Problem with vmnet-natd

[chrno]

Hi,

I'm running Vmware Server v1.07 on a Ubuntu 7.10 Gutsy Server (Kernel: 2.6.22-15).

For some unknown reason, vmnet-natd seems to keep crashed after a day or two. Is there anyway to just restart NAT service on /dev/vmnet8 without having to do a /etc/init.d/vmware restart ? Because if it's possible, I'll just put it in crontab until I figure out why this is happening ...

Also any advice on how to troubleshoot and find out why vmnet-natd is crashing?

Thanks in advance for any suggestions or recommendations :D

chrno

[falko]

Also any advice on how to troubleshoot and find out why vmnet-natd is crashing?
Did you check your logs?

[chrno]

hey falko,

Thanks for replying! Apologies for not replying earlier as I have been away for a couple of days.

There is nothing obvious in the following logs:

/var/log/syslog
or
/var/log/vmware/vmware-serverd.log

Are there any other logs that I should look at?

I've also checked out the Vmware Communities and someone mentioned that vmnet-natd will crash or experience performance degradation when the count in the following folder is near the ulimit max. (Mine default is 1024):

ls -l /proc/`pidof vmnet-natd`/fd/

Which kinda make sense as that is the amount of resources that can be allocated to a process. What do you reckon?

As a workaround, I have swapped to host-only networking and use iptables (on Vmware Host) to perform the NAT between eth0 and vmnet1 (host-only networking).

Seems to work out great for now. But still curious as to why vmnet-natd keeps failing on me ...

chrno

[falko]

I've also checked out the Vmware Communities and someone mentioned that vmnet-natd will crash or experience performance degradation when the count in the following folder is near the ulimit max.Can you post the link?

(Mine default is 1024):

ls -l /proc/`pidof vmnet-natd`/fd/

What's the output?