UNSOLVED
Getting errors when restoring an NDMP backup
I am trying to backup&restore a Celerra via NDMP to a Legato 7.2.1 server running Soalris-9 sparc using disk as the backup media.
The backup is working fine, mminfo reports confirm the backup happens but I cannot perform a restore using nwrecover or nsrndmp_recover.
The error I get for nwrecover is
Reading Data .....
xdr_getbytes failed
recover failed for internal reason
NDMP Service Warning: Read failed on archive volume 1
Data server halted: The restore is aborted bu operator
Error durning File NDMP extraction
NDMP retrieval: child failed with a status of 1
llpatch# nsrndmp_recover -s llpatch -c cdm1 -S 4041078215 /testmigfs/testmigfs/dan_test_doc.txt2
nsrndmp_recover:ssid'4041078215': Performing recover from Non-NDMP type of device
nsrndmp_recover:ssid'4041078215': NDMP Service Debug: The process id for NDMP service is 0xd2ebe010
nsrndmp_recover:ssid'4041078215': Host = llpatch (155.34.24.227) port = 9671
nsrndmp_recover:ssid'4041078215': Performing non-DAR Recovery..
nsrdsa_recover:ssid'4041078215': READY:(host, ipaddr, port) = (llpatch, 155.34.24.227, 9671)
nsrdsa_recover:ssid'4041078215': error, bad last fragment
nsrndmp_recover:ssid'4041078215': NDMP Service Warning: Read failed on archive volume 1
nsrdsa_recover:ssid'4041078215': Reading data...DONE.
nsrndmp_recover:ssid'4041078215': NDMP Service Error: Sorry, unable to determine archive format.
nsrndmp_recover:ssid'4041078215': Data server halted: Error during the restore.
nsrndmp_recover:ssid'4041078215': NDMP data server has an internal error.
nsrndmp_recover:ssid'4041078215': Error during File NDMP Extraction.
nsrndmp_recover:ssid'4041078215': NDMP recover failed.
nsrndmp_recover:ssid'4041078215': Error during NDMP recover
llpatch# nsrndmp_recover -s llpatch -c cdm1 -S 4041078215 /testmigfs/dan_test_doc.txt2
nsrndmp_recover: Entry '/testmigfs/dan_test_doc.txt2' not found in NW index database
nsrndmp_recover: Error during NDMP recover
Does anyone have any suggestions?
thanks
Dan
The backup is working fine, mminfo reports confirm the backup happens but I cannot perform a restore using nwrecover or nsrndmp_recover.
The error I get for nwrecover is
Reading Data .....
xdr_getbytes failed
recover failed for internal reason
NDMP Service Warning: Read failed on archive volume 1
Data server halted: The restore is aborted bu operator
Error durning File NDMP extraction
NDMP retrieval: child failed with a status of 1
llpatch# nsrndmp_recover -s llpatch -c cdm1 -S 4041078215 /testmigfs/testmigfs/dan_test_doc.txt2
nsrndmp_recover:ssid'4041078215': Performing recover from Non-NDMP type of device
nsrndmp_recover:ssid'4041078215': NDMP Service Debug: The process id for NDMP service is 0xd2ebe010
nsrndmp_recover:ssid'4041078215': Host = llpatch (155.34.24.227) port = 9671
nsrndmp_recover:ssid'4041078215': Performing non-DAR Recovery..
nsrdsa_recover:ssid'4041078215': READY:(host, ipaddr, port) = (llpatch, 155.34.24.227, 9671)
nsrdsa_recover:ssid'4041078215': error, bad last fragment
nsrndmp_recover:ssid'4041078215': NDMP Service Warning: Read failed on archive volume 1
nsrdsa_recover:ssid'4041078215': Reading data...DONE.
nsrndmp_recover:ssid'4041078215': NDMP Service Error: Sorry, unable to determine archive format.
nsrndmp_recover:ssid'4041078215': Data server halted: Error during the restore.
nsrndmp_recover:ssid'4041078215': NDMP data server has an internal error.
nsrndmp_recover:ssid'4041078215': Error during File NDMP Extraction.
nsrndmp_recover:ssid'4041078215': NDMP recover failed.
nsrndmp_recover:ssid'4041078215': Error during NDMP recover
llpatch# nsrndmp_recover -s llpatch -c cdm1 -S 4041078215 /testmigfs/dan_test_doc.txt2
nsrndmp_recover: Entry '/testmigfs/dan_test_doc.txt2' not found in NW index database
nsrndmp_recover: Error during NDMP recover
Does anyone have any suggestions?
thanks
Dan
- llpatch# mminfo -S -q ssid=4041078215
ssid=4041078215 savetime=02/22/07 14:57:58 (1172174278) cdm1:/testmigfs
level=1 sflags=vNsF size=4977 files=3 insert=02/22/07
create=02/22/07 complete=02/22/07 browse=03/25/07 15:57:59 retent=02/22/08 14:57:59
clientid=5bcc909b-00000004-45db37fb-45db670a-000a0000-9b2218e3
*DIRECT: y;
*FILESYSTEM: /testmigfs;
*HIST: y;
*HOST: cdm1;
*LEVEL: 1;
*NDMP_BUTYPE: tar;
*OPTIONS: NT;
Clone #1: cloneid=1172174278 time=02/22/07 14:57:58 flags=F
frag@ 0 volid=4292721069 file/rec= 0/0 rn=0 last=02/22/07
Clone #2: cloneid=1172174279 time=02/22/07 14:57:59 flags=F
frag@ 0 volid= 14530987 file/rec= 0/0 rn=0 last=02/22/07 - Strange indeed. I'm not sure if this could be bug or something else yet. Is there any additional information from ndmp log on Celerra? It looks as ndmpd on Celerra can't read ssid and recognize it properly. At this point that could bug on both sides unless something else obvious that we are missing is going on.
nsrdsa_recover is clearly not being called here. Is the tape being loaded into a non-NDMP configured device? Usually in the output
we should see something like "Performing immedidate recover" if the recover is being launched from the Storage Node where the
tape/backup device is loaded. OR we should "Performing non-immediate recover" if it being launched from a machine that is NOT
the Storage Node. The celerra clearly thinks it should be able to read the data directly and NetWorker believes it is ndmpdsa.
What is the Recover Storage Node value in the Client?
Are you using the same Storage Node that the backup was done with? If no, set NSR_DSA_NODE= in the application
information section of the client resource. The value of NSR_DSA_NODE is the NetWorker Storage Node that will do the recover. Then
make sure you set the Recover Storage Node in the Client to be same as the value you specified for NSR_DSA_NODE. Last but not
least launch the recover from the machine that is specified as the value of NSR_DSA_NODE.
ble1
6 Operator
•
14354 Posts
•
56186 Points
721
0
Posted February 23rd, 2007 06:00