“Missing Path” or “Unused element address”

There can be many reasons causing this error, but you can try the folllowing:

Update your drivers from the following url, then delete and re-configure tape drives on Netbackup.

http://seer.entsupport.symantec.com/docs/287850.htm

server name not found in the NetBackup configuration

1. You were trying to expire a media from emm database:

bpexpdate -m media_id -d 0

Solution: Try to add -host media_server_hostname to end of the command

bpexpdate -m media_id -d 0 -host media_server

Please add your knowledge about this error…

Thanks.

2. Clean a tape drive manually
tpclean -C <drive name>

3. Find backup images of a client for spesific date range.
bpimagelist -U -client <CLIENT_NAME> -d <START_DATE> -e <END_DATE>

4.  Find tape media used for backup of a client for spesific date range.
bpimagelist -U -client <CLIENT_NAME> -d <START_DATE> -e <END_DATE> -media

5. Monitor backup jobs
/usr/openv/netbackup/bin/admincmd/bpdbjobs -report

6. Howto re-use used tape media (That is: to re-use a tape media used by another system or older netbackup software)
“Expire that media”
bpexpdate -ev media_id -d 0 -force -host NBUHostname
“De-assign that media”
vmquery -deassignbyid media_id 4 0
“Move that media to a volume pool (preferably a Scratch pool)”
vmchange -m media_id -p Scratch_Pool_Number
“Re-label that media”
bplabel -ev media_id -d robot_type -p Pool_name

7. Check drive usage
/usr/openv/volmgr/bin/vmoprcmd

8. Take a drive down/up
/usr/openv/volmgr/vmoprcmd -down <drive_index>
/usr/openv/volmgr/vmoprcmd -up <drive_index>

9. Howto unfreeze a media
“Find the frozen media id’s”
/usr/openv/netbackup/bin/goodies/available_media | grep -i frozen
“Unfreeze the media according to the media id found above”
bpmedia -unfreeze -ev <media_id> -h <media_server>

10. Delete a tape media from NetBackup
“Make sure no images exist on the tape”
bpimmedia -mediaid media_id -L
“Expire the tape”
bpexpdate -ev media_id -d 0 -host NBUHostname -force
“Find the pool name of the tape and make sure the status of that tape”
vmquery -m media_id
“De-assign the tape”
vmquery -deassignbyid <media_id> <pool_number> <status from vmquery -m>
vmquery -deassignbyid media_id 4 0×0
“Delete the tape from NetBackup”
vmdelete -m media_id

12. Properties of a volume pool
vmpool -listall

RMAN> run {
1> ALLOCATE CHANNEL chtape TYPE ‘SBT_TAPE’;
2> set until time ’2008-11-05:14:30:00′;
3> SEND ‘NB_ORA_CLIENT=realdb_host,NB_ORA_POLICY=rman_realdb,NB_ORA_SERV=veritas,NB_ORA_SCHED=Default-Application-Backup’;
4> RESTORE CONTROLFILE;
5> RELEASE CHANNEL chtape;
6> }

executing command: SET until clause
released channel: chtape
RMAN-00571: ===========================================================
RMAN-00569: =============== ERROR MESSAGE STACK FOLLOWS ===============
RMAN-00571: ===========================================================
RMAN-03002: failure of set command at 01/06/2009 15:07:58
ORA-01861: literal does not match format string

This can be caused by your date format. Set your NLS date format accordingly and retry your restore procedure.
/home/oracle> export NLS_DATE_FORMAT=”YYYY-MM-DD:HH24:MI:SS”

RMAN> run {
1> ALLOCATE CHANNEL chtape TYPE ‘SBT_TAPE’;
2> set until time ’2009-01-05:14:30:00′;
3> SEND ‘NB_ORA_CLIENT=realdb_host,NB_ORA_POLICY=rman_realdb,NB_ORA_SERV=veritas,NB_ORA_SCHED=Default-Application-Backup’;
3> RESTORE CONTROLFILE;
4> RELEASE CHANNEL chtape;
5> }

allocated channel: chtape
channel chtape: sid=1014 devtype=SBT_TAPE
channel chtape: Veritas NetBackup for Oracle – Release 6.5 (2008052301)

executing command: SET until clause

Application connection failed for servername\. Please select a different SQL server host and instance, or press ‘cancel’ to exit the NetBackup for SQL Server Graphical User Interface.

According to this page, this error is caused by windows 2008′s “User Account Control”. You need to run”Netbackup Ms-Sql client” with “Run As Administrator” privilege.

Right click “Netbackup Ms-Sql client” and select “Run as Administrator” and “Apply”. Try to connect to ms-sql server again and you should be able to see databases now…

You have a unix based netbackup master server and have an windows server with oracle that you want to backup through rman catalog backup, but you are unable to use the templates created by the “Backup, Archive and Restore” application on windows.

On your windows server, you can convert your templates to rman scripts using the following line…

c:\Program Files\Veritas\Netbackup\bin\bpdbsbora -backup -g Full_backup.cmd -t Full_backup.tpl

Now you can enter this script on unix side as a backup selection script…

Switch to user oracle

cd $ORACLE_HOME/lib
rm -f libobk.so
ln -s /usr/openv/netbackup/bin/libobk.so64.1 libobk.so
(use libobk.so.1 instead of libobk.so64.1 if you’re not using 64Bit OS)
ls -al libobk.so
lrwxrwxrwx 1 oracle dba 38 Dec 23 15:13 libobk.so -> /usr/openv/netbackup/bin/libobk.so64.1

Now you can retry your backup

There may be more than one cause for this problem, here is a solution for me.
You might have two seperate ip definitons on /etc/hosts file with the same hostname for your netbackup master server. e.g
172.6.2.10 veritas
192.168.14.10 veritas
You are connecting through 192.168.14.10 interface but in /etc/hosts file 172.6.2.10 comes first, so netbackup client doesnt recognize netbackup master server correctly.
So, correct this problem and it should run through…

Since it won’t have any media id, you will not be able to see it on “netbackup administration console” and won’t be able to eject the tape out of library.

Instead of manually removing the tape from the library, you can use “robtest” to eject the tape. Follow these steps:You need to know which slot is tape inserted in the library. In this example tape is inserted on the slot 10

1. Start robtest, and choose you tape library, here TLD is selected.

/usr/openv/volmgr/bin/robtest
Configured robots with local control supporting test utilities:
TLD(0)    robotic-path = /dev/sg1

Robot Selection
———————
1) TLD 0
2) None/quit
Enter Choice: 1

Robot selected: TLD(0)   robotic-path = /dev/sg1
Opening /dev/sg1
MODE_SENSE complete

…..

Enter tld commands

2. Move your media from library slot 10 to media access port 1

m s10 i1

Initiating MOVE_MEDIUM from address 1009 to 10
MOVE_MEDIUM complete
Quit

3. Your tape is now on media access port slot 1, and can be removed easily.

vmquery -m <cleaning tape #>

C:\>vmquery -m CLN007
======================================
media ID: CLN007
media type: 1/2″ cleaning tape 3 (25)
barcode: CLN107L3
media description: —
volume pool: None (0)
robot type: TLD – Tape Library DLT (8)
robot number: 0
robot slot: 4
robot control host: Netbackup
volume group: 00_000_TLD
vault name: —
vault sent date: —
vault return date: —
vault slot: —
vault session id: —
vault container id: -
created: 7/11/2008 6:29:48 PM
assigned: —
last mounted: 8/11/2008 8:20:22 AM
first mount: 7/11/2008 7:17:04 PM
expiration date: —
cleanings left: 27
======================================

Frequency of cleaning with cleaning tapes
With modern drives, cleaning is done only when drives say they need…
This is done via “Tape Alert” feature. You configure NetBackup so that the cleaning frequency is set to 0, and it will clean drives only when they request to be cleaned.

Keep in mind that, cleaning a drive too often can cause more harm than good. You should find a good frequency that works for you.

If you face an error like “daemon failed accepting connection 59”

This error can be caused by different things. But first of all, check that there are enough “cleanings remaining” on the cleaning tape used.

Cleaning Tapes from command line

Show the drive name to be cleaned

tpclean -L

Manually clean the drive:

tpclean -C <drive_name>

Add

VERBOSE = x

line to bp.conf, where x is the number to set how verbose the logs will be.

Where are Netbackup logs?

Under /usr/openv/netbackup/logs on Unix or <INSTALL DIRECTORY>\VERITAS\netbackup\logs on windows systems.

Where are tape drive errors?

/usr/openv/netbackup/db/media/errors on Unix or <INSTALL DIRECTORY>\VERITAS\NetBackup\db\media\errors on windows systems.

2. The partiton where netbackup is installed must not be 100% full, or netbackup will basicaly stop functioning.

3.If dns does not work and there are no records on /etc/hosts netbackup will again stop functioning and netbackup will need to restart to be able to work again.

4. Some commands for troubleshooting:

a. /usr/openv/netbackup/bin/bpclntcmd -hn client_name

Returns client_name’s ip address if netbackup can do a successful lookup.

b. /usr/openv/netbackup/bin/bpclntcmd -hn client_ip_address

Returns client’s address, if netbackup can do a successfull reverse dns lookup

c. /usr/openv/netbackup/bin/bpexpdate -m 00m4l3 -d 0

Allows you to expire a media when needed.

d.bpstulist

Will show you all drives and hosts connected to master server

e. /usr/openv/volmgr/bin/tpautoconf -a

Will scan all local device on media manager.

f. /usr/openv/volmgr/bin/vmoprcmd

Shows information about your media servers.

g. /usr/openv/netbackup/bin/admincmd/bpmedialist -rl 1

Shows info about media that’s available, full, expired or other.

h. /usr/openv/netbackup/bin/admincmd/bpmedia -unfreeze -m 06m03l3

Unfreezes a media.

i. /usr/openv/volmgr/bin/vmupdate -rt tld -rn 1 -full [-interactive] -use_barcode_rules

Inventory a robot.

j. /usr/openv/netbackup/bin/bpmedialist -rl 1

List which disks are expired, full or available. (-rl used for retention level. 1 means retention of 2 weeks)

k. /usr/openv/netbackup/bin/bpps -x

All running netbackup processes are listed

l. /usr/openv/netbackup/bin/goodies/available_media

All media and its states are listed. The script runs other commands and formats the outputs to a usable format.

m. /usr/openv/netbackup/bin/admincmd/bpdbjobs

Activity monitor.

n. bplabel -ev media_id_number -d <tape density> -p Pool_Name

Label a media with “media_id_number” in pool “Pool_Name”.

o. bpimmedia -mediaid 0m0687 -L

List images on a tape with media id 0m0687

p. vmdelete -m 000687

Delete a media from netbackup emm database