Hi guys!
Here’s a case that alarms are generated indicating that disks respond slowly when a loop contains more than 10 disk enclosures.
Symptoms
In the OceanStor V200R001C00 environment,
when a single loop in the storage system contains more than 10 disk enclosures,
the ISM reports alarms, indicating that disks respond slowly.
Since a loop contains a maximum of four disk enclosures in a HUAWEI OceanStor 18000 environment, no alarm is generated.
Alarm Information
The ISM reports the alarm "Slow Disk Response".
Possible Causes
Identifying a slow disk is based on the average service time spent on each I/O operation (the average I/O return time) within a period of time.
When multiple disk enclosures are connected in a loop, disks in the loop take longer time to respond to the I/Os than the average response time.
Therefore, the ISM incorrectly assumes that the disks have slow response, and the alarms are generated.
Identification Method
1.On the ISM or on the CLI, run show alarm to check whether the alarm whose ID is F00A000D is generated.
2.Check the working environment starting from SAS interface modules on the controller enclosures.
More than 10 disk enclosures are connected in the same loop.
Troubleshooting Procedure
1.Stop services if any and reduce the number of disk enclosures to not more than 10. In doing this,
the storage pool degradation or other faults may occur.
2.Connect the disk enclosures to another loop and check whether the storage pool is recovered.
3.Re-enable the services and check whether the services are running correctly.
Check After Recovery
On the ISM or on the CLI, run show alarm to check whether the alarms are cleared, the storage pool faults are rectified, and the services are recovered.
