August 12, 2025Aug 12 I've had this issue for weeks now, whenever I start a VM that requires a lot of RAM, my shares cease to exist and the only fix is to reboot the entire server every single time which is once a day. I have run some memory tests via the Live Memory Tester Plugin but no issues seem to be logged when completing the test with the array spun down and using all except 500mb of RAM. I'm thinking about getting a new set of RAM to see if that does the trick, but I'm not so sure that it will help. Can anyone recommend anything else to test and help nail down what is going on? unraid-diagnostics-20250812-1243.zip
August 13, 2025Aug 13 Community Expert Server is running out of RAM and shfs was the processe killed:Aug 12 06:00:41 Unraid kernel: Out of memory: Killed process 14365 (shfs)
August 13, 2025Aug 13 Author This happens whether i give the VM 16G or 20G of RAM. The least available RAM I leave free is 2-3G for anything else that needs it.
August 15, 2025Aug 15 Author On 8/13/2025 at 1:53 AM, itimpi said:How much RAM are you trying to give to the VM?16-20G for this particular VM
August 15, 2025Aug 15 Author On 8/13/2025 at 1:48 AM, JorgeB said:Server is running out of RAM and shfs was the processe killed:Aug 12 06:00:41 Unraid kernel: Out of memory: Killed process 14365 (shfs)Why was the process killed for it if there's a few gigs of RAM still available?
August 16, 2025Aug 16 Community Expert Since it was killed, it suggests not enough RAM was actually available. Note that KVM has some overhead; retest with less RAM assigned to the VM.
August 16, 2025Aug 16 Community Expert 2 hours ago, dtanton said:This happens whether i give the VM 16G or 20G of RAM. The least available RAM I leave free is 2-3G for anything else that needs it.OK. You might try giving it much less (i.e something like 8GB) to see if that avoids the issue. I would suggest leaving something like 4GB for the base OS, plus more if you run docker containers.It could also be worth installing the swapfile plugin as that can help if memory fragmentation is aggravating the issue.
August 16, 2025Aug 16 Author 5 hours ago, JorgeB said:Since it was killed, it suggests not enough RAM was actually available. Note that KVM has some overhead; retest with less RAM assigned to the VM.I have done this by testing with 16 instead of 20. Same result.
August 17, 2025Aug 17 Author Solution On 8/16/2025 at 3:23 AM, itimpi said:OK. You might try giving it much less (i.e something like 8GB) to see if that avoids the issue. I would suggest leaving something like 4GB for the base OS, plus more if you run docker containers.It could also be worth installing the swapfile plugin as that can help if memory fragmentation is aggravating the issue.testing with 16GB leaves more than 4GB for anything else. I did not test with 8GB for the VM, I downgraded to 6.12.14 and the issue is resolved. I guess 7.x.x isn't ready for stable production.8 hours ago, JorgeB said:Try with 8GB and post new diags if it still fails.testing with 16GB leaves more than 4GB for anything else. I did not test with 8GB for the VM, I downgraded to 6.12.14 and the issue is resolved. I guess 7.x.x isn't ready for stable production.
August 17, 2025Aug 17 Community Expert 1 hour ago, dtanton said:testing with 16GB leaves more than 4GB for anything else.The OS uses probably about 2, whatever other services you run will use some, and the total used shouldn't exceed about 80% of total capacity or apps and the kernel may start to take drastic measures to save RAM / kill things.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.