Maybe I'm thinking incorrectly about this but shouldn't I just be able to:
Download all 'Security' patches
Place them in an http depot
Query them via an ESX 3.5 Update 1 host
Have it be smart enough to install all 'missing' patches?
My experience to date has been:
Downloaded all 3.5 Security Patches
Placed them in an http depot
Was able to successfully query from an ESX 3.5 U1 host via 'esxupdate -d -b "*SG" scan (and --test update)
Test update informed me that I was missing dependency patches (which were actually categorized as 'General') so I d/l'd and added those to the http depot and re-ran --test update.
I was then informed that things could not progress due to already installed patches being there and obsolete patches being present (according to the Update .pdf esxupdate is supposed to be smart enough to ignore these at patch time, no??)
I removed the already installed / obsolete patches from the http depot (ran --test update and everything looked good) then actually attempted to Update for real.
Now all I get is:
# esxupdate -d -b "*SG" --test update
INFO: Configuring...
ERROR: Bundle ESX350-200802304-SG is already installed. Nothing for esxupdate to do.
Why isn't esxupdate 'smart enough' to move on past this vs. returning to shell?
The more time I spend with this the lower my IQ becomes unfortunately.
Thanks for the help!!
Check out esx-autopatch.pl at http://vmprofessional.com/index.php?content=utilities
It takes a little updating to get the patches listed in order along with md5sums.
I use this script at the end of my scripted build process since 3.5 U3 doesn't work on my IBM HS20 blades.
Anyone having similar pain?
Am I in the stone ages by even attempting this approach?
I would look at VUM but we are at Update 1 currently and it doesn't seem worth it until Update 2 at a minimum.
Thank you.
Check out esx-autopatch.pl at http://vmprofessional.com/index.php?content=utilities
It takes a little updating to get the patches listed in order along with md5sums.
I use this script at the end of my scripted build process since 3.5 U3 doesn't work on my IBM HS20 blades.
I also use this script in the same manner and am very content with it.