cancel
Showing results for 
Search instead for 
Did you mean: 

upgrade to 6.5 from 6.0 mp7 failing "unable to create/upgrade database"

dennis_sexton
Level 5

Have a solaris 10 v890 i'm trying to upgrade from netbackup 6.0 mp7 to netbackp 6.5. All seemed to go pretty well until

the database buildl. After  says it is unable to create/update databaseInstallation completed successfully

Starting the NetBackup Event Manager.

Database server is NB_carnival
Language resource file (dblgen9.res) version mismatch.
Language resource file (dblgen9.res) version mismatch.
Database server [NB_carnival] not available.

Unable to create/upgrade the NB database.  Refer to the
log file in /usr/openv/netbackup/logs/nbdb for more
information.  Rerun /usr/openv/netbackup/bin/install_bp
when the problem has been resolved.

The above log file lookes like

10:57:14.257 [12206] <4> create_nbdb: Entering.
10:57:14.257 [12206] <4> create_nbdb: /usr/openv/db/bin/create_nbdb
10:57:14.258 [12206] <4> create_nbdb: ASANY = /usr/openv/db
10:57:14.258 [12206] <4> create_nbdb: ASANY9 = /usr/openv/db
10:57:14.258 [12206] <4> create_nbdb: LD_LIBRARY_PATH = /usr/openv/db/lib
10:57:14.258 [12206] <4> create_nbdb: ASTMP = /usr/openv/tmp
10:57:14.259 [12206] <4> pingDBServer: Entering.
10:57:14.259 [12206] <4> run_nbdb_cmd: Entering.
10:57:14.259 [12206] <4> run_nbdb_cmd: Running . /usr/openv/db/vxdbms_env.sh && "/usr/openv/db/bin/dbping" -q -
c "ENG=NB_carnival;LINKS=shmem,tcpip{PORT=13785}"
10:57:14.302 [12206] <16> run_nbdb_cmd: . /usr/openv/db/vxdbms_env.sh && "/usr/openv/db/bin/dbping" -q -c "ENG=
NB_carnival;LINKS=shmem,tcpip{PORT=13785}" failed.
10:57:14.302 [12206] <4> run_nbdb_cmd: Exiting with rc = 77
10:57:14.302 [12206] <4> pingDBServer: Sleeping for 60 seconds and will retry.
10:58:14.309 [12206] <4> run_nbdb_cmd: Entering.
10:58:14.309 [12206] <4> run_nbdb_cmd: Running . /usr/openv/db/vxdbms_env.sh && "/usr/openv/db/bin/dbping" -q -
c "ENG=NB_carnival;LINKS=shmem,tcpip{PORT=13785}"
10:58:14.352 [12206] <16> run_nbdb_cmd: . /usr/openv/db/vxdbms_env.sh && "/usr/openv/db/bin/dbping" -q -c "ENG=
NB_carnival;LINKS=shmem,tcpip{PORT=13785}" failed.
10:58:14.352 [12206] <4> run_nbdb_cmd: Exiting with rc = 77
10:58:14.352 [12206] <16> pingDBServer: Database server [NB_carnival] not available.
10:58:14.352 [12206] <4> pingDBServer: Exiting with rc = 12
10:58:14.352 [12206] <4> getExistingDBFile: Entering.
10:58:14.354 [12206] <4> getExistingDBFile: Exiting with rc = 0
10:58:14.354 [12206] <16> create_nbdb: Database server required to continue.
10:58:14.354 [12206] <4> create_nbdb: Exiting with rc = 12

 

any help appreciated. i've been dropped from the phone tree 3 times already.

7 REPLIES 7

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Please find the installation log under /usr/openv/tmp. It should be something like 'trace_install_bp.####'

Please post as attachment.

 

As far as Support call is concerned - log a call online first: https://mysupport.symantec.com/

You will receive a case id. Then phone again and enter case # when prompted.

dennis_sexton
Level 5

Yes. I did log call  online first. After I called in and   put in the queue, I've been dropped 3 times :(

install_trace attached

Marianne
Moderator
Moderator
Partner    VIP    Accredited Certified

Sorry Dennis - Installation log contains same error...

I've been searching for the following: Language resource file (dblgen9.res) version mismatch
 - I cannot find any reference....

I have been searching our lab master for dblgen9.res, but we're already on 7.0.1 - the file does not exist.

Hope you get through to an engineer soon....

MCron
Level 4
Partner

I have seen this a couple of times in the past.  I believe in most of the times that I have seen this, it came down to a communcations issue.  The way that NBU communicates internally has been tightened up over time, so 6.5 is tighter than 6.0, and 7.0 is even more so.  Things that you could get away with in your name resolution for instance has just gotten harder.

Taking a wild guess here..  how many interfaces does this server have, and does each one have a listing in the bp.conf file?  Do all of them resolve?

Also the entries in bp.conf for the non reserved ports has given me grief as well.

-Mark

dennis_sexton
Level 5

system has only one interface so there is only 1 entry in bp.conf. However there is a difference in bp.conf. Don't

SERVER = $hostname.$Domain.EDU
CLIENT_NAME = $hostname.$domain.edu
EMMSERVER = $hostname.$domain.edu

Note the caps in domain and edu.

/usr/openv/db was a link to a zfs file system in the original 6.0 environment. If I move that link out of the way and

create /usr/openv/db as a directory bp_install still fails and says it can't find /usr/openv/db/bin/nbdbms_server_install

that file isn't there and apparently bp_install doesn't fill in the /usr/openv/db directory. I am considering tar'ing

from the install dvd everythning under solaris/SMCnetbp/reloc/lopev/db into /usr/openv/db and re-running bp_install

but am slightly hesitant to do that. Does this seem like a reasonable thing to try?

(still waiting for a call back :( )

Yogesh9881
Level 6
Accredited

What is your current NBU version ???

you can run NBCC and post O/P

dennis_sexton
Level 5

Well I finally heard from symantec. We managed to fix the problem by putting data from the 6.0 mp7 /usr/openv/db into that directory and re-running install_bp.  So I am now up and running on 6.5.

Of course there still seems to be some issue with the configuration since I'm now freezing tapes with no data but I'm also getting some backups and am able to do restores.