RE: Migration from IDS 7.25 to 7.31
We have been on 7.31.UC4 for almost two weeks now and it appears to have
fixed the bug 115327 for us. We were on 7.31.UC2 and had terrible
performance. Everything is okay now. onstat -R and onstat -P are good and
we have not had any complaints and have not needed to bounce the database at
all. (HP-UX 10.20)
Dianne
-----Original Message-----
From: Pratt, Fred [mailto:prattf@hastings-ent.com]
Sent: November 04,1999 10:19 AM
To: 'kagel@bloomberg.net'; informix-list@iiug.org
Subject: RE: Migration from IDS 7.25 to 7.31
FYI. According to Informix tech support, the bug 115327 fix is "coded, but
not tested" in 7.31.uc4 and up. Likewise for 9.20.uc1 and higher.
7.30.uc10 and above have been coded *and* tested.
HTH
> -----------
> Regards,
>
> Fred Pratt
> Informix DBA PHONE
> (806) 351-2300 x/3660
> Hastings Entertainment, Inc. FAX (806)
> 351-2727
> http://www.hastings-ent.com EMAIL
> prattf@hastings-ent.com
>
>
>
> -----Original Message-----
> From: Art S. Kagel [SMTP:kagel@bloomberg.net]
> Sent: Tuesday, November 02, 1999 3:30 PM
> To: informix-list@iiug.org
> Subject: Re: Migration from IDS 7.25 to 7.31
>
> Daniel Esquerre Fernandez wrote:
> >
> > Hi ALL
> >
> > Excuse me i am an old infromix adm, but my expirence was
> developed
> > in Online 5.x and 4GL RDS 4.x , this year migrate to IDS 7.25 and 4gl
> 7.20
> > intalled on an RS6000 f50 AIX 4.3, in these moths we get some
> performance
> > problems, concentrate in a few days of high activity, we get a bug know
> :
> >
> > Bug de Informix ---
> >
> > Bug: 105813 SM_POLL THREAD CAN ASSERT FAIL BECAUSE OF TIMING ISSUE
> > BETWEEN AN INTERUPTED SHARED MEMORY CONNECTION AND ITSELF
>
> Someone interrupted a user program using kill -9 or kill -TERM or ^C and
> left some shared memory data corrupted. The engine crashed because it
> could not figure out how to correct the problem.
>
> > we still waiting for a 7.24.UC9 from Agost, a few days ago we have
> recived a
> > 7.31 as an UPDATE. are there some advice about this new version?,
> somebody
> > knows a migration way, is it a good solution to fix the bug?
>
> Version 7.31 can upgrade you from 7.24 automatically, as long as you do
> not have any incomplete In-Place ALTERs in progress. You can check this
> with oncheck -pT database:table. At the end of the report is a report of
> the different record versions contained in the table and how many of each
> there are. For any tables which have any rows for a version other than
> (current) execute an update of every row by setting some column to itself.
>
> This will force the engine to complete the ALTER and update every row to
> the new format.
>
> IDS 7.31 is significantly faster than 7.24 and less buggy to boot. If you
>
> have 7.31UC3 or earlier upgrade to get a version that does not have a
> particular bug (#115327) which is either 7.31UC3 or UC4 I don't remember.
>
> Art S. Kagel