Video Screencast Help

failed to restore ORACLE DB to alternate client

Created: 14 Mar 2013 | 11 comments
moxes john's picture

i encounter a problem to restore oracle DB to other client. alternate client is set as the same as source including user and file/folder permission. error came out as error 25 but flatfile backup and restore able to be executed and the alternate client. i've set the environment to be the same as well and run the restore but failed. i had call symantec support engineer for a solution and till now no solution had. anyone who have the idea on this please help.

restore status.

03/06/2013 16:19:16 - Info bpbrm (pid=8061172) brudr is the host to restore to

03/06/2013 16:19:16 - Info bpbrm (pid=8061172) telling media manager to start restore on client

03/06/2013 16:19:17 - Info bpbrm (pid=8061172) spawning a brm child process

03/06/2013 16:19:17 - Info bpbrm (pid=8061172) child pid: 7209100

03/06/2013 16:19:17 - Info bpbrm (pid=8061172) listening for client connection

03/06/2013 16:19:23 - Info bptm (pid=7864434) Waiting for mount of media id B00289 (copy 2) on server htvamdia1.

03/06/2013 16:19:23 - started process bptm (pid=7864434)

03/06/2013 16:19:23 - mounting B00289

03/06/2013 16:19:24 - Info bptm (pid=7864434) INF - Waiting for mount of media id B00289 on server htvamdia1 for reading.

03/06/2013 16:20:09 - mounted B00289; mount time: 0:00:46

03/06/2013 16:20:09 - Info bptm (pid=7864434) B00289

03/06/2013 16:20:11 - Info bptm (pid=7864434) INF - Waiting for positioning of media id B00289 on server htvamdia1 for reading.

03/06/2013 16:20:11 - positioning B00289 to file 2

03/06/2013 16:20:11 - positioned B00289; position time: 0:00:00

03/06/2013 16:20:11 - begin reading

03/06/2013 16:21:17 - Info bpbrm (pid=8061172) media manager for backup id bkkicba01-nbu_1362084707 exited with status 25: cannot connect on socket

03/06/2013 16:21:17 - Info bpbrm (pid=8061172) terminating bpbrm child 7209100 jobid=2631104

03/06/2013 16:21:17 - Error bpbrm (pid=7209100) listen for client protocol error - couldn't accept from data socket, A file or directory in the path name does not exist. (2)

03/06/2013 16:26:35 - begin Restore

03/06/2013 16:26:37 - number of images required: 1

03/06/2013 16:26:37 - media needed: B00289

03/06/2013 16:26:41 - restoring from image bkkicba01-nbu_1362084707

03/06/2013 16:26:47 - requesting resource B00289

03/06/2013 16:26:48 - Waiting for scan drive stop HTV2-HP.ULTRIUM4-SCSI.005, Media server: htvamdia1

03/06/2013 16:26:50 - granted resource  B00289

03/06/2013 16:26:50 - granted resource  HTV2-HP.ULTRIUM4-SCSI.005

 

 

Operating Systems:

Comments 11 CommentsJump to latest comment

Marianne's picture

Please tell us exact steps that were taken to prepare for redirected restore.

We can see that source client is bkkicba01-nbu, destinatio client is brudr, media server is htvamdia1.

Hopefully you have created logs on destination client before you attempted the restore?

We need to see media server's bpbrm log as well as client's bpcd and dbclient logs.

Please rename all logs to reflect process names (e.g. bpcd.txt) and post as File attachments.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

moxes john's picture

our DBA had prepare all the instance as per production server. we have set credential also file n folder permission to be exactly the same as our production server. before we run the the restore script, we have set the ORACLE_SID and NB_ORA_CLIENT. and when we are in RMAN, we set the DBID as per production server. i only have bpbrm from media and dbclient from client. this case had been logged as P1 to symantec netbackup senior engineer but its already pending for 2 weeks without any good result. im a bit confuse with oracle restore process as this is my 1st time perfoming it.

AttachmentSize
Media bpbrm.txt 178.6 KB
client dbclient.txt 56.83 KB
Marianne's picture

status 25: cannot connect on socket 

 

 <16> bpbrm listen_for_client: listen for client timeout during accept from data listen socket after 60 seconds

bpcd on the client logs connection attempts. 

We need to see this log.

PS: is there any reason why you cannot restore to a supported, NBU 7.x client?

 

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

moxes john's picture

im not sure why but i tested backup and restore normal file it can be done. the only thing failed is to backup and restore oracle database from that machine. i can only get the bpcd logs earliest on monday for you to see. is there any extra tweaks need to be done on the configuration coz i just follow the steps from the admin guide to do this..

Marianne's picture

The comms path for database backup and restore is different from OS backups.

We really need the bpcd log. Surely the Symantec engineer asked for it as well?

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

moxes john's picture

Hi Marianne,

sorrry for the delay..i did provide them with the logs but they could find a solution for this.

here's the bpcd logs

AttachmentSize
log.030613.txt 239.82 KB
log.030713.txt 19.87 KB
Marianne's picture

I see repeated occurences of the following:

 

17:48:54.127 [13324] <2> get_short: (2) premature end of file (byte 1)
17:48:54.127 [13324] <16> bpcd main: token read: -9
17:48:54.127 [13324] <4> bpcd main:    errno = 5 - Input/output error

These bpcd logs do not correspond with dbclient or bpbrm logs supplied previously, so I cannot say if this is related to the status 25 seen in bpbrm log above.

Hopefully you have provided Symantec with a consistent set of logs from the same restore attempt?

Repeat of my previous question:

Is there any reason why you cannot restore to a supported, NBU 7.x client?

 

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

moxes john's picture

i dont know why i cant restore to a supported NBU client for db, but for a normal flatfile it can.

i did provide them all of the logs. i even try other region support for suggestion but till now i still dont have any updates. sad

Marianne's picture

i dont know why i cant restore to a supported NBU client for db

NBU 6.5 is no longer supported. Your client is on version 6.5.

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links

moxes john's picture

if thats the case i will upgrade the version and try again and see the result.

thanks for the advice marianne

Marianne's picture

I cannot say for sure that upgrading the client will solve the problem but certainly worth a try. You may also receive a more favourable response from Support when servers and clients are on supported versions. Phone your local Support number, ask to speak to the duty manager and ask for your call to be escalated. 

It is impossible to pinpoint the problem with the unrelated sets of logs that were shared in this discussion.

I have found the following TN with exact same errors in bpcd log: http://www.symantec.com/docs/TECH182557

Seems error is caused by 'I/O issue' on client's file system.

May be worth trying the restore to a different client. 

*** PS: 
Seems you won't be able to upgrade NBU on the client. 
We see this info in dbclient log:

Release: 2.4.21-4.ELsmp

Version: #1 SMP Fri Oct 3 17:52:56 EDT 2003

Machine: i686

 

NBU 7.x is only supported on 64-bit Linux.
Please find a different (supported) client to restore to as suggested previously.

 

Supporting Storage Foundation and VCS on Unix and Windows as well as NetBackup on Unix and Windows
Handy NBU Links