Don't you lose all the HA preferences when you disable HA?
I had the 'shortnames' issue also - which editing the hosts file fixed. But I also had an issue with connecting to my SQL DB after the upgrade. It appears that the VC upgrade wants to use the 'Service Account' for the DB. We have it configured for 'Windows Auth.' and we used our existing DB. Once we verified the ODBC connections and the DB properties both used the same account everything worked.
Hello,
I have HA enabled and the upgrade works just fine. Granted, I do have everything in /etc/hosts that is required.
Best regards,
Edward
the mulptiple short names?
I had an issue with 2 hosts when Configuring HA....the multiple shortnames error.
No, at least the "Isolation response" settings of the VMs are preserved. The others are easy to remember and to restore...
Hi John,
thank you sooooooooooooooooooooooooooooooooooooooooo
and again soooooooooooooooooooooooooooooooooooo
muchhhh
i change the /etc/hosts to lower case and everting was ok
thanksthanksthanksthanksthanksthanksthanksthanks
i think the clean install of 3.0.1 and upgrade that was ok... it's the 3.0.0 that I upgraded to 3.0.1 and then 3.0.1 that caused a problem.
I was more concerned about the Restart priority settings...but it looks like those are preserved per VM so disabling HA is not a big deal.
My settings were preserved as well. (I have a number of vms set to manual restart.)
Hi,
had no luck with update
Got the error "Error 25014. Setup failed to upgrade the vpxd_stats_rollup procedure in the VirtualCenter database." We have a Oracle-Database.
Search this forum, but found no solution.
Thx
Hi esxfax, read the dokumentation carefully, there some point with Oracle Database.
Yesterday i updated an Oracle Database too, this was ok.
Grand vpxadmin DBMS_LOCK Permissions
\- grant execute on dbms_lock to vpxadmin;
I upgraded from VC 2.0.1 non-patched directly to 2.0.2 and all went well. HA was turned off previous to the upgrade. I have 9 local hosts and 5 remote hosts and about half of them failed to upgrade the VC client. Restarting the services on these hosts and then reconnecting to them in the VI client fixed everything. Not too painful.
Grand vpxadmin DBMS_LOCK Permissions
- grant execute on dbms_lock to vpxadmin;
Hi,
you must also grant execute on dbms_lock to public...
Execute was granted to vpxadmin, but we also had the error: "Error 25014. Setup failed to upgrade the vpxd_stats_rollup procedure in the VirtualCenter database".
Googling the Oracle error gave the solution to grant execute also to public and say what, the error disappeared...
"identifier 'SYS.DBMS_LOCK' must be declared"
-> You or your DBA must connect to SYS and issue this command:
GRANT EXECUTE ON DBMS_LOCK TO PUBLIC;
Thanks for your help Nautilus!
Update processes now without problems.
CU,
esxfan
i'll add my experience:
environment is vc 2.0.1, seven ESX 3.0.1 hosts (6 enterprise licenses, one std for an AMD box).
i had to create /tmp/vmware-root on 4 of my ESX hosts prior to starting.
I upgraded each component individually.
1) upgraded License Server first (it asks about upgrading if you run that install separately and not as part of VC update).
2) After that, I upgraded VC, choosing to use my existing MSDE instance. 3) Then I ran the VI client update.
smooth as silk.
I too upgraded to VC 2.0.2 from 2.0.1. Then upgraded my six ESX hosts to 3.0.2 from 3.0.1. All went well, except now VC does not show the traffic light statuses for any VM's when viewed in the Virtual Machines tab of VC. The statuses do show in the Hosts tab window. Wondering what's going on to cause this.
just no Problem esxfan, i'am not a database consultant and knows nearly nothing about oracle, but i think some thime its better when we read exactly the documentation ( i'am doing this sometime )
I have seen a problem with VI client not updated the column "uptime" when looking at the hosts. Anyone else seen this?
/Magnus
I had similar issues as described in this thread. I did not turn off HA before doing the update, and had the errors in VC when trying to get the agents to deploy. At which point, all of our hosts were up and running still, but HA would not work at all.
We had 2 errors showing, one is the error about not being able to contact the primary, and the other the default HA error. After spending a few hours working with this, checking hosts files, rechecking all the settings, I finally found a log file with the answer in it. HA was getting confused on what IP address on the box to use. The IP that is in DNS, or 127.0.0.1.
The default installation of ESX has 127.0.0.1 localhost.localdomain localhost in the hosts file. However, when I changed this to:
DNSIP FQDN shortname .. HA configured right away and began working. I am not sure if this should have been changed all along, or if anyone else has had something like this happen, but thought I would share this incase others run into this.
EDIT: I just looked at a brand new install of 3.0.2 and noticed that the host file now has the "127.0.0.1 localhost.localdomain localhost" aswell as "dnsip fqdn" in there. I am not sure if this was in previous versions and ours were edited wrong or not.
Message was edited by:
dduell