virtual segment question

Posted in 2014

Summary

User on IDS 10.00.FC6 saw ~47 virtual shared memory segments and huge memory attributed to the rsam pool, and onmode -F wouldn't free it. Suggestions were to use onstat -g mem to find which sessions consumed the memory. The poster then found from online.log that dynamic lock allocation was the cause. Advice: size the LOCKS parameter properly to avoid repeated doubling, raise it in onconfig, and find/fix applications using excessive locks (e.g. break huge deletes/updates/inserts into smaller transactions or use truncate). No confirmation of the final outcome is recorded.

Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.

Topics: General Discussion

CHUAN LU — Tue, 03 Jun 2014 02:36:39 — source: IIUG Forums & Mailing Lists

HI, I use IDS10.0FC6 and found about 47 virtual segment. SHMVIRTSIZE= 1024000 SHMADD= 256000 ' onstat -g mem' show there are lots of memory usage in rsam. rsam V 7000001501b7040 11665981440 110064 117571 111 Whu ISAM can use so many memory? any suggestion would be appreciate.

↪ replying to CHUAN LU

Art Kagel — Tue, 03 Jun 2014 03:16:18 — source: IIUG Forums & Mailing Lists

Does onmode -F release the extra segments? Art Art S. Kagel, Principal Consultant ASK Database Management Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Mon, Jun 2, 2014 at 10:36 PM, CHUAN LU <luchuan@cn.ibm.com> wrote: > HI, > > I use IDS10.0FC6 and found about 47 virtual segment. > > SHMVIRTSIZE= 1024000 > > SHMADD= 256000 > > ' onstat -g mem' show there are lots of memory usage in rsam. > > rsam V 7000001501b7040 11665981440 110064 117571 111 > > Whu ISAM can use so many memory? > > any suggestion would be appreciate. > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --001a11c3cca229477804fae5efa6

↪ replying to Art Kagel

CHUAN LU — Tue, 03 Jun 2014 05:17:33 — source: IIUG Forums & Mailing Lists

onmode -F doesn't do help.

↪ replying to CHUAN LU

david@smooth1.co.uk — Tue, 03 Jun 2014 07:43:39 — source: IIUG Forums & Mailing Lists

Run onstat -g mem and see which sessions are using the memory. David. > On 03 June 2014 at 06:17 CHUAN LU <luchuan@cn.ibm.com> wrote: > > > onmode -F doesn't do help.> > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. >

↪ replying to david@smooth1.co.uk

CHUAN LU — Tue, 03 Jun 2014 08:39:52 — source: IIUG Forums & Mailing Lists

I read the online.log, it's dynamic lock allocation lead to the problem

↪ replying to CHUAN LU

Art Kagel — Tue, 03 Jun 2014 10:18:23 — source: IIUG Forums & Mailing Lists

Ahh, then you have to determine a) how many locks you really need and correctly configure LOCKS to avoid the doubling behavior, and b) figure out what applications or users are using excessive numbers of locks and fix that problem. For example, if you have an application that is performing massive deletes, either change it to use truncate or use my dbdelete utility (or at least the method that dbdelete uses to avoid using huge numbers of locks) instead. If it's an interactive user performing massive global updates or a batch job inserting millions of rows in a single transaction, change the procedure to break up the changes into smaller chunks. Art Art S. Kagel, Principal Consultant ASK Database Management Blog: http://informix-myview.blogspot.com/ Disclaimer: Please keep in mind that my own opinions are my own opinions and do not reflect on the IIUG, nor any other organization with which I am associated either explicitly, implicitly, or by inference. Neither do those opinions reflect those of other individuals affiliated with any entity with which I am affiliated nor those of the entities themselves. On Tue, Jun 3, 2014 at 4:39 AM, CHUAN LU <luchuan@cn.ibm.com> wrote: > I read the online.log, it's dynamic lock allocation lead to the problem > > > > ******************************************************************************* > Forum Note: Use "Reply" to post a response in the discussion forum. > > --047d7b3a8aa893c55a04faebd479

↪ replying to CHUAN LU

LLOYD SERRAO — Wed, 04 Jun 2014 10:48:08 — source: IIUG Forums & Mailing Lists

Did you try to increase the number of locks in the LOCKS parameter in Onconfig? Lloyd

Related threads

Glossary

onstat