problem with lanfree

[lmonya]

Hello,
we have a problem with our lanfree.


our environment:
- client_old (solaris 9) backup/restore lanfree without any problem
- client_new (solaris 10) StorageAgent is configured (identically client_old), session is ok.
- TSM_SERVER_1(5.5.2)
- TSM_NODE_A(5.5.2.0)
- TDPOracle (5.5.2.0), rman, 4 session parrallel
- StorageAgent (5.3.2.1)
- VTL(Eternus)
- MAXNUMMounts=12 and 24 drives/paths
- identically version of tape drivers on all servers
- zoning is correct
- all drives and paths are ok.


Situation 1:
a client_new is logged on TSM_SERVER_1 as TSM_NODE_A and tried to restore the database (with 4(!) session parrallel). it is working partialy. Two session are opened and data will be copied, two other session are closed immediately with:
01/25/2012 13:22:12 ANR8302E (Session: 96907, Origin: TSM_NODE_A_SA) I/O
error on drive <drive_name>
(/tsm/dev/<drive_name>) with volume <volume_name>
(OP=READ, Error Number=5, CC=0, KEY=00, ASC=00, ASCQ=00,
SENSE=70.00.00.00.00.00.00.58.00.00.00.00.00.00.00.00.00-
.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0-
0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.-
00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00-
.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0-
0.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.-
00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00-
.00.00.01.00.00.00.01.FE.C0.A2.00.02.D8.DC.D4.00.04.00.0-
0.00.00.00.00.00.00.00.20.00.04.00.00.00.04.00.00.00.04.-
00.00.00.04.00.00.03.49.62.7C.00.00.00.00.00.00.00.00.01-
.59.DC.00.01.59.DD.5C.00.00.00.00.FD.67.BD.88.01.38.EE.1-
8.03.49.3B.7C.03.AE.CC.F4.03.A7.F0.00.00.04.00.00.FD.67.-
BD.78.00.00.00.03.02.9C.DF.84.03.49.3B.7C.00.00.00.02.00-
.00.00.00.00.00.00.00.00.00.00.00.00.00.00,
Description=An undetermined error has occurred). Refer
to Appendix D in the 'Messages' manual for recommended
action. (SESSION: 96907)


Situation 2:
TSM_NODE_A makes backup from client_old (TDPOracle, rman with 2(!) session parrallel). it needs two tapes per session for full backup.
a client_new is logged on TSM_SERVER_1 as TSM_NODE_A and tried to restore the database(with 2(!) session parrallel). it is working but only with first two tapes.
as the next tape is mounted we get the same error:
TSM_NODE_A_SA) I/O
error on drive <drive_name>
(/tsm/dev/<drive_name>) with volume <volume_name>
(OP=READ, Error Number=5, CC=0, KEY=00, ASC=00, ASCQ=00,
SENSE=70.00.00.00.00.00.00.58.00.00.00.00.00.00.00.00.00-
.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.00.0-
.......................


Situation 3:
TSM_NODE_A makes backup from client_old (TDPOracle, rman with 2(!) session parrallel).
a client_new is logged on TSM_SERVER_1 as TSM_NODE_A (lanfree is not enabled!!!!).
restore and recovery are working without any problems(with 2(!) session parrallel).



all logs of OS are clean.
all logs of VTL are clean.

where and what is(can be) wrong?

Thanks

 

[oguri]

Hiii in previous i have faced a problem like this,

Each and every OS have a name for the drive (like in aix /dev/rmt) , Give the device name in the definition of the path from storage agent to the drive.

let us know the status

 

[chad_small]

Yeah usually the path is incorrectly defined. You need to make sure the device listed in the Storage Agent path matches the correct device on the TSM server.

Example:

TSM Server
------------
Name: DRIVE1
Library: TS3310-2
Serial: 30123001A
WWN: BLAHBLAHBLAH
Device: /dev/rmt12

Storage Agent
---------------
Name: DRIVE1
Library: TS3310-2
Serial: 30123001A
WWN: BLAHBLAHBLAH
Device: /dev/rmt3

 

[lmonya]

devices configured correctly.
we have a small overhead: we have a link for every device on tsm server and on tsm client with storageagent (cluster switch is easier when devices have the same names on both sides )... like this:

TSM Server(cluster):
root@>tsm_server> [/] # ls -ltra /tsm/dev/V020
lrwxrwxrwx 1 root root 14 Aug 31 2010 /tsm/dev/V020 -> /dev/rmt/164st

TSM Client(cluster):
root@<tsm_client>[/] # ls -ltra /tsm/dev/V020
lrwxrwxrwx 1 t root root 13 Jan 18 13:03 /tsm/dev/V020 -> /dev/rmt/63st


if I check a device, I get the same serial number on TSm Server and on TSM Client(StorageAgent) like this:
################################
Device /tsm/dev/V20 opened.
+----------------------------------------------------------------------------+
| IBM SCSI TAPE & MEDIUM CHANGER UTILITY PROGRAM |
+------------------------+--------------------------+------------------------+
| << GENERAL COMMANDS >> |
| 1: Open Device M: Menu Refresh D: Device Type |
| 2: Close Device Q: Quit Program |
| |
| << SERVICE COMMANDS >> |
| 3: Serial/Subsys Number 6: Store Dump 53: Enable/Disable Path |
| 4: Query Microcode Level 7: Download Microcode 54: Query Device Paths |
| 5: Force Dump 52: Get Driver Info 58: EKM Test |
| |
| << BASIC COMMANDS >> |
| 9: Test Unit Ready 12: Reserve 50: Log Sense Page |
| 10: Inquiry 13: Release 51: Mode Sense Page |
| 11: Request Sense |
+------------------------+--------------------------+------------------------+
| << TAPE DRIVE COMMANDS >> |
| 20: Read Data 29: Backward Space File 38: Get Position |
| 21: Write Data 30: Forward Space Rec 39: Set Position |
| 22: Write File Mark 31: Backward Space Rec 42: Sync Buffer |
| 23: Erase Tape 32: Locate End Of Data 43: Display Message |
| 24: Rewind 33: Get Record Size 45: Report Density |
| 25: Retension 34: Set Record Size 48: Query/Set Parameters |
| 26: Offline 35: Get Device Status 49: Read/Write Tests |
| 27: Load/Unload Tape 36: Get Device Info 55: Read Block Limits |
| 28: Forward Space File 37: Get Media Info 57: Get Encryption State |
| 59: Get Density |
+----------------------------------------------------------------------------+

ENTER COMMAND ('M' for Menu) ==> 3

Serial Number: 00C9D67D0020
IBM Library Device Subsystem Number: d00200
################################

 

[lmonya]

we have problem with this storageagent only wehn more as two tape will be mounted at the same time.
wehn two tape are mounted, it is working correctly till the next tape will be mounted...

very incredible situation...