Update Manager : There are errors during the scan operation

[AndyShine]

VC 2.5 (virtual machine), 2 x ESX 3.5 in a cluster

I've just installed Update Manger, a vanila install, and setup a schedule to download ESX patches only. I attached the 'Critical Host updates' baseline to my cluster. I then scanned one of the ESX hosts. The scan showed the critical updates need to be applied so I remediated. The patches installed , the server rebooted and everything was fine.

I then attempted to repeat this for the second ESX host. When I trigger the scan i can see the task open the firewall ports but the 'Scan Entity' reports "There are errors during the scan operation'. If i launch the scan from the cluster level Scan Entity reports 'Vmware update Manager had a failure'. In both cases the firewall ports are closed again after the error.

I imagine this is a configuration problem with the second ESX host but can't see anything.

Any suggestions?

Thanks in advance

Andrew

More information:

I have a second cluster with 3 ESX hosts. I attached the 'Critical updates' baseline and scanned the esx hosts without a problem. I haven't yet tried to remediate since this my production cluster, however I'm pretty sure its going to work (yes I'm an optimist). Anyway I think this further suggests that there is a configuration problem with the second ESX host in my other cluster.

Can anyone make a suggestion on what the problem is?

Thanks again

Message was edited by: AndyShine

[MobiusJB]

I am seeing the same issue, I've made a post on this problem as well. Same with you, my issue came about when trying to update my 2nd ESX host, the first one updated without any issue at all. In my situation both installs were exactly the same other than the host name and IP address of the service console, so I'm really confused as I do not believe anything was configured incorrectly.

Sorry, not really an answer for you, just wanted to bump your post, maybe someone else has seen this?

[RParker]

Did you try removing the baseline, and readding and then do a rescan?

[AndyShine]

I've managed to fix my problem. It turns out that the ESX host that was not updating had a bad build (a faulty upgrade from 3.0.2). I re-ran the upgrade and all is well.

(I spotted the bad build issue by running 'esxupdate -i query' : at the end of the output there was a bunch of messages about missing or wrong version packages)

Good Luck

Andrew

[MobiusJB]

I guess your post hasn't show up yet Andy, but I saw you said in an email notification that your re-ran the upgrade and that ended up resolving it. For my scenario I ended up just removing the host from VC which removed the VC agent from the (vpxa I believe it is) and then I re-added the host which of course re-added the agent. Afterwords, attachine the baselines and doing the updates worked for me. Thanks for getting back on this.

[mforbes]

Thanks Mobius,

I was running into a similar issue. I removed the host, re-introducted the host, re-introduced the baseline and re-scanned fine, then was able to remediate as expected.

Mike

[bgardner2001]

Same thing happened to me in Vsphere 4. After updates on first host had the 2nd give "There are errors during the scan operation". I updated VC 4 to update 2 and error went away on that host. Until I tried to update my 3rd host (which was previously working before the VC update) and I did a disconnect/re-connect of my 3rd host and was then able to scan/stage/remediate all 3 of my hosts again. So for me remove/re-add was not necessary in VC4.

[ph0bia]

Same problem here - of 22 existing 4.1 hosts, one would not remediate with no human-readable clue as to why in the log. Very disappointed in the logs, they are pretty much worthless.

I then added four brand new 4.1 hosts to a new cluster, in a new datacenter and they too will scan and stage but not remediate.

I tried all the tricks I've read about in these forums and others including rebooting Host & VCS, removing all affected Hosts from VCS and re-introducing them then re-scanning, staging and remediating. I have verified that the security profile is permitting the traffic, even temprarily disabled the firewall to make certain it wasn't that.

Have not been able to resolve this and it is mind-bogglingly frustrating because these 26 hosts are identical so there is no good reason why it shouldn't work. I want them consistent but now I have 21 which are patched and 5 which are not...

[ph0bia]

My boss was able to fix this one, although we still really don't understand why it happened. If we choose the 'Disable HA' option at the end of the Remediation dialog, it works. Strangely this wasn't required for the other hosts which patched without incident, and sadly VMware doesn't report this as the problem in the cryptic and unhelpful error message.

While researching I also ran across a posting which suggested that having orphaned VMs can caue this, or other orphaned or inaccessible resources such as disc images. So if you have this problem here is a checklist of things to check/try:

- Check the security profile/firewall on the hostmake sure it is allowing Update Manager

- Check windows firewall on VC, make sure nothing is blocking 80, 9000-9100 between the VCS and hosts

- Check for oprhaned VMs, datastores, ISO images, etc.

- Reboot Host, Reboot VCS

- Choose the Disable HA option in the remediation dialog

- Try removing the host from the VCS and re-adding, then re-scan,re-stage, and remediate