Tilo,

Here are the results of the db_restartinfo:

dbmcli on sandbox : DEV>db_restartinfo
OK
Used LOG Page           0
First LOG Page          2147483647
Restartable             1
Id Restart Record       twobase:DEV_20040830_195645
Id LOG Info             twobase:DEV_20040830_195645
Consistent              1

(The "twobase" name in those lines above refers to the source server.
The server upon which I'm trying to restore this data is named
"sandbox".)

It took me a while to figure out how to use backint to retrieve the log
data from Data Protector, but this was the process I followed:

----------------------------------------
/usr/sbin/mknod /var/opt/omni/tmp/DEV.BACKDP-Archive.1 p

/opt/omni/lbin/sapdb_backint -u DEV -f restore -t file -p 1111164226 -i
/var/opt/omni/tmp/DEV.bsi_in -c
----------------------------------------

The file /var/opt/omni/tmp/DEV.bsi_in contained:

----------------------------------------
0503091:1 Stream /var/opt/omni/tmp/DEV.BACKDP-Archive.1
----------------------------------------

However, even this did not work until I set the following environment
variables:

----------------------------------------
setenv OB2APPNAME DEV
setenv OB2BARHOSTNAME sandbox.hb.local
setenv TimeoutWaitFiles 30
setenv OB2BACKUPAPPNAME DEV
setenv OB2BACKUPHOSTNAME twobase.hb.local
setenv BI_CALLER DBMSRV
setenv BI_REQUEST NEW
setenv BI_BACKUP ARCHIVE
----------------------------------------

After these were set, sapdb_backint finally called Data Protector to
start sending the data.  While I was waiting, I had my
medium_labeloffline process ready in another window.  Unfortunately,
this was all the about I received:

----------------------------------------
dbmcli on sandbox : DEV>medium_labeloffline LOGPipe
OK
Returncode              0
Date
Time
Server
Database
Kernel Version
Pages Transferred       0
Pages Left              0
Volumes
Medianame
Location                /var/opt/omni/tmp/DEV.BACKDP-Archive.1
Errortext
Label
Is Consistent
First LOG Page          0
Last LOG Page
DB Stamp 1 Date
DB Stamp 1 Time
DB Stamp 2 Date
DB Stamp 2 Time
Page Count              0
Devices Used            0
Database ID
Max Used Data Page
----------------------------------------

As you can see, it's empty.  I will note, however, that the
medium_labeloffline command waited until my Data Protector GUI indicated
that data was actually being read from tape.  As soon as the first bytes
of data were transmitted, the medium_labeloffline command immediately
returned the output noted above.  The Data Protector process ran for at
least a minute after that, reading more data from the tape.

So from my perspective, it's like SAP DB is trying to read the log data
before any substantial amount arrives from tape.

Here's something else that's been troubling me regarding the whole issue
of logs...  During normal operations, we run with AUTOLOG enabled.  This
causes log segments to create version files in (for us)
/sapdb_backup0/logs/log.xyz.  These version files, in turn, are
transmitted to Data Protector via the pipe
/var/opt/omni/tmp/DEV.BACKUP-Archive.1.  Now then...  When we start
starting about restores, should I assume that this process would
basically be reversed -- data should flow from Data Protector through
the pipe and wind up in /sapdb_backup0/logs?  Because if that's the way
it should work, it's not, because this system's /sapdb_backup0/logs
directory hasn't been written to in over two weeks.

Thanks, Tilo, and everyone else for your help,
~Fred

--
MaxDB Discussion Mailing List
For list archives: http://lists.mysql.com/maxdb
To unsubscribe:    http://lists.mysql.com/[EMAIL PROTECTED]

Reply via email to