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

E9000 Power Shortage y Predictive failure asserted alarms in MM1- e8

Publication Date:  2015-12-03 Views:  195 Downloads:  0
Issue Description

The customer reported issues with an alarm that appeared with the next message:

"Power Shortage y Predictive failure asserted alarms in MM1- e8"

 

Alarm Information

The customer sent the next screenshot of the alarm:




Handling Process
  1. 1 Run the smmget -l shelf -d powercappingenable command on the MM910 command-line interface (CLI) to check whether the power capping function is enabled for the chassis. If enabled is displayed in the command output, the power capping function is enabled.

  2. 2 Check whether a new Compute Node or another component (such as a Switch Module or fan module) is installed.

  3. 3 Run the smmget -d listpresent command on the MM910 CLI to query the installed Compute Nodes.

    bladeN indicates a Compute Node name. For example, if the following command output is displayed, the system has installed blade 1 and blade 13.

    root@SMM:/# smmget -d listpresent
    List Present Information:
    system
    shelf
    smm
    blade1
    blade13
    pem
    fantray
    swi1
    swi2
    swi4

  4. 4 Run the smmget -l bladeN -d powercapping command on the MM910 CLI to query the power capping value for each Compute Node (bladeN indicates a Compute Node name).

    If the following command output is displayed, the power capping value for the Compute Node 13 is 883 W.

    root@SMM:/# smmget -l blade13 -d powercapping
    blade power capping is:883
    blade power capping is disabled

  5. 5 Run the smmget -l bladeN -d powerreference command on the MM910 CLI to query the power lower threshold for each Compute Node (bladeN indicates a Compute Node name).

    If the following command output is displayed, the power lower threshold for the Compute Node 13 is 66 W.

    root@SMM:/# smmget -l blade13 -d powerreference
    Min Power:66 Watts

  6. 6 Check whether the power capping value of each Compute Node is less than the power lower threshold.

  7. 7 Run the smmset -l bladeN -d powercapping -v value command on the MM910 CLI to change each Compute Node's power capping value that is less than the power lower threshold. (bladeN indicates a Compute Node name and value indicates the power capping value for a Compute Node, which must be greater than the power lower threshold.) Then check whether the alarm is cleared.

    • If yes, no further action is required.
    • If no, go to Step 8.

  8. 8 Run the smmset -l shelf -d powercapping -v value command on the MM910 CLI to change the power capping value for the chassis. Then check whether the alarm is cleared.

    • If yes, no further action is required.
    • If no, go to Step 9.

  9. 9 Run the smmget -l bladeN -t fru —d all command on the MM910 CLI to check whether the Compute Node CH242 exists.

  10. 10 Run the smmset -l shelf -d powercappingenable —v disable command on the MM910 CLI. Then check whether the alarm is cleared.

    • If yes, no further action is required.
    • If no, go to Step 11.

  11. 11 Contact Huawei technical support for help.

Clearing

This alarm is cleared when the sensor detects that the power capping value of the Compute Node is higher than the minor alarm threshold.

After the fault is rectified, the system automatically clears the alarm.

Root Cause

The customer when the power capping value was configured he had 6 blade servers CH240, after that he modified the E9000 structure and remove 5 blade servers and inserted 4 blade servers CH121 and just kept one CH240, the power capping value never was configured after the modification then the alarm appeared.

Solution

 

The values in the slots 2, 13 and 14 of the device E9000 had a bad configuration:

After the customer applied the command that I suggested, the alarm was cleared.

 



Suggestions
Every time that the customer add new servers or remove, he needs to check in the configuration the power capping value neded depending of his needs in order to verify that will be working fine going further.

END