Hi. Since I installed a new LSI controller (9305-24i), new SAS cables, additional RAM, and a GPU, I've been getting errors. For example, every morning requires that I reboot the server because it's "unstable" (I can't access the docker panel, I can launch a VM, unassigned drives have gone missing, etc.).

 

I"m not sure what could be the cause. I'm attaching my diagnostics if anyone can provide some help.

tower-diagnostics-20190922-1347.zip

Edited September 22, 20196 yr by Kuddy
attaching diagnostics

If you look at your syslog files, they are filled with errors.  You seem to have a lot of problems with the HD's assigned to sdj, sdk, and sdl.  You also seem to have problems with the mpt3sas driver which (I believe) is used by the LSI controller.  (I never did finish with the syslog because the quantity of errors.)  

 

You should probably be double checking that you have installed and seated the card properly and that all of the connectors are secure.  You should probably be double checking the power feeds to those three HD's.  

 

You should probably also check that the power supply is capable of handling the extra power requirements after adding all that new gear.

 

Did you run a memtst after your installed the new RAM?  In cases like this-- where there is a problem and lots of installed RAM--- you should probably test it for 24 hours.  

There's a constant problem with the HBA, make sure it's well seated, sufficiently cooled and not a fake

 

Sep 22 05:45:07 Tower kernel: mpt3sas_cm0: fault_state(0x6004)!