Disk Down and Clear events (0xc40060 and 0xc40061) occur in the same second in Spectrum 9.2

Document ID : KB000019478
Last Modified Date : 14/02/2018
Show Technical Document Details

Description:

Customer may also receive an alarm on a disk being down on a host device and the alarm gets cleared in the same second.

The disk 'E:\' also known as 'E:\ Label: Temp Serial Number <SN>'is no longer available.
This disk was being monitored by model <model name>. Host_Device (name - <model name>).

Solution:

Spectrum is checking the disk status twice within the trig timer interval. First time it got no response from disk and raised the "DISK NO LONGER AVAILABLE" alarm and when it got response second time , it has cleared the alarm. So it is working as per design.

We have a workaround, to avoid generating an alarm ("DISK NO LONGER AVAILABLE") if in case there is a clear event 0xc40061 within 60 seconds.

For this we have to create an EventPair rule for the 0xc40060 event to have some custom event 0xfff00001 to generate alarm "DISK NO LONGER AVAILABLE", if clear event 0xc40061 is not generated within 60 seconds.

  1. Add below entries to <SPECROOT>/custom/Events/EventDisp file.

    0xc40060 E 50 R CA.EventPair, 0xc40061, "0xfff00001 -:-", 600xfff00001 E 50
    A 3,0xc40060

  2. Create EvFormat file Eventfff00001 for custom event 0xfff00001 under <SPECROOT>/custom/Events/EvFormat with below content.

    {d "%w- %d %m-, %Y - %T"} - The disk {S 1} is no longer available. This disk was being monitored by model {S 2}. {t} (name - {m}). -(event [{e}])

  3. Run "Update Event Configuration" from Oneclick (select VNM->Information tab ->SpectroSERVER control -> Run Update Event COnfiguration

  4. Reload EvFormat files from OneClick Administration page. (Click on "Update Event and Alarm files -> click Reload)

Otherwise we can do the same from Event Configuration editor.