Oracle Backups - Status code 6 - The backups failed to back up the requested files

[Ashleigh_Steven]

Hi,

Doing a oracle backup through RMAN, but getting Status code 6 - The backups failed to back up the requested files.  Can someone please help to explain this?

18:35:18.826 [5460.3880] <2> int_CloseImage: INF - Backup - closing <bk_a4l0vvdt_1_1_705691069>
18:35:48.998 [5460.3880] <2> int_logDateRange: INF - Start Time = 12/15/09 09:17:49
18:35:48.998 [5460.3880] <2> int_logDateRange: INF - End Time = 12/17/09 09:17:49
18:35:49.982 [5460.3880] <2> vnet_vnetd_service_socket: ..\libvlibs\vnet_vnetd.c.2034: VN_REQUEST_SERVICE_SOCKET: 6 0x00000006
18:35:49.982 [5460.3880] <2> vnet_vnetd_service_socket: ..\libvlibs\vnet_vnetd.c.2048: service: bprd
18:35:49.982 [5460.3880] <2> logconnections: BPRD CONNECT FROM 10.64.115.11.3742 TO 10.64.114.2.13724
18:36:02.638 [5460.3880] <2> xbsa_ProgressLogSetup: INF - No progress log: Job will proceed without progress logging
18:36:04.794 [5460.3880] <2> vnet_vnetd_service_socket: ..\libvlibs\vnet_vnetd.c.2034: VN_REQUEST_SERVICE_SOCKET: 6 0x00000006
18:36:04.794 [5460.3880] <2> vnet_vnetd_service_socket: ..\libvlibs\vnet_vnetd.c.2048: service: bprd
18:36:04.794 [5460.3880] <2> logconnections: BPRD CONNECT FROM 10.64.115.11.3750 TO 10.64.114.2.13724
18:37:09.857 [5460.3880] <16> handshake: ERR - didn't receive the right filename from server. sent </bk_a5l1040f_1_1_705695759>, received <SET SNAP_ID=itrdbdop01-bkup_1260861171>.
18:37:09.857 [5460.3880] <16> CreateNewImage: ERR - handshake() failed
18:37:09.982 [5460.3880] <16> VxBSACreateObject: ERR - Could not create new image with file /bk_a5l1040f_1_1_705695759.
18:37:09.982 [5460.3880] <16> xbsa_CreateObject: ERR - VxBSACreateObject: Failed with error:
   Server Status:  Communication with the server has not been iniatated or the server status has not been retrieved from the server.
18:37:15.888 [5460.3880] <4> sbtend: INF - --- END of SESSION ---
18:37:15.888 [5460.3880] <8> close_image: Session being terminated abnormally, cleaning up
18:37:15.888 [5460.3880] <4> close_image: INF - backup FAILED
18:37:15.888 [5460.3880] <4> close_image: INF ---- end of Backup ---

18:37:15.888 [5460.3880] <16> VxBSAEndTxn: ERR - Transaction ended with active Backup/Restore.
18:37:15.888 [5460.3880] <16> xbsa_EndTransaction: ERR - VxBSAEndTxn: Failed with error:
   The transaction was aborted.
 

[Marianne]

There seems to be a comms problem between client and master server:
<16> handshake: ERR - didn't receive the right filename from server.
<16> CreateNewImage: ERR - handshake() failed
<16> xbsa_CreateObject: ERR - VxBSACreateObject: Failed with error:
   Server Status:  Communication with the server has not been iniatated or the server status has not been retrieved from the server.

The client needs to connect to bprd (via vnetd) on the master. Check in bprd log on the master for the incoming request from client's IP address. See in bprd log if master is able to resolve IP address to client name that corresponds with client name in policy.

Do the following test on the master server (ensure you have bpcd log on the client):
bptestbpcd -client <client_name>

The easiest solution normally is to add /etc/hosts entries on both sides.
If client is behind firewall, ensure vnetd is open in both directions.

[Symontec]

Would you try a bpclntcmd -pn from the client to check cominucation?

Regards,
SAF

[NBUTSE]

  Ashleigh,
It appears that your dbclient logs are not at full verbosity, you should increase the verbose levels.  Get a remote connection to the Windows Oracle server, from the start menu go to All Programs -> Veritas NetBackup -> Backup Archive and Restore.  In the Backup Archive and Restore tool, and go to the file menu, and down to NetBackup Client Properties, then to the trouble shooting tab on the right hand side.  Take note of the current settings, and then set the general logging level at 2, the verbose at 5, and the database at 9.  Click okay.

So what sticks out ot me in the logs that you've got is:

18:36:02.638 [5460.3880] <2> xbsa_ProgressLogSetup: INF - No progress log: Job will proceed without progress logging

That's rather odd, it suggests that the orasbt.dll (Oracle Agent for NetBackup) was unable to create a progress file under the netbackup\logs\user_ops\dbext\logs directory.  Ensure that this directory structure exists, and that the Oracle user has full read/write access to that directory.

With the verobse levels set, and the directory structure and permissions verified, please re-attempt the operation and take another look at the dbclient logs.

[NBUTSE]

What didn't you like to reflect the down vote?  Looking for honest feed back here.  Not enough explaination?


My rationale for my reply, the progress file is sometimes called a COMM file.  A COMM file is created upon initiation of an RMAN operation using SBT_TAPE channels when the NetBackup for Oracle agent has been installed.  This COMM file is passed to BPRD after the license key has been verified in the backup or restore request.  So if that file can't be created prior to the backup or restore request it would seemingly fail.  The COMM file receives updates from BPRD and NBJM out on the master server, and then also BPBRM from the media server.  The purpose of this COMM file is to provide the agent with status updates from the NetBackup master and media servers.

In Ashleigh's case, it appears that the COMM file was not created.  I know the logs are not at max verbosity because I work NetBackup for Oracle cases every day here at Symantec.  If they were at full verbose levels we should see all the parameters of the BPRD request enumerated, which we do not see here.

[Baski]

Thanks for good explanation.

[Marianne]

IMHO there was MORE THAN ENOUGH info in level 0 of dbclient log to pinpoint the problem.
Comms tests to and from the client (bptestbpcd and bpclntcmd) will reveal where the problem is.

[Ashleigh_Steven]

Will do the comms test and reveal the answer for you.

Thanks for your feedback so far.

[Ashleigh_Steven]

From the client;

C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -self
gethostname() returned: itrdbdop01-bkup
host itrdbdop01-bkup: itrdbdop01-bkup.toll.com.au at
checkhname: aliases:

C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -pn
expecting response from server tardis-bkup
aitrdbdop01-bkup.toll.com.au itrdbdop01-bkup
C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -hn tardis-bkup
host tardis-bkup: tardis-bkup.toll.com.au at
checkhname: aliases:

C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -hn itrdbdop01-bkup
host itrdbdop01-bkup: itrdbdop01-bkup.toll.com.au at
checkhname: aliases:

C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -ip
checkhaddr: host   : tardis-bkup: tardis-bkup.toll.com.au at
checkhaddr: aliases:

C:\Program Files\VERITAS\NetBackup\bin>bpclntcmd -ip
checkhaddr: host   : aitrdbdop01-bkup: aitrdbdop01-bkup.toll.com.au at

checkhaddr: aliases:     itrdbdop01-bkup.toll.com.au

From the master server;

$ sudo /usr/openv/netbackup/bin/bpclntcmd -hn itrdbdop01-bkup
host itrdbdop01-bkup: itrdbdop01-bkup.toll.com.au at
aliases:

sudo /usr/openv/netbackup/bin/bpclntcmd -ip
checkhaddr: host   : aitrdbdop01-bkup: aitrdbdop01-bkup.toll.com.au at

checkhaddr: aliases:     itrdbdop01-bkup.toll.com.au
 

[Marianne]

There are two possibilities here: a timeout or name resolution problem.
How long does it take for the backup to fail? The recommendation for database backups is to increase CLIENT_CONNECT_TIMEOUT and CLIENT_READ_TIMEOUT to 1800.
To double-check comms on master server, check bprd log.
Find the connection request from the client IP address. Follow the process through the bprd log to see if the master server resolves the IP to correct hostname as per the policy. All of this need to succeed before bprd hands over to bpdbm for image creation.

[anasser]

Hello see what time backups are failing with status code 6 and check how many backups are queued at that time, there can be also a possibility of performance tunning here the Oracle backups fails with status code 6 and File system backups with status code 47. Let us know what you see there.
And also what version of netbackup are you running.?

[rj_nbu]

Is it 6.5.2 ?

18:37:09.857 [5460.3880] <16> handshake: ERR - didn't receive the right filename from server. sent </bk_a5l1040f_1_1_705695759>, received <SET SNAP_ID=itrdbdop01-bkup_1260861171>.


check this:

http://support.veritas.com/docs/306316

[MariusD]

I have also this problem with the Oracle data base backup.

The backup schedule for Oracle finish with: "the backup failed to back up the requested files(6)" , but if I start a manual backup, works.
What can be?


MfG
Marius Demeter

[Will_Restore]

BUG REPORT: At 6.5.2, SAP and SQL-Server automatic backups intermittently report status 1 and application backups fail with status 6

Workaround:

If this problem is encountered, please contact Symantec Technical Services.

http://seer.entsupport.symantec.com/docs/306393.htm