itagsurya
Contributor
Contributor

ESX 3i 3.5 u3 server unresponsive from Virtual center

Hi

We have esx 3i 3.5 u3 server. Suddenly uresponsive from virtual center, but server is up and running . vm's also fine. how to solve the issue

Regards

Surya

if you found this or other information useful, please consider awarding points for "Correct" or "Helpful"
0 Kudos
13 Replies
Dave_Mishchenko
Immortal
Immortal

Did you try to restart the management agents at the console DCUI (i.e. Press F2, login and then select to restart the management agents)?

0 Kudos
itagsurya
Contributor
Contributor

hi

I have restart the the agents still iam facing same problem.

Thanks

Surya

if you found this or other information useful, please consider awarding points for "Correct" or "Helpful"
0 Kudos
Dave_Mishchenko
Immortal
Immortal

Are you able to connect directly to the host with the VI client?

0 Kudos
itagsurya
Contributor
Contributor

Hi

I am not able to connect to esx server through VI client and Rcli .

if you found this or other information useful, please consider awarding points for "Correct" or "Helpful"
0 Kudos
bjmoore
Enthusiast
Enthusiast

I was in the same boat yesterday with 4 of our ESXi U2 boxes. They all were disconnected, couldn't manage them directly via the VI client. I restarted network and management agents multiple times with no luck. Ultimately I had to power down and re-register all of the VMs on that cluster and reboot the hosts manually. What type of hardware are you using? Definitely interested in figuring out what causes this problem.

0 Kudos
Dave_Mishchenko
Immortal
Immortal

What sort of hardware are you running?

0 Kudos
itagsurya
Contributor
Contributor

Hi

my hardware is Bl20p g3 and BL 460C g1 servers. finally we shut down the all the VM's and we rebooted the esx server. now its working fine.

Regards

Surya

if you found this or other information useful, please consider awarding points for "Correct" or "Helpful"
0 Kudos
bjmoore
Enthusiast
Enthusiast

IBM HS21 XM blades, onboard solid state drives for the ESXi installation, fiber attached storage via a Qlogic HBA, 32GB RAM and dual quad core 5450s.

0 Kudos
jparnell
Hot Shot
Hot Shot

I've also got this problem on one of our ESXi servers - ESXi 3.5 U3 Build 130755.

Tried restarting the agent from the console but no luck. I fear the only option is going to be to restart the server. We used to have this problem a lot with ESX but i had hoped ESXi would be more reliable.

0 Kudos
michelb
Enthusiast
Enthusiast

As weird as it sounds...

Had a similar problem. Found it was a Windows 2003 server running as a VM that was locking ESXi up on a regular basis meaning a hard-boot of the ESXi box.

As soon as I had shut down that vm, and left it off, ESXi was fine again.

Ended up reinstalling and reconfiguring the W2K3 server.

Don't know how this happened as the HAL should prevent this. Maybe mine was just an oddity.

0 Kudos
jparnell
Hot Shot
Hot Shot

I'm pretty sure I've identified what's causing this. If you have a read

of this site

http://blog.scottlowe.org/2008/12/23/random-reboots-with-vmware-esx-35-u

pdate-3/ you'll see a post from Mike Karolow. Basically it's a problem

with HA logging on the ESXi server. I've tried running the command he

suggests and it returns 0 free inodes (my logs also show 'not enough

space on the device' error messages when trying to restart the

management agents).

Interestingly VMware have just released vCenter 2.5 U4. One fix in this

is:

"ESX Server 3i Installations Run Out of Inodes and Stop Responding to

the VirtualCenter Server

In previous releases, when an ESX Server 3i host in an HA-enabled

cluster is isolated from the VirtualCenter network and reconnected back

to the VirtualCenter network, after a while no free inodes might be

available for the ESX Server 3i and the ESX Server 3i host might stop

responding to the VirtualCenter Server and the VI Client.

This issue is resolved"

I've just got to decide whether to disable HA for the moment or install

a less than one day old upgrade to vCenter.

Hope this helps.

James

0 Kudos
amgenung23
Contributor
Contributor

What work for me was turning off HA.

Confugring the cluster options with the das.failuredetectiontime = 60000

Then re-enabling HA on the cluster.

0 Kudos
mattg_navistar
Contributor
Contributor

I just ran into this issue for the 3rd time.

I found there are a lot of aam log files in /var/log/vmware/aam/rule

use the following command to clear them all out (rm *.log doesn't work)

for i in 1 2 3 4 5 6 7 8 9; do rm /var/log/vmware/aam/rule/$i*.log ; done; services.sh restart

This has worked for me many times now.

0 Kudos