VMware Cloud Community
ludz1
Enthusiast
Enthusiast
Jump to solution

VSAN Alerts when performing maintenance

Hi all,

I cant remember in VSAN 6.6 but in 6.2 when you place a host in maintenance mode, and the host reboots you naturally get all the vSAN health alarms, object health, DB object failed etc.

The big red failed (test result) pops up but is there anyone else that thinks this is overkill?, I got a shock when I spotted it thinking something was wrong.

vSphere should know the host is in maintenance mode and maybe should give an amber warning? anyone any thoughts on this?

Cheers

newbie (here)

1 Solution

Accepted Solutions
TheBobkin
Champion
Champion
Jump to solution

Hello ludz1​,

Welcome to Communities. Some useful info on participating here:

https://communities.vmware.com/docs/DOC-12286

"I cant remember in VSAN 6.6 but in 6.2 when you place a host in maintenance mode, and the host reboots you naturally get all the vSAN health alarms, object health, DB object failed etc."

Yes, this is the same in all versions of vSAN/Health check.

"The big red failed (test result) pops up but is there anyone else that thinks this is overkill?, I got a shock when I spotted it thinking something was wrong."

I don't think it is overkill - one should always be aware of the state of their data:

When a host enters MM (decomState: 6 in CMMDS) with 'Ensure Accessibility' selected, all the data-components residing locally on this node are marked as Absent - this means their data is becoming staler and staler the longer they are out of sync with the remaining copy/copies and these components cannot be used until they are resynced withe the data changes that occurred during their absence.

The Health check for 'Data' in this case is informing you that the FTT-1 data is currently FTT-0 which is something I am sure you want to be informed of. If you are concerned you can always look at the drill-down information which will show you why this alert is red e.g. Objects that are 'Reduced Availability With No Rebuild' (as the CLOM timer is still ticking down and rebuild has not been initiated).

"vSphere should know the host is in maintenance mode and maybe should give an amber warning? anyone any thoughts on this?"

Sure it does, but this is vSAN Health check not ESXi-with-no-shared-but-local-storage Health check :smileygrin:

PS: To any Mods reading - What's up with new questions (like this one) appearing as 'This question is Assumed Answered' when no-one has commented?

Bob

View solution in original post

3 Replies
TheBobkin
Champion
Champion
Jump to solution

Hello ludz1​,

Welcome to Communities. Some useful info on participating here:

https://communities.vmware.com/docs/DOC-12286

"I cant remember in VSAN 6.6 but in 6.2 when you place a host in maintenance mode, and the host reboots you naturally get all the vSAN health alarms, object health, DB object failed etc."

Yes, this is the same in all versions of vSAN/Health check.

"The big red failed (test result) pops up but is there anyone else that thinks this is overkill?, I got a shock when I spotted it thinking something was wrong."

I don't think it is overkill - one should always be aware of the state of their data:

When a host enters MM (decomState: 6 in CMMDS) with 'Ensure Accessibility' selected, all the data-components residing locally on this node are marked as Absent - this means their data is becoming staler and staler the longer they are out of sync with the remaining copy/copies and these components cannot be used until they are resynced withe the data changes that occurred during their absence.

The Health check for 'Data' in this case is informing you that the FTT-1 data is currently FTT-0 which is something I am sure you want to be informed of. If you are concerned you can always look at the drill-down information which will show you why this alert is red e.g. Objects that are 'Reduced Availability With No Rebuild' (as the CLOM timer is still ticking down and rebuild has not been initiated).

"vSphere should know the host is in maintenance mode and maybe should give an amber warning? anyone any thoughts on this?"

Sure it does, but this is vSAN Health check not ESXi-with-no-shared-but-local-storage Health check :smileygrin:

PS: To any Mods reading - What's up with new questions (like this one) appearing as 'This question is Assumed Answered' when no-one has commented?

Bob

srodenburg
Expert
Expert
Jump to solution

Bob has it spot on with his statement "I don't think it is overkill - one should always be aware of the state of their data:".

Don't forget that your ESXi Servers are now ALSO your storage-controllers and one being out of action, for what ever reason, impairs the redundancy of some of the data. So it's a big deal.

ludz1
Enthusiast
Enthusiast
Jump to solution

Thanks Bob & Srodenburg,

Yes valid points alright, i just had a split second of a "oh did I do something wrong" to cause the "RED" alert warning, when I'm actually performing maintenance that I knew about but I do take your points.

Cheers guys

Reply
0 Kudos