Heres several ways to do some checking.
1. q volume access=readonly and q volume access=uanvailable
This will give you list of tapes that have been put in this state most likely
by the system due to some type of error. (Unavlaibale can be caused by ADSM
trying to mount a tape that is not in the library - unavailble) Then you can
work with these tapes on the list.
2. select volume_name, read_errors,write_errors from volumes where (read_errors
> 0 or write_errors > 0)
This will give a list of tapes that have more than 1 read or write error. In
other words you canm conecntrate on tapes with a lot of errors first.
3. I have not used the "audit" commands. A sort of way I have done this if I
know/think there is a problem with a tape is to do this. Make a list of the
tapes. Do a "move data" on the tapes. This will move all data off of the tape
onto another scratch tape or another tape in that stgpool. When done the tape
is a scratch. Then see if that tape comes up on the problem list again, if so
I move data again and then chunk the tape/get a replacement.
If there are problems reading the tape and it can't move all data then I do a
"restore volume" if this is a primary volume, and it will get data from
copytapes to restore this volumes data to another tape. (Note: Use the
preview=yes for each tape first and then look at actlog and it will show you
all the tapes that are needed in case some are not in library you can request
them.)
Last of all is the "delete volume" command. I don't know if anyone has another
solution if you get to this point. It may depend on how youir stgpools and
copy pools are set up if there is another option.
David B. Longo
Systems Administrator
Health First, Inc. I/T
3300 Fiske Blvd.
Rockledge, FL 32955-4305
PH 321.434.5536
Pager 321.634.8230
Fax: 321.434.5525
david.longo AT health-first DOT org
>>> Hintze.Kai AT AMSTR DOT COM 12/08/00 12:31PM >>>
Greetings all!
Summary: Does anyone have a script to audit contents of all our tapes, or
other suggestion on how to verify what they actually contain?
Discussion:
As I have mentioned before, we have had serious, ongoing problems with our DLT
drives. We think we may have them resolved, finally. At this point I have no
known problems with drives or media, but I am concerned that some of the tapes
may have been damaged. I don't want to throw them all out on a maybe, but I
think I should do a surface scan of all our tapes to try to spot damage or
incomplete backups. Then I can throw out the few that don't repair immediately.
Reading the manuals it looks like I want to "audit contents" on each volume in
the library. But just doing an "audit contents *" would start up 300+
processes all waiting for their turn at a drive. So I need a script to start
one at a time, and wait until it finishes before it starts a new one. It will
need to run nohupped or be interuptible because extrapolating from the couple
of audits I have run as tests this will run for a couple of weeks, and my PC
isn't that stable.
Platform is ADSM server 3.1.2.20, command line interface 3.1.0.0, running on
AIX 4.1.5.
Before I invent a new wheel, I wonder if anyone has one they will let me copy?
Or is there a better way to verify that all our media are intact?
Please respond to me directly at hintze.kai AT amstr DOT com, as I get the list
in
digest form. That is sufficient to keep a finger on what is happening, but
sometimes I miss details in the jumble.
Thanks,
Kai.
|
