Change Details

xenon.eqiad.wmnet became unresponsive this afternoon, and was found to be under extremely high CPU utilization. The culprit was a number of kernel acpi_pad processes. {P3608} Google turned up no shortage of hits for this, most of which seemed to indicate this could happen after disabling hyperthreading (which doesn't seem to be applicable here). Several recommended removing the `acpi_pad` kernel module, which I did (ephemerally, ala `rmmod acpi_pad`), which seems to have done the trick (the load is now normal).

xenon.eqiad.wmnet became unresponsive this afternoon, and was found to be under extremely high CPU utilization. The culprit was a number of kernel acpi_pad processes. {P3608} After a reboot, the machine came back up in the same state, 11 `acpi_pad` processes saturating the CPU. Google turned up no shortage of hits for this, most of which seemed to indicate this could happen after disabling hyperthreading (which doesn't seem to be applicable here). Several recommended removing the `acpi_pad` kernel module, which I did (ephemerally, ala `rmmod acpi_pad`), which seems to have done the trick (the load is now normal).

xenon.eqiad.wmnet became unresponsive this afternoon, and was found to be under extremely high CPU utilization. The culprit was a number of kernel acpi_pad processes. {P3608} After a reboot, the machine came back up in the same state, 11 `acpi_pad` processes saturating the CPU. Google turned up no shortage of hits for this, most of which seemed to indicate this could happen after disabling hyperthreading (which doesn't seem to be applicable here). Several recommended removing the `acpi_pad` kernel module, which I did (ephemerally, ala `rmmod acpi_pad`), which seems to have done the trick (the load is now normal).