While ase was starting last night, it went through it’s normal
start-up procedures of recovering each database in turn. However
when it got to one particular database the following was noted in
the log “00:00000:00001:2004/09/09 23:09:04.34 server
Database 'bo_prd' is now online.
00:00000:00001:2004/09/09 23:09:05.59 server Recovering
database 'xamin_mdl'.
00:00000:00001:2004/09/09 23:09:06.64 server Redo pass of
recovery has processed 506 committed and 0 aborted transactions.”
And it just sat there for 2 hours doing nothing. My problem was
that the main database that users require had a higher id number,
hence would only be recovered after this troublesome database. I
had to quickly somehow get ase to skip recovery of this database
an carry on, I edited sysdatabases table and gave the troublesome
database a higher id number, recycled the ase, and all the other
dbs recovered fine. My questions:
1. any ideas/has anyone seen sybase hang like this before.
2. is changing the id of the database a bad thing to do !
3. what should I have done in this situation.

For your info, we routinely shutdown with nowait.

Any help would be appreciated. Thanks
--
[ Remove No.Reply.For_ from email address to reply via email ]

"Anthony Mandic" <am@*t.com.au> wrote

: marek wrote:
: >
: > While ase was starting last night, it went through it’s normal
: > start-up procedures of recovering each database in turn. However
: > when it got to one particular database the following was noted in
: > the log “00:00000:00001:2004/09/09 23:09:04.34 server
: > Database 'bo_prd' is now online.
: > 00:00000:00001:2004/09/09 23:09:05.59 server Recovering
: > database 'xamin_mdl'.
: > 00:00000:00001:2004/09/09 23:09:06.64 server Redo pass of
: > recovery has processed 506 committed and 0 aborted transactions.”
: > And it just sat there for 2 hours doing nothing. My problem was
: > that the main database that users require had a higher id number,
: > hence would only be recovered after this troublesome database. I
: > had to quickly somehow get ase to skip recovery of this database
: > an carry on, I edited sysdatabases table and gave the troublesome
: > database a higher id number, recycled the ase, and all the other
: > dbs recovered fine. My questions:
: > 1. any ideas/has anyone seen sybase hang like this before.
:
: Never!
:
: > 2. is changing the id of the database a bad thing to do !
:
: Its not the smartest thing to do. There is an acceptable
: method of setting the recovery order. Unfortunately, I can't
: recall all the details and I don't have access to a server to
: look it up (and, oddly, its not in Rob Verschoor's CQRG nor
: in Jeff Garbus's ASE Administrator's Guide). Anyway, there's
: a field in one of the system tables that lets you set the
: recovery order for databases.

sp_dbrecovery_order [database_name [, rec_order [, force]]]

See Ref Man for details.

HTH,

/Stefan

You're looking for 'sp_dbrecovery_order'...

It's described in the first edition of my ASE QRG on page 73/74; in the 2nd
edition on page 83 and in the 3rd on page 87.


HTH,

Rob V.

Thanks. Didn't realise you could change the start order.