We are currently running 24x7 2.3.6 Version and appear to be getting "memory leaks" which can stop 24 x 7 from running all together.

In order for this not to occur, we shut down the product every 2-3 days or so. Once we shut down and bring it back up, all is ok again for another 2-3 days.

Any suggestions?

P.S. Other than this....I love the product.

1. Enable "job execution statistics" in the 24x7 system options.
2. Let it run for a day.
3. Check the statistics in stat.log file and find out which job(s) is causing the problem.
4. Let us know more about this job so we can start helping you.

: We are currently running 24x7 2.3.6 Version and appear to be getting
: "memory leaks" which can stop 24 x 7 from running all together.

: In order for this not to occur, we shut down the product every 2-3 days or
: so. Once we shut down and bring it back up, all is ok again for another
: 2-3 days.

: Any suggestions?

: P.S. Other than this....I love the product.

We have been running the job execution statistics and only notice a gradual decrease in memory as each job runs. One job in particular is listed in this log a lot (it runs every 15 min). Here is the plan....

1) I've restarted 24x7 this a.m.
2) Enabled Job execution statistics
3) Will collect data on all jobs until noon
4) The job which runs every 15 minutes will be turned off at noon.
5) I'll review the job log at 5:00 pm today and let you know my findings.

If you have other ideas....let me know. Thanks.

: 1. Enable "job execution statistics" in the 24x7 system options.
: 2. Let it run for a day.
: 3. Check the statistics in stat.log file and find out which job(s) is causing
: the problem.
: 4. Let us know more about this job so we can start helping you.

1. You should disable logging for that job
2. If this is an async job, make sure the job ends before another instance is started.
3. Make sure the job releases all resources that it allocates, this especially applies to to all external programs that it starts. Note that these programs share 24x7 memory space.

4. Because you now know which job is causing the memory problem, you have a workaround for forced "clean" job exit. Make this job schedule as [no schedule] and create another job that will run instead. That job can have [run program] type with the command line "24x7 /JOB [first job id or name here]"

: We have been running the job execution statistics and only notice a gradual
: decrease in memory as each job runs. One job in particular is listed in
: this log a lot (it runs every 15 min). Here is the plan....

: 1) I've restarted 24x7 this a.m.
: 2) Enabled Job execution statistics
: 3) Will collect data on all jobs until noon
: 4) The job which runs every 15 minutes will be turned off at noon.
: 5) I'll review the job log at 5:00 pm today and let you know my findings.

: If you have other ideas....let me know. Thanks.

Regarding:

"4. Because you now know which job is causing the memory problem, you have a : workaround for forced "clean" job exit. Make this job schedule
as [no schedule] and create another job that will run instead. That job
can have [run program] type with the command line "24x7 /JOB [first
job id or name here]" "

This has worked perfectly for us and is an excellent solution. We have 12 jobs that work as "job callers", loading another 24x7 instance.

Caution, tho, if you have enabled the HTML Status Update feature. It will try to update every time 24x7 loads and cause all sorts of heartache.

I'm not quite sure WHAT job is causing this.....let me give you a few examples of my log so you can see what is happening. (This is a consecutive section of the stat.log) And....we rebooted 24x7 right before job no. 83

Just so you know.....my Free Resource: starts at around 50%....and gradually decreases. But if I'm reading these stats correctly not one of these jobs is causing the leak....it seems to be something external.

Am I reading these correctly? Can you interpret?

Thanks!

------------------------------------------------------------------------------------------------------------
Job No: 81 Job Name: Bek_Wells_Ack Process ID: 502
Creation Time: 25-Oct-2000 9:33:25:0005 Exit Time: 25-Oct-2000 9:33:26:0007
Duration: 1 seconds
Kernel Time: 0:00:00:0000 User Time: 0:00:00:0000
Exit Code: 0

Free Resource: Before: 9.257% After: 9.282%
Memory Load: Before: 0% After: 0%
Available Physical Memory: Before: 2126692352 (2,076,848 Kb) After: 2126622720 (2,076,780 Kb)
Available Virtual Memory: Before: 1411444736 (1,378,364 Kb) After: 1411313664 (1,378,236 Kb)
Available Page File Size: Before: 40443904 (39,496 Kb) After: 40554496 (39,604 Kb)

------------------------------------------------------------------------------------------------------------
Job No: 81 Job Name: Bek_Wells_Ack Process ID: 172
Creation Time: 25-Oct-2000 9:33:26:0007 Exit Time: 25-Oct-2000 9:33:29:0000
Duration: 3 seconds
Kernel Time: 0:00:00:0000 User Time: 0:00:00:0000
Exit Code: 0

Free Resource: Before: 9.251% After: 9.276%
Memory Load: Before: 0% After: 0%
Available Physical Memory: Before: 2127052800 (2,077,200 Kb) After: 2127040512 (2,077,188 Kb)
Available Virtual Memory: Before: 1411248128 (1,378,172 Kb) After: 1411248128 (1,378,172 Kb)
Available Page File Size: Before: 40419328 (39,472 Kb) After: 40525824 (39,576 Kb)

------------------------------------------------------------------------------------------------------------
Job No: 81 Job Name: Bek_Wells_Ack Process ID: 508
Creation Time: 25-Oct-2000 9:33:30:0009 Exit Time: 25-Oct-2000 9:33:31:0000
Duration: 1 seconds
Kernel Time: 0:00:00:0000 User Time: 0:00:00:0000
Exit Code: 0

Free Resource: Before: 9.215% After: 9.242%
Memory Load: Before: 0% After: 0%
Available Physical Memory: Before: 2126880768 (2,077,032 Kb) After: 2126925824 (2,077,076 Kb)
Available Virtual Memory: Before: 1409413120 (1,376,380 Kb) After: 1409413120 (1,376,380 Kb)
Available Page File Size: Before: 40259584 (39,316 Kb) After: 40378368 (39,432 Kb)

------------------------------------------------------------------------------------------------------------
Job No: 82 Job Name: Bek_Wells_Outbound_Orders Process ID: 486
Creation Time: 25-Oct-2000 10:57:03:0008 Exit Time: 25-Oct-2000 10:57:05:0007
Duration: 2 seconds
Kernel Time: 0:00:00:0000 User Time: 0:00:00:0000
Exit Code: 0

Free Resource: Before: 7.818% After: 7.838%
Memory Load: Before: 0% After: 0%
Available Physical Memory: Before: 2125053952 (2,075,248 Kb) After: 2124722176 (2,074,924 Kb)
Available Virtual Memory: Before: 1392046080 (1,359,420 Kb) After: 1392177152 (1,359,548 Kb)
Available Page File Size: Before: 34156544 (33,356 Kb) After: 34246656 (33,444 Kb)

------------------------------------------------------------------------------------------------------------
Job No: 82 Job Name: Bek_Wells_Outbound_Orders Process ID: 75
Creation Time: 25-Oct-2000 11:19:04:0000 Exit Time: 25-Oct-2000 11:19:05:0009
Duration: 1 seconds
Kernel Time: 0:00:00:0000 User Time: 0:00:00:0000
Exit Code: 0

Free Resource: Before: 7.428% After: 7.449%
Memory Load: Before: 0% After: 0%
Available Physical Memory: Before: 2126422016 (2,076,584 Kb) After: 2126176256 (2,076,344 Kb)
Available Virtual Memory: Before: 1387589632 (1,355,068 Kb) After: 1387589632 (1,355,068 Kb)
Available Page File Size: Before: 32452608 (31,692 Kb) After: 32546816 (31,784 Kb)

------------------------------------------------------------------------------------------------------------
Job No: 83 Job Name: Bek_Wells_Release_Update Process ID: 445
Creation Time: 25-Oct-2000 13:17:29:0002 Exit Time: 25-Oct-2000 13:17:30:0003
Duration: 1 seconds
Kernel Time: 0:00:00:0000 User Time: 0:00:00:0000
Exit Code: 0

Free Resource: Before: 54.649% After: 54.675%
Memory Load: Before: 0% After: 0%
Available Physical Memory: Before: 2146799616 (2,096,484 Kb) After: 2146672640 (2,096,360 Kb)
Available Virtual Memory: Before: 2014457856 (1,967,244 Kb) After: 2014392320 (1,967,180 Kb)
Available Page File Size: Before: 238764032 (233,168 Kb) After: 238878720 (233,280 Kb)

------------------------------------------------------------------------------------------------------------

: 1. You should disable logging for that job
: 2. If this is an async job, make sure the job ends before another instance is
: started.
: 3. Make sure the job releases all resources that it allocates, this
: especially applies to to all external programs that it starts. Note that
: these programs share 24x7 memory space.

: 4. Because you now know which job is causing the memory problem, you have a
: workaround for forced "clean" job exit. Make this job schedule
: as [no schedule] and create another job that will run instead. That job
: can have [run program] type with the command line "24x7 /JOB [first
: job id or name here]"

It is difficult to find out trends and dependencies from this small portion of the log, but a few issues that can be highlighted.

First of all, job 81 started when the machine had only 9% of free resources (the machine was loaded heavily at that time, with little free space left in the page file.). This is bad sign.

The job worked just 1 second (or was it an async. program launch?) then another instance started in 1 second and ran during 3 seconds. Then the third instance started again in the next 2 seconds.
How can it be that a job restarts within 1-2 seconds interval when the 24x7 has 1 minute time increments?

Because of the very small log subset provided for analysis, it is impossible to say for sure what's wrong, but it seems to me that after every run of job 81 and job 82, the free space in the page file decreases, but again, this could be just a coincident as there other programs running on the same computer at the same time.

After you rebooted the computer, it had only 54% free resources, This is a bad sign (not 95%, not 90%, not even 80%).

Have you checked the page file settings whether they match what NT is recomending foryou system?
Have you checked the free disk space on the page file drive? Did you cleaned up temp files from the %TEMP% directory?
Do you run an anti-virus program or disk caching program?
What else are you running on that system?

: I'm not quite sure WHAT job is causing this.....let me give you a few
: examples of my log so you can see what is happening. (This is a
: consecutive section of the stat.log) And....we rebooted 24x7 right before
: job no. 83

: Just so you know.....my Free Resource: starts at around 50%....and gradually
: decreases. But if I'm reading these stats correctly not one of these jobs
: is causing the leak....it seems to be something external.

: Am I reading these correctly? Can you interpret?

: Thanks!