6:0x0000000103124118 L7:0x0000000103124118
ANR9999D
ANR9999D I0:0x0000000000000000 I1:0x0000000103123688 I2:0x0000000000000008
ANR9999D I3:0x0000000000000000 I4:0x00000001030BE520 I5:0x0000000101B7BA28
ANR9999D I6:0xFFFFFFFF7FFFC1D1 I7:0x00000001007D6838
ANR9999D
ANR9999D Trace-back of called functions:
ANR9999D 0x00000001000B1E48 pkShowCallChain
ANR9999D 0x000000010008A1C0 AbortServer
ANR9999D 0x000000010008AB04 TrapHandler
ANR9999D 0xFFFFFFFF7EB1EDD8 *UNKNOWN*
ANR9999D 0xFFFFFFFF7EB1B668 *UNKNOWN*
ANR9999D 0x00000001000811E8 pkAcquireMutex
ANR9999D 0x00000001007D6838 IcCopyEnd
ANR9999D 0x00000001007F3CB8 IcRestoreDbData
ANR9999D 0x00000001007C854C icRestoreOneImageCopy
ANR9999D 0x00000001001114CC AdmRestoreDb
ANR9999D 0x00000001002BDEFC admRestoreDatabase
ANR9999D 0x0000000100067A38 RestoreDb
ANR9999D 0x0000000100061A2C main
ANR9999D 0x000000010005E7DC _start
ANR7820S Server thread 1 terminated in response to program abort.
ANR7820S Server thread 2 terminated in response to program abort.
ANR7820S Server thread 3 terminated in response to program abort.
ANR7820S Server thread 4 terminated in response to program abort.
ANR7820S Server thread 5 terminated in response to program abort.
ANR7820S Server thread 6 terminated in response to program abort.
ANR7820S Server thread 7 terminated in response to program abort.
ANR7820S Server thread 8 terminated in response to program abort.
ANR7820S Server thread 9 terminated in response to program abort.
ANR7820S Server thread 10 terminated in response to program abort.
Abort - core dumped...

If I look at the tape drive.The system ejects the tape out.
I have a standalone tape drive configured for this restore purpose..

 

[heada]

Is it the same type drive type as what the tape was written to? Is the firmware version the same? TSM failed when trying to read the type of media (OP=GET_MEDIUM_INFO) and so I'd look at either the drive or the tape itself.

-Aaron

 

[jdresang]

Looks like you also defined your dev class with a name of lto3, not lto

 

[menonk]

The tape drive and media is good.
tape drive works at the OS level. I had performed backup/restore at OS level before using the tsm drivers to associate it.

One more intresting thing is that the tape comes out of the drive during this database restore activity.

 

[zatogo]

I think jdresang is on to something...

Looking at your initial post...you define a device class called LTO3...but your restore command calls a device class called LTO...that is definitely going to cause an error of some sort.

 

[menonk]

Well now I modified my devconfig file..

root@hdqof1 # more devconfig
/* Device Configuration */
DEFINE DEVCLASS LTO DEVTYPE=LTO FORMAT=ULTRIUM3 MOUNTLIMIT=1 MOUNTWAIT=60 MOUNTR
ETENTION=60 LIBRARY=SL500A WORM=NO
SET SERVERNAME TAECADSM
DEFINE LIBRARY SL500A LIBTYPE=MANUAL
DEFINE DRIVE SL500A ST0 ONLINE=Yes
DEFINE PATH TAECADSM ST0 SRCTYPE=SERVER DESTTYPE=DRIVE LIBRARY=SL500A DEVICE=/de
v/rmt/2mt ONLINE=YES

The issue still persists

 

[jdresang]

Did you copy and paste your devconfig file? If so, make sure all the lines that are supposed to, wrap and aren't 2 lines that look like one. I have been caught by that before. Otherwise, do you have a utility to the tape drive itself, independent of TSM?

 

[Efim]

1. Try to change to FORMAT=DRIVE
2. Do you use only TSM driver ? This is the similar problem:
http://www-1.ibm.com/support/docvie...2E&uid=swg21170253&loc=en_US&cs=utf-8&lang=en
Efim

 

[ttrinh7975]

menonk-

What type of error are you getting now?
Check your device path....
You have in your devconfig file as /dev/rmt/2mt....


According to the error, it mounted the tape on /dev/rmt/1mt...

ANR8326I 001: Mount LTO volume 07L072 R/O in drive LTOMANUAL01 (/dev/rmt/1mt)

 

[menonk]

Hi,
I was advised to go for a reinstallation of the whole DR Setup.
Also I checked in google that 5.3.6 is better than 5.3.3 which I was using.
Accordingly I created 2 LUN's one for Database & other for Log.
The size allocated to these LUN's were little more than the volume capacity allocated in Production environment.

NOw.

1. OS is installed
2. OS patches were updated
3. TSM 5.3.6 was installed
4. I formated using
dsmfmt -log /tivolidb/vol1 283720
dsmfmt -log /tivolilog/log1 7164
All was good till now
5. Now I performed the following step
dsmserv format 1 /tivolilog/log1 1 /tivolidb/vol1

The log format happens without any issues. However Database format gives an issue
ANR0367I Database format completed.
ANR0207E Page address mismatch detected on database volume /tivolidb/vol1,
logical page 0 (physical page 256); actual: 67108864.
ANR9999D dballoc.c(1341): ThreadId <0> Error reading space map page from disk.
ANR9999D ThreadId <0> issued message 9999 from:
ANR9999D ThreadId <0> 0x0000000101669384 outDiagf
ANR9999D ThreadId <0> 0x00000001007430C8 DbAllocInit
ANR9999D ThreadId <0> 0x000000010074DC98 dbInit
ANR9999D ThreadId <0> 0x00000001002C3108 admInstallServer
ANR9999D ThreadId <0> 0x0000000100066D08 InstallServer
ANR9999D ThreadId <0> 0x0000000100062354 main
ANR9999D ThreadId <0> 0x000000010005F0BC _start
ANR0361E Database initialization failed: error initializing database page
allocator.

 

[Efim]

Why you format db as log ? Is this misprint ?

4. I formated using
dsmfmt -log /tivolidb/vol1 283720
dsmfmt -log /tivolilog/log1 7164

chdnge to

dsmfmt -db /tivolidb/vol1 283720
dsmfmt -log /tivolilog/log1 7164

Efim

 

[menonk]

Sorry its a misprint
1)cd /opt/tivoli/tsm/server/bin
a)dsmfmt -log /tivolilog/log1 7164
b)dsmfmt

 

[Efim]

Do you revise OS errorlogs ?
Try to delete vol1 and log1 and repeat your steps.
Efim

 

[menonk]

I deleted the log1 and vol1. Still the issue is persisting

 

[menonk]

I was able to resolve the issue.
There were 2 issues related to it.
a. I SAN box from which I used the LUN had throughput issue. I used a SAN with higher speed. It worked. Its advised not to use SATA disks for TSM Lun
b. Each partition size for DB LUN should be less than 100G.I had a LUN of 300G. This I divided into 4 slices that helped

 

[edata]

please, could you detail the following point

I was able to resolve the issue.
There were 2 issues related to it.
a. I SAN box from which I used the LUN had throughput issue. I used a SAN with higher speed.

thank you

 

[menonk]

Hi Edata,
The SAN issue was as follows
1. I have Netapp filer. The 1st filer I used had limited speed wrt to LUN performance.RPM of disks was 10K.
2. It had Sata disks. TSM does not recommend to have SATA drives for its Database. One can have TSM disk volumes on SATA disk.

3. The new disks were FC drives with RPM of 15K and that helped.