barcode list corrupted 1

[atquevale]

Due to an unfortunate break-down in a SCSI controller of the acssa server
which manages the juke-box (e il nome vero???), Legato shows the following
message: "barcode list corrupted".

Now, even if I tried the nsrjb -IE and nsrjb -HE commands, I'm not able to
mount tapes in Legato anymore (I get a nasty I/O error).
The tape is immediately unmounted and ejected as soon as I mount it
(mmmhh... seems like the barcode isn't recognized...).
What's more, the label associated to the barcode of the volume I'm trying to
mount is cleared in the window showing the mounted tapes.
Can anyone help me?

 

[ag100]

Hey,

Is the problem with the controller fixed? If not, the system may not see the jukebox properly at all. Have you deleted, then re-added the jukebox?

If so, what happens if you set bar code reader and match bar codes to &quot;No&quot;. Will it let you load tapes through the jukebox? (e.g. nsrjb -l -S<slot number> -f <device>

Let me know..

-ag100

 

[atquevale]

I wasn't able to try the suggested solution because in the meanwhile Legato
started labeling five or six times the same tape in different pools. So I
was forced to shutdown the cluster.
After several attempt to reset and restart the IPL, the juke-box just
remains OFFLINE. Basically, the ACCSA shows as ONLINE only the following
drives:
ACSSA> q port 0,0
2001-05-15 12:43:17 Port Status
Identifier State
0, 0 offline

ACSSA> q acs 0
2001-05-15 12:43:26 ACS Status
Identifier State Free Cell Audit Mount Dismount Enter
Eject
Count C/P C/P C/P C/P
C/P
0 offline 1115 0/0 0/0 0/0 0/0
0/0

ACSSA> q drive
...
2001-05-15 12:44:35 Drive Status
Identifier State Status Volume Type
0, 1, 1, 0 online in use 002320 9840
0, 1, 1, 1 online in use 002020 9840
0, 1, 1, 2 online in use 002277 9840


Note: tapes have been unmounted by hand and, since the stk-acs cannot see
the juke-box, there was no update of the database.

When the juke-box was still online, we tried an audit, but no changes in the
situation occurred.

Nowadays, the juke-box IPL stop at init8 (during the audit).
The ACSSS database in up and running (the Oracle DBA checked the logs; after
the disk recovery the situation became normal once again).

Is there something you would advice me to do WITHOUT restart Legato?

Thanks.

 

[jkalbus]

Hi,

we had the same problem after a crash. What helped was to manually edit the nsrjb.res file located in /nsr/res.

There are lines that should look like that (or similar) when no tapes are loaded.

loaded bar codes: &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;;
loaded slots: &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;;
loaded volumes: &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;, &quot;&quot;;

They had an asterik or a 999999 in some places, and so networker refused to work with us. Editing these lines so that they looked like the above example helped and after an audit everything went fine.

Although Legato says never ever to edit .res files manually I think it should be save to do so when networker is down and you know what you are doing, else you could cause a lot of havoc.

Johanes

 

[ericmo]

What I do to reset :

$ nsradmin
NetWorker administration program.
Use the "help" command for help, "visual" for full-screen mode.
nsradmin> . type: NSR jukebox; name:"Silo"
Current query set
nsradmin> update loaded volumes:" "," "
loaded volumes: " ", " ";
Update? y
updated resource id 0.23.83.217.65.173.243.184.10.4.131.210(16550)
nsradmin> update loaded bar codes:" "," "
loaded bar codes: " ", " ";
Update? y
updated resource id 0.23.83.217.65.173.243.184.10.4.131.210(16550)
nsradmin> update loaded slots:" "," "
loaded slots: " ", " ";
Update? y
updated resource id 0.23.83.217.65.173.243.184.10.4.131.210(16550)
nsradmin> quit

 

[ericmo]

No space :
update loaded volumes:"",""