ESX host keep on disconnecting from VC

I had a problem where randomly mu hosts would disconnect from the VC.  This was really worrying as DRS and HA would try and kick in a move all the guests to another host but then all of a sudden the host would reconnect.

After I had recovered from my heart attach I discovered that the service console was low on memory.  To find this out log on the host and type ‘free’.  This should tell you how much memory is used.  Look out for ‘swap’ and then ‘used’.  If your used swap figure is similar to your total then you need to make a change.  I’ve been told that 80mb is fine and a guide line but if it reports considerably higher then you need to change it.

Log in the client and click the configuration tab of the affected server.  Then from the left select ‘Memory’.  You should then see the service console memory allocation which could be “272MB’ which is the default.  I have upped mine to 800mb which I believe is the maximum (ESX3.5).  Click on ‘Properties’ on the top right and update appropriately.  Once you have done this you need to reboot your host.

I hope this helps!

ESX health check e-mail direct from host

This is a useful script I have come across which sits on the ESX host.  It send you an e-mail with an attached HTML report file.  It’s well detailed and provides useful information however on the downside, you have to manage it on a per host basis.  Maybe there’s some more centralised Powershell scripts that run again VC!

http://sourceforge.net/projects/esxhealthscript/