ESX host deteted during maintenance mode... a lot of issues...

[olegarr]

Hello friends,

I was trying to take out of the cluster one of my ESX host. I click "put in maintenance mode", but it was taking too long (oh, where was my paitience!!!:), I click "cancel" on task bar on this task, but it still was taking long time... So, I decided that if I will reboot my ESX it it will help... I rebooted it, but the task was still running and on my VIC clicked "delete" (I wanted to remove it from VC and add it back after that)...

Now I do not see this ESX under my VC, however the tast ""etering

maintenance mode" is in "in progress status" and my ESX server right now do NOT see iSCSI.

I read that the situation when server in maintenance mode will not see iSCSI pretty usual.

My questions are:

How can I check that server is really in maintenance mode? How can I exit maintenance mode? What is the best steps to add it back to my VC? Should I add it first and after that try to exit maintenance mode, or exit maintenance mode and add after that?

Please help me if you know how to resolve my issue.

Best regards,

Olegarr

[christianZ]

Have you tried to restart the VC server ?

[olegarr]

No... you mean restart Windows machine where VC located? still no?

do you recommend to do so?

But how can I check that my ESX in (or not in) maintenance mode?

thansk

[christianZ]

No... you mean restart Windows machine where VC

located? still no?

do you recommend to do so?

YES

But how can I check that my ESX in (or not in)

maintenance mode?

On Esx in /var/log/vmware/hostd.log search for "MaintenanceMode" -

you will see the task beginning and maybe the finishing.

[olegarr]

Christian, thank you for your help.

If it is still in maintenance mode is there any command that will help me exit it? or should I add host to VC first and only after that try to exit maintenance mode, through VC command?

thanks a lot

[christianZ]

What can you see in hostd.log first ?

[mbrkic]

You can check if the host is in maintenance mode or not from the SC with:

vimsh -n -e hostsvc/runtimeinfo | grep inMaintenanceMode

This will show you:

inMaintenanceMode = false,

or,

inMaintenanceMode = true,

You can also exit mainenance mode from SC with:

vimsh -n -e hostsvc/mainenance_mode_exit

[dedfred]

Hello,

I had a similar issue with Maintenance mode but the SC commands provided are great and I've been able to bring my server out of Maint mode. However, when I select the server in VI I get the below msg:

"HA agent disabled on 'servername' in cluster 'clustername'"

any idea why this is happening.

Thanks for you assistance in advance.

Regards

[dedfred]

OK...I think I solved the issue. By choosing the disconnect and then reconnect options on the affected host HA configuration came good during the reconnect process.

However, if there's a VM Guru who can explain this process to me it would be most appreciated.

Is this something that needs to be done each time a host is taken in Maintenance mode or is it some bug?

What does the diconnect and reconnect do?

Thanks again

[Jae_Ellers]

VC, ESX, and your VC client get confused. I find the best thing to do is this:

1. On the esx host: service mgmt-vmware restart

2. On the VC server: restart the VMware VirtualCenter Server service

3. Kill & restart the client.

sometimes it takes 2x #1, but this usually shakes loose any confusion along the VC chain.

[dedfred]

That is kind of annoying isnt it. Is this is bug that VMWare a looking at fixing or this is just the way things are going to be?