Repositories Crashing - Patch 2

[tkeith]

I don't remember having such issue previously but it seems in the last week or so we are seeing random lockups on some of our repositories killing jobs or preventing them from even starting. Thought i would see if anyone else has been seeing this issue lately.
The only recent changes we had made was upgrading to Patch 2. We have a pretty large environment and in the last week approximately 3 or 4 Repositories have locked up to the point a reset in vCenter is required.

Repo OS: Linux RedHat 6 (64bit)
Storage: NFS mounted volume
Concurrent Connections: 12
CPU: 2 vCPU
RAM: 8GB
Case # 00933631

[Gostev]

Hi, Keith.

As per the What's New document, Update 2 has introduced a 64-bit Linux data mover, so this would explain any potential difference in behavior after upgrading.

Looking at your system spec, your backup repository server clearly does not meet the minimum System Requirements on the amount of RAM. In fact, 8GB was way below minimum requirements for 12 concurrent jobs even for legacy 32-bit data mover, and 64-bit data mover consumes twice as much memory due to additional performance optimizations we were able to add (which use RAM cache), thanks to the fact that 64-bit processes can address much more memory.

If increasing RAM is not an option, then a quick and dirty solution in your case would be to reduce the amount of concurrent tasks from 12 to 4 or so.

Thanks!