Operating System of a Storage Node Is Abnormally Reset
The vmcore file used to locate the cause of the abnormal reset cannot be found.<?xml:namespace prefix = "o" ns = "urn:schemas-microsoft-com:office:office" />
Symptom
After the operating system of a storage node is abnormally reset, the vmcore file used to
locate the cause of the abnormal reset cannot be found in the /corefile directory of the
operating system disk. In the serial log downloaded from the iBMC, there are slow disks (in
the log line that contains [sda] Attached SCSI disk, the number in the first brackets is larger
than 60).
Possible Causes
Possible cause 1: The kdump service is not enabled.
Possible cause 2: The operating system disk is damaged.
Possible cause 3: The kdump service cannot be provided after the system resets due to a
hardware fault.
Procedure
Possible cause 1: The kdump service is not enabled.
a. Log in to the operating system of the faulty node. Run rckdump status to view
whether the status of the kdump service is running.
If yes, go to possible cause 2.
If no, run rckdump start to start the kdump service.
b. Check whether the vmcore file can be found after the storage node is abnormally
reset again.
If yes, no further action is required.
If no, go to possible cause 2.
Possible cause 2: The operating system disk is damaged.
a. Use the Mgmt port to log in to the iBMC remotely. Choose Diagnosis > Serial Port
Data > Download. In the downloaded log file, search for [sda] Attached SCSI
disk. In the log line, check whether the first number in the first square brackets is
larger than 100.
If yes, the operating system disk is faulty. In this case, replace the faulty disk
by referring to the Parts Replacement document of the relevant version.
If no, go to possible cause 3.
b. Check whether the vmcore file can be found after the storage node is abnormally
reset again.
If yes, no further action is required.
If no, go to possible cause 3.
Possible cause 3: The kdump service cannot be provided after the system resets due to a
hardware fault.
a. Contact technical support engineers and provide assistance in troubleshooting the
hardware fault.
b. Check whether the vmcore file can be found after the storage node is abnormally
reset again.
If yes, no further action is required.
If no, contact technical support.
----End
