No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

Reminder

To have a better experience, please upgrade your IE browser.

upgrade

Single Path or Failure to Access for a New Disk That Replaces a Slow Disk

Publication Date:  2013-07-04 Views:  90 Downloads:  3
Issue Description
For T series V100R002SPC012 or earlier versions, a slow disk is replaced by a new disk after a slow response alarm is reported. There is a probability that the new disk can be recognized by only one controller or cannot be accessed.
The replacement disk can be recognized by only one controller or cannot be accessed.
Alarm Information
None
Handling Process
1. On the CLI, run the showupgradepkginfo -t 1 command to view the system software version. V100R002C00SPC010 and later versions are security requirement versions, and others are non-security requirement versions.
2. For security requirement versions, remove and reinsert the single-path disk at an interval of 30s. After the disk indicator becomes steady on, run the showdisk –physic command to view the disk status. It is NOMAL. Then, run the showdisk –logic command. The disk is in reconstruction. The disk is recovered.
3. For non-security requirement versions, go to the debug mode (Contact technical support engineers to obtain the dynamic password required for logging in to the debug mode. For V100R002, enter developer and the obtained password on the CLI. Then, enter debug. For V100R001, enter debug and the obtained password). Run mml to go to the mml mode. Run the dev diskinfo command to view the disk status. If the disk status is link[1,0], the disk can be recognized by only controller A. If the disk status is link[0,1], the disk can be recognized by only controller B.
4. If the disk can be recognized by only controller A, log in to controller A's mml mode (the same for controller B). Run the dev diskinset enclosureid slotid command, where enclosureid refers to the enclosure ID, and slotid refers to the slot ID.
5. Run the command dev diskinfo to view the disk status. link[0,0] indicates the disk status is normal.

The ISM shows that the single path status of the disk is recovered. Or on the CLI, run the showdisk –physic command. The disk status is NORMAL (not SINGLEPATH).
Root Cause
When a slow disk is replaced, one disk controller is OUT and first processed. The other controller is GOOD and has the slow disk mark, triggering the slow disk processing process. Therefore, the disk PHY is closed. However, the driver enables disk PHY when the disk is removed. If the operation of closing the disk PHY occurs after the driver enabling it, the newly inserted disk can be recognized by only one controller or cannot be accessed.

1. Decompress the log file and find the event folder. Open it and view the event information as shown in the following figure. Confirm that the disk has slow response.

2. View the SES log information for the command output of disable disk phy.

The command output shows that the disk PHY is closed when the disk is removed.
3. View the messages log to determine whether the slow disk processing is implemented.

The log shows that the slow disk processing is implemented when the disk is removed. The RAID mechanism closes the disk PHY.
Suggestions
None

END