Alex,

What files are you seeing in vmware-tmp?

either by using  ls -l or lsof |grep vmware-tmp

Aaron



On 10/21/10 12:32 PM, Alexander Patterson wrote:
Hello,

This is an ongoing question about how the VCL scripts clean up the vmware
tmp directory.

I have a few questions about the process and what gets cleaned.

What files are deleted in the vmware tmp, when the vcl runs its cleanup
scripts. Are there any directory or files that are safe to manually delete
and where are they located in that directory?

Our campus has created local storage on the each blade for the vmware tmp
files for it’s called /vmware-tmp and adding the string to to put the tmp
directory in vmware to its own folder.



Here is our current vmware configuration setup

serverd.init.fullpath = "/usr/lib/vmware/serverd/init.pl"

authd.client.port = "904"

control.fullpath = "/usr/bin/vmware-cmd"

authd.fullpath = "/usr/sbin/vmware-authd"

loop.fullpath = "/usr/bin/vmware-loop"

libdir = "/usr/lib/vmware"

vmware.fullpath = "/usr/bin/vmware"

vmdir = "/virtualmachines"

dhcpd.fullpath = "/usr/bin/vmnet-dhcpd"

serverd.fullpath = "/usr/sbin/vmware-serverd"

tmpDirectory = "/vmware-tmp/"



datastore.name = "local"



datastore.localpath = "/virtualmachines/"



prefvmx.useRecommendedLockedMemSize = "TRUE"

prefvmx.minVmMemPct = "100"

defaultVMPath = "/virtualmachines"

priority.grabbed = "normal"

priority.ungrabbed = "normal"





This is very helpful before we were having our tmp fill up to %100 and crash
our blades.

Our currently DF looks like this.

Filesystem           1K-blocks      Used Available Use% Mounted on

/dev/sda2             60948136   3384160  54418048   6% /

/dev/sda6              8123168    293208   7410668   4% /home

/dev/sda5             10153988    695628   8934244   8% /var

/dev/sda1               101086     11878     83989  13% /boot

tmpfs                 24716364         0  24716364   0% /dev/shm

netappfiler02:/vol/sata1/virtualmachines

5610301032 1500130872 4110170160  27% /virtualmachines

/dev/sda7             49414916  35444872  11459892  76% /vmware-tmp

We are running a NFS mount to run our images off our current hardware.

The /vmware-tmp directory is slowly filling up again. This has been a huge
ongoing issue, we have already given the directory more space. We have it
limited before but within about 6 days, it’s getting full.

I know that the VCL scripts don’t point to this current custom directory and
run a cleanup.

Is there a way to change to code to clean up to a custom directory? Is there
a way to setup a cleanup script or cron job to clean up the directory after
X amount of time? I need to know what I can and can’t delete if I have to do
this manually as well?




--

Aaron Peeler
Program Manager
Virtual Computing Lab
NC State University
aaron_pee...@ncsu.edu
919-513-4571

Reply via email to