ViceVersa Pro 2.5 2504 Invalid argument has encountered

[RForSho]

Currently using VVPro 2.5 Bld 2504. today we received an error:
Invalid argument has occurred.
This has never happened before and it is critical that is does not happen again. The Script states that it has completed successfully but in fact has not. When trying to run the comparison, it completes cut cannot execute. It Appears that it think there have been no changes and thinks it is finished successfully.

Server 2003 Standard R2
3GB RAM
1.5 million files (-/+) to compare @ 19.98 GB

I have read a couple similar post that describe a memory issue but do not detail it. i Never see the memory increase above 1.38 GB while VVPro is running. This is all that this server does. Any Ideas?

[TGRMN Software]

Hi , please install the latest ViceVersa PRO 2.5 build (it's free). Is it on 32 bit? Could you split the single profile into two profiles to run sequentially? thanks
_________________
--
TGRMN Software Support
http://www.tgrmn.com
http://www.compareandmerge.com

[RForSho]

- Yes it is on a 32 bis OS.
- Unfortunately we do not have the option to split this job.

We have the latest build of 2.5 (2513) on another 32 bit 2003 server with this exact same script/profile for a back up. It is a bit better as far as processing power and has 4GB RAM. We were able to run a comparisong for the exact same mirror on it sucessfully and then execute successfully as well. But.. the nest time it was scheduled to run it failed only this time it did state that it could not complete and emailed the following notification:

Profile: ArchiveSync
Result: Execution not completed.
Exit Code: 2
[ERROR] Out of memory, too many files in source and target. Split job into multiple profiles [Error Code mC1] Profile Start Time: 4/30/2014 3:55:50 PM Profile End Time: 4/30/2014 4:04:21 PM

Since we cannot split the job we tried to manage the file count down and took it from approx. 1.5 million (19.9GB )to 1.3 million files (17.8GB).
Ran it again and received notification that it was successful. It appears that everything is there.
Since originally it would say completed successfully but actually had not I added a test file to the source this morning and have ran it again to make sure it mirrors.

From this point on we will not be able to manage the count any lower, I hope that at this point it proves to be successful.

[RForSho]

[RForSho]

Update:

It appears the profile ran successfully on the production server (build 2504), the test file mirrored just fine this morning. This was the first run of the production profile since managing down the file count. But as the file count grows we will not always have the option of managing down like that, so that does not appear to be a feasible resolution. Is there nothing more that can be done other than splitting the job or shrinking it?

Like the original post states, we do not see the memory usage go anywhere near the max so i am not clear how it is running out of memory. The only other Application running is Virus protection which runs briefly in the AM for a very short time and even then does not use hardly any additional memory.
Please let me know your thoughts/ideas. I appreciate the time and input. Thanks!

[TGRMN Software]

On Windows 32-bit there's a limit to the amount a memory each process can use (from 2GB to 4GB, depending on configuration): no matter how much physical memory is installed, that is the top logical limit.

On 64-bit there's no such limit (actually is 8 TB in x64 to 128 TB in the latest Windows 8.1 and Windows Server 2012 R2).

I was wondering if you already tried using file filters to split the job as shown here: http://www.tgrmn.com/web/kb/item104.htm

thanks
_________________
--
TGRMN Software Support
http://www.tgrmn.com
http://www.compareandmerge.com

[RForSho]

No, I had not tried to split the job yet since we were able to manage the file #'s down.
It seems to be consistently successful the past couple days. if it recurs I will again revisit the option for splitting the job