Latch crashing engine

Latch crashing engine

Post by TRiever8 » Sun, 31 Dec 1899 09:00:00



Can anyone provide any insight about "Process Aborted Abnormally (latch);"
message crashing the engine.?

This is happening on Solaris 2.6 and Informix 5.10 systems that have been
recently upgraded from Solaris 2.5 and Informix 5.03 or higher.  These older
system did latch on occassion, but not with the frequency that they do now.

We have worked with Informix and used a debugable engine and have determined
that signals coming from our 4gl application are the cause, Informix has not
provided much insight other than don't allow control-c during critical
processes.
We have tried blocking any interreupts from the application with no success.

This problem seems to go away with an upgrade to Informix 7.x, which is not
always cost effective.

Is there anything to look for or do or not do?.

 
 
 

Latch crashing engine

Post by David William » Sun, 31 Dec 1899 09:00:00




Quote:>Can anyone provide any insight about "Process Aborted Abnormally (latch);"
>message crashing the engine.?

>This is happening on Solaris 2.6 and Informix 5.10 systems that have been
>recently upgraded from Solaris 2.5 and Informix 5.03 or higher.  These older
>system did latch on occassion, but not with the frequency that they do now.

>We have worked with Informix and used a debugable engine and have determined
>that signals coming from our 4gl application are the cause, Informix has not
>provided much insight other than don't allow control-c during critical
>processes.
>We have tried blocking any interreupts from the application with no success.

  Do not send signals to the sqlturbo process!

Quote:>This problem seems to go away with an upgrade to Informix 7.x, which is not
>always cost effective.

>Is there anything to look for or do or not do?.

--
David Williams

 
 
 

1. latch levels and latch deadlock

Hello,

I've some questions:

- If higher level latch is requiested in willing-to-wait mode while
holding lower level latch, will process sleep whether or not this
higher level latch is available ?

- how does process continue its processing in no-wait mode ? what is
the mechanism ? to prevent from latch deadlock, how does process
release its holding higher level latches ? is there a latch recovery ?

regards...

Sent via Deja.com http://www.deja.com/
Before you buy.

2. Dynamic SQL -Howto set a col to NULL

3. Redo copy latch and redo allocation latch

4. Migrating btrieve 5 to 6.15

5. Jet Engine Crashes

6. Jet Engine crashes

7. Jet Engine Crashing

8. DBD::Informix Update statements crashing engine

9. IDS 9.21 Sysmaster Query and Engine Crash

10. 7.12 engine crashes on SCO Openserver 5

11. 4GL crashing my Online Engine