home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.databases.sybase
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!darwin.sura.net!spool.mu.edu!uwm.edu!linac!sunova!higgs.ssc.gov!jbaron
- From: jbaron@higgs.ssc.gov (Jeff Baron)
- Subject: Re: Recovering from a log segment full condition
- Message-ID: <1992Dec22.173634.4259@sunova.ssc.gov>
- Sender: usenet@sunova.ssc.gov (News Admin)
- Nntp-Posting-Host: higgs.ssc.gov
- Reply-To: jbaron@higgs.ssc.gov (Jeff Baron)
- Organization: Superconducting Super Collider Lab
- References: <1992Dec16.205611.19776@spectrum.xerox.com> <dave.724703604@rincon> <WFINNERT.92Dec21194611@larry.shearson.com>
- Date: Tue, 22 Dec 1992 17:36:34 GMT
- Lines: 23
-
- Warren Finnerty writes:
- -
- - Strange that model being full would be a problem.
- - ( I'll take your word for it. :) )
- - Do any of you SYBASE types want to confess to what futz'ing around in model
- - goes on in conjunction with createing objects in tempdb. ( Not at startup )
-
- This has happened to me as well. At an old job, they have a very
- small Sybase setup. One day, almost 18 months after I had left,
- they called with a very strange problem; turned out the the
- transaction log allocated to model (ie. the entire space) had
- filled.
-
- There is one piece of information that *is* kept in model; perhaps
- it is a duplicate pointer to other transaction logs, or something of
- a like systemish nature. It does change, and the transaction log
- *will* fill eventually. (This was server version 4.0 or somesuch.)
-
- Take a look at yours today!
-
- --
- Jeff Baron
- jbaron@gauss.ssc.gov
-