I suddenly gotten a read error on the parity drive which disabled the parity disk. I know that there is nothing wrong with the parity disk as I only completed a parity test successfully yesterday evening and another one a week before. However now it appears that the SAS driver for the LSI controller is having issues and therefore is not able to shutdown the server. I am trying to download the diagnostic file but that was so far not possible. I have the syslog. I am currently not able to resolve this, any suggestions to fix this are highly appreciated.

tower-syslog-20190731-1836.zip

after about 30min of me asking the server to shutdown it finally did, without me doing anything.

It looks like the system entered sleep mode and upon restore the 11:0:*:0 devices went into a repetitive rescan of the disk information.

 

I'm only guessing here but was the sleep intended? It seems like the controller did not successfully recover from the wakeup. Does the controller have it's own power cord? It might be loose or the card itself loose.

I did put the array into sleep mode and pressed a keyboard button. The LSI raid controllers do not have power cords itself. So does this mean I should not use sleep mode? This is a new motherboard and with my old board I was not able to use sleep mode, but I do not want to run into this problem continuesly. Also how can I recover from this, will I have to redo the parity check again? Is the only way to avoid this to not use sleep mode with LSI controllers?

You should update the HBA's firmware, it's on an ancient release.