Atom issue? Unfortunately we have detected a kernel crash (panic)



  • After years of rock solid service (never a crash), my pfSense box crashed today.

    On coming up the logs suggest it's hardware related with NMI indicates hardware failure. Machine has an Intel C2558 (which famously have a CPU bug which causes them to fail eventually). Wonderning if anything in the logs can suggest if the CPU is on its way out?

    Attached ddb.txt.

    ddb.txt


  • Rebel Alliance Netgate Administrator

    I don't think it's the CPU that has failed you.

    I would start out with normal troubleshooting; First step I would do is take a backup.

    I would then run a File System Check, seeing if that resolves the issue. If it does not you might need to get a bit deeper, looking at the RAM or the HardDisk to see if there are errors on either.

    Or as it's an older unit (Years of rock solid service), you might just want to look at replacing the unit.


  • Netgate Administrator

    Yeah, that doesn't look like the CPU issue you are referring to.

    Steve



  • Just documenting here. Ran fsck, no errors found. Ran memtest overnight (4 passes), no errors found.

    Machine posts really slowly. Sometimes hangs on 'System initializing F1'.

    After posting it did this once:
    Screenshot 2019-04-14 at 13.09.58.jpg

    Also machine is only 3 years old.


  • Netgate Administrator

    That looks like a hardware issue but it's still processing. It's something different.

    Steve