Let's back up a minute. All things being equal, aside from the delay in
CUPS printing, what else is not working at or beyond par? Is the printing
issue the ONLY issue? How long is the delay? Is it between pages or
between jobs or just a lag for a given job to start from a cold start
(nothing else printing at the moment)? How do the users log/use the
system?  If telnet or ssh, is there any delay in connecting there? Any
changes elsewhere in the environment as part of this upgrade (such as
network equipment/configuration, version of linux on the guests, DNS
servers, etc)? Anything out of the ordinary in /var/log/messages or the
tail of the 'dmesg' output that is present on the guest in question and
not when it was physical? I assume the guest's physical form still exists
somewhere? How are the cups printers connected to the network and likewise
configured in CUPS on the guest? If TCP socket based setup (ie, TCP port
9100), what happens if you open a terminal window on the guest in question
and execute a telnet to the printer's IP address on port 9100 (ie, 'telnet
<ipaddr> 9100')?  Does it connect immediately or is there a delay? Not
sure if you have more than one printer, but if so, do all printers exhibit
the slowness or just certain ones?  In vSphere Client, as I'm sure you are
aware, the host and guest both have tabs for "Resource Allocation",
"Performance" and "Events".  Between periods of printing and non-printing,
is there anything that seems out of whack or wildly varying?  

Chris is right, you need to install VMware Tools.  It is very easy on
CentOS 6.  When you tell vSphere Client to install VMWare Tools on the
CentOS guest, it will connect a small .iso image to the guest, and, if you
are running a GUI console for the guest, and are logged in as root, it
will likely automount it for you.  However it does, you need to locate the
mount point, then from, say, /root, extract the tarball on the mounted
image.  There is a perl script in the extracted folder to install vmware
tools.  Run it and just take the default for everything (initially at
least).  You should probably reboot the guest too once it's done as
depending on how it finds the guest's environment, it may build a new
kernel boot image.  Once VMWare Tools is running, the guest will
communicate a lot more info back to the host and this will provide
additional information about resource utilization at the guest level with
respect the ESXi and the host and may help identify bottlenecks at the
VMWare level.

-----Original Message-----
From: [email protected] [mailto:[email protected]] On
Behalf Of Howard White
Sent: Tuesday, November 20, 2012 2:05 PM
To: [email protected]
Subject: Re: [nlug] the continuing saga of the reluctant systems
administrator

On 11/20/2012 01:42 PM, Chris McQuistion wrote:
> What's your disk subsystem with the new ESXi host?  How much physical 
> memory and CPU does the box have and how much have you designated to 
> your VM's?
>
> Do you VMWare Tools installed on the Guest VM's?
>
> Chris
>

300GB of mirrored SAS drives on a 3ware RAID controller.
16GB of physical memory with 8GB allocated to guest 1 and 4gb allocated to
guest 2.
Has one four-core CPU; guest 1 has "3 cores" allocated and guest 2 has
one.
Top is showing 24k of swap used but current memory used is quite a bit
less than 8GB.
And yes, Tilghman, vm.swappiness=85.

Reading up on VMWare Tools

Howard

--
You received this message because you are subscribed to the Google Groups
"NLUG" group.
To post to this group, send email to [email protected] To
unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/nlug-talk?hl=en

-- 
You received this message because you are subscribed to the Google Groups 
"NLUG" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/nlug-talk?hl=en

Reply via email to