IBM storewize SSH - new mode vdisk-status #5263
Open
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Community contributors
Description
if a Hyperswap configuration seem's actually not really supported and the returned data are not interpreted in the correct way. Teh components "vdisk" check has not enough filter possibilities and not the possibility to override the critical status with the needed variables. So i created a new vdisk-status mode to handle the status in correct way for all needs.
For the hyperswap configuration from 4 disk 1 disk is always offline:
A hyperswap configuration can be identified when the column RC_name is not empty and has a value like rcrelXX.
The right logic to monitor this configuration should be:
every disk with function master_change or aux_change must be online, otherwise should be a error
one of the disks with function master or aux is allways offline, the other one is online
to indentify which of this 2 disks should be online you need to launch the lsrcrelationship command.
It's much more faster to call it generally once wihout rc_name (lsrcrelationship rcrel95 ) as 100 times with rc_name.
So this case is OK because the offline ones is not the primary vdisk of the RC
default critical status
(%{rc_name} ne "" && %{primary} == 1 || %{rc_name} eq "") && %{status} =~ /offline/i
so this means the "primary" is only checked fore RC's. for single vdsisk only the status will be checked.
The vsidk component check can be removed
Linked PR:
#5106
Type of change
How this pull request can be tested ?
All data are still the same. So use the test since yet.