alarmprogram.sh when LTAPEDEV /dev/null
Posted in 2012
On HP-UX with IDS 11.50.FC6, the poster was flooded (100k+) with messages showing alarmprogram.sh being invoked recursively and exiting with return code 1, despite LTAPEDEV=/dev/null, and assumed log backups were the cause. Replies noted LTAPEDEV=/dev/null doesn't stop the alarm program from firing on log-complete events, and the real trigger was ALRM_ALL_EVENTS set to 1 (0 in production), which calls the script for nearly every online.log line. Suggested checks: script permissions (fine, 755), presence of /bin/ksh, and adding echo/stderr redirection inside the script to debug the return code 1. No confirmed fix is recorded.
Auto-generated by DrWatson from the posts below — may be imperfect; read the full thread.
Topics: Migration, Import/Export & Data Conversion, Platform-Specific Issues, Versions, Editions & End-of-Life
HP-UX wuast130 B.11.23 U 9000/800
IBM Informix Dynamic Server Version 11.50.FC6
hello
does anyone know why this is happening
I thought if LTAPEDEV /dev/null
then it would not try to back up log - I asume this is whats happening >
I am getting thousands of these messages and tasks running - so many running
now I cant do any more command with informix . I am doing a dbimport
-c /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with
return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 1 42
"Event occured" "Process exited with return code 1: /bin/sh /bin/sh -c
/opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with
return
alarmprogram.sh
I don't think that is the logical log backup...
It looks much worse, but I may be wrong.... At first glance it looks like
an event with class ID 42 has happened.
The Admin Ref Guide has this info about it:
"The database server encountered an error
while validating the tablespace page.
Online log: Assertion warning with details
of the table.
Server state: Online.
User action: Examine the online log to
determine which database.owner.tablename
the issue occurred on. Run the oncheck -pt
command on the table. Correct any errors
identified by the oncheck utility and retry
the operation. If the operation fails again,
note all circumstances and contact IBM
Software Support"
Do you have strange messages on the online.log?
On the other hand, severity 1 looks pretty low....
Regards.
On Mon, Aug 13, 2012 at 10:07 PM, KARL OLIVER <karl.oliver@maf.govt.nz>wrote:
> HP-UX wuast130 B.11.23 U 9000/800
>
> IBM Informix Dynamic Server Version 11.50.FC6
>
> hello
> does anyone know why this is happening
> I thought if LTAPEDEV /dev/null
>
> then it would not try to back up log - I asume this is whats happening >
> I am getting thousands of these messages and tasks running - so many
> running
> now I cant do any more command with informix . I am doing a dbimport
>
> -c /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited
> with
> return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 1 42
> "Event occured" "Process exited with return code 1: /bin/sh /bin/sh -c
> /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with
> return
>
> alarmprogram.sh
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
--20cf300e4d7d777cfb04c72c3e38
so far 131267 of these messages 09:24:51 Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 2 23 "" "Logical Log 10222 Complete, timestamp: 0x147eff26." """ """ "" Maybe it is just saying logical log is completed ?
Original post:
HP-UX wuast130 B.11.23 U 9000/800
IBM Informix Dynamic Server Version 11.50.FC6
hello
does anyone know why this is happening
I thought if LTAPEDEV /dev/null
then it would not try to back up log - I asume this is whats happening >
I am getting thousands of these messages and tasks running - so many running
now I cant do any more command with informix . I am doing a dbimport
-c /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with
return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 1 42
"Event occured" "Process exited with return code 1: /bin/sh /bin/sh -c
/opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with
return
alarmprogram.sh
Response:
Have you modified /opt/informix/etc/alarmprogram.sh? From a quick test here,
it looks like even with LTAPEDEV set to /dev/null whenever I switch log files
(which is the same as a logical log completing), it still tries to invoke the
alarmprogram.sh. Are the permissions on $INFORMIXDIR/etc/alarmprogram.sh 755,
as I think if the permissions are messed up you can get those sorts of errors.
Jacques Renaut
IBM Informix Advanced Support
APD Team
permissions are -rwxr-xr-x
Please post the following:
a)
ls -liad /opt /opt/informix /opt/informix/etc
/opt/informix/etc/alarmprogram.sh
b)
head -5 /opt/informix/etc/alarmprogram.sh
c)
onstat -c | grep ALRM
You have two events there. One is logical log complete. The other is event
with class ID 42 which I really never noticed...
Regards.
On Mon, Aug 13, 2012 at 10:26 PM, KARL OLIVER <karl.oliver@maf.govt.nz>wrote:
> so far 131267
> of these messages
> 09:24:51 Process exited with return code 1: /bin/sh /bin/sh -c
> /opt/informix/etc/alarmprogram.sh 1 42 "Event occured" "Process exited with
> return code 1: /bin/sh /bin/sh -c /opt/informix/etc/alarmprogram.sh 1 42
> "Event occured" "Process exited with return code 1: /bin/sh /bin/sh -c
> /opt/informix/etc/alarmprogram.sh 2 23 "" "Logical Log 10222 Complete,
> timestamp: 0x147eff26." """ """ ""
>
> Maybe it is just saying logical log is completed ?
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
--20cf303640b344001d04c72d5fb6
2 dr-xr-xr-x 92 bin bin 8192 Apr 6 2011 /opt
61926 drwxr-xr-x 38 informix informix 8192 Sep 13 2011 /opt/informix
61927 drwxrwxr-x 4 informix informix 8192 Aug 13 09:34 /opt/informix/etc
83027 -rwxr-xr-x 1 informix informix 19932 Aug 4 2010
/opt/informix/etc/alarmprogram.sh
head -5 /opt/informix/etc/alarmprogram.sh
head -30 /opt/informix/etc/alarmprogram.sh
#!/bin/ksh
# /**************************************************************************/
# /* */
# /* Licensed Materials - Property of IBM */
# /* */
# /* "Restricted Materials of IBM" */
# /* */
# /* IBM Informix Dynamic Server */
# /* (c) Copyright IBM Corporation 1996, 2006 All rights reserved. */
# /* */
# /**************************************************************************/
#
# Name : alarmprogram.sh
# Created : May 2002
# Description : Automates many IDS events using event alarms from the
# database server. To install this script, add the following
# line to the ONCONFIG file :
# ALARMPROGRAM <INFORMIXDIR>/etc/alarmprogram.sh
# where <INFORMIXDIR> is replaced by the full value of $INFORMIXDIR
# This script sends email and pages the systems group when necessary.
#
# /**************************************************************************/
# ########################################
#
# PUBLIC SECTION : CONFIGURATION VARIABLES
#
# ########################################
/opt/informix/etc [paqis wuast130 ast130_ecms_dba] onstat -c | grep ALRM
ALRM_ALL_EVENTS 1 # Triggers ALARMPROGRAM for any event occur
I see for above I have ALRM_ALL_EVENTS set to 0 for production
Yes... ALRM_ALL_EVENTS set to 1 will trigger the alarmprogram.sh for each
line (or nearly that) written in the online.log
That's the cause for thos class ID 42 events.
Unfortunately this parameter is not dynamic...
But in any case I still don't know why the calls fail...
Do you have /bin/ksh on that machine?
You could try something like:
1) Add "echo $0 $* >> /tmp/alarmprogram.debug" to the begining of the
alarmprogram.sh script
2) If that works and you get entries in the file you could try something
more elaborated like "exec 2>> /tmp/alarmprogram.err" and see what you get
I may be missing something very obvious, but typically these erros are
simple to track (permissions, /bin/ksh not available etc.)
Regards.
On Mon, Aug 13, 2012 at 11:47 PM, KARL OLIVER <karl.oliver@maf.govt.nz>wrote:
> 2 dr-xr-xr-x 92 bin bin 8192 Apr 6 2011 /opt
> 61926 drwxr-xr-x 38 informix informix 8192 Sep 13 2011 /opt/informix
> 61927 drwxrwxr-x 4 informix informix 8192 Aug 13 09:34 /opt/informix/etc
> 83027 -rwxr-xr-x 1 informix informix 19932 Aug 4 2010
> /opt/informix/etc/alarmprogram.sh
>
> head -5 /opt/informix/etc/alarmprogram.sh
>
> head -30 /opt/informix/etc/alarmprogram.sh
> #!/bin/ksh
> #
> /**************************************************************************/
> # /* */
> # /* Licensed Materials - Property of IBM */
> # /* */
> # /* "Restricted Materials of IBM" */
> # /* */
> # /* IBM Informix Dynamic Server */
> # /* (c) Copyright IBM Corporation 1996, 2006 All rights reserved. */
> # /* */
> #
> /**************************************************************************/
> #
> # Name : alarmprogram.sh
> # Created : May 2002
> # Description : Automates many IDS events using event alarms from the
> # database server. To install this script, add the following
> # line to the ONCONFIG file :
> # ALARMPROGRAM <INFORMIXDIR>/etc/alarmprogram.sh
> # where <INFORMIXDIR> is replaced by the full value of $INFORMIXDIR
> # This script sends email and pages the systems group when necessary.
> #
> #
> /**************************************************************************/
>
> # ########################################
> #
> # PUBLIC SECTION : CONFIGURATION VARIABLES
> #
> # ########################################
>
> /opt/informix/etc [paqis wuast130 ast130_ecms_dba] onstat -c | grep ALRM
> ALRM_ALL_EVENTS 1 # Triggers ALARMPROGRAM for any event occur>
> I see for above I have ALRM_ALL_EVENTS set to 0 for production
>
>
>
>
*******************************************************************************
> Forum Note: Use "Reply" to post a response in the discussion forum.
>
>
--
Fernando Nunes
Portugal
http://informix-technology.blogspot.com
My email works... but I don't check it frequently...
--20cf3066815765544704c72dbc3d