2

I'm evaluating PlasticSCM on a VMWare Machine with 4GB RAM and 4Core CPU. Since I've ported our trunk into the server (about 6GB of Data), the service ran out of memory (started swapping). I've increased the the VM RAM to 6GB This is actually more than I'd like to load the host system with, since I've also got VMs for PlasticSCM Client, TeamCity Server, TeamCity Agent.

I was trying to find a spec with details on hardware requirement for running PlasticSCM server which incorporates scaling. So far, I've only found the minimum requirement (512MB RAM etc.) and the system information of your heavy load and scale test. As far as I can see, it's all about RAM. :)

Anyway is there a detailed spec with recommendations for the hardware being used?

P.S.: Of course, in case of switching to Plastic we'd run the service on a real machine instead of VM.

Jan
  • 171
  • 1
  • 6
  • Hi DenDev. Look, 4GB of RAM should be more than enough. I've 18GB of repos in my laptop and the process uses like 150-200MB. Are you intensively using under heavy load (tons of users?). Since I guess it is not the case, I can only think of eating too much memory during the "port" which I guess was a gitsync or similar, correct? If so, yes, during import it can use more mem (typically the client, not the server). Restart, it should use less. Otherwise contact our support team and we'll look into it immediately. – pablo May 21 '15 at 16:17
  • TeamCity was updating the workspace. In the meanwhile, I was testing some access stuff with the ACLs. By this the TeamServer got accidentally ripped of the server (access right revoked during update) which might have left the server in an undefined state. At the end, the server was consuming 4,5GB (swapping), SQL was consuming 2GB. Since restarting the service, its running fine with about 200MB RAM, MySQL consuming 2.7GB. Anyway, I would expect the server to handle those operations gracefully. Still, I'd like to have some hardware spec I can provide to our IT. – Jan May 22 '15 at 13:14

0 Answers0