i can only guess.
Are the cards overclocked? Do you use the auto-fan feature of the miner?
FyI: the codebase of vertminer is nearly the same as with cgminer-3.7.2 but the opencl kernel is tweaked -> the cards are pushed harder
My 7950s run since > 1 week without any problem (only yesterday i had to restart because here we had a 2 second power loss)
Not overclocked much, stock mem on the 7950 (1250MHz) and 1020 MHz engine. They are undervolted to 1.081V, but had no issues at all on these settings using normal cgminer. No auto-fan, 85% fixed fan, 60 degrees. Note I had the same issue on R9 280X, different rig (different PSU brand, different motherboard etc). I've been running those 280X without touching it and without any problems on cgminer.
Hm, the settings look safe. My 7950s are at 1000/1500 with 1.012V (at 80°C with BIOS Fan control)
The only relation to the 0 RPM i can see is with the fixed Fan.
Did this happen also before with vertminer < 0.5.2? Or only now with vertminer-0.5.3?
I was on vertminer 0.5.3 from the start. Also it is not related to fixed fan because my 280X had fan set to something like 60-90 and they had the same problem.
I did now remember it could have coincided with some power fluctuations (we had rain and subzero temps, a lot of problems with powerlines), however in the settings topic there were several other people who also reported having the same problem, so I don't think that's it.
Also for some reason I am getting a small amount of HW errors when running 280X on intensity 13, but none at 12. Not sure what I could try to get those to zero, because I am not comfortable running with those errors as I am afraid it could have something to do with that problem, and don't wanna end up with broken cards. I tried lowering TC under 8k but that only increased HW errors, I tried raising it to around 10k but did not help at all.