SCO OpenServer 6.0 Network Performance

[Vmwared]

We have successfully installed and configured SCO Openserver 6 on an ESXi 3.5 update 3 host.

Applications are working however we are experiencing very poor performance on the network throughput.

Modified the .VMX file to add the "E1000" network, replacing the "Flexible" AMD nic. Performance improved but did not clear issue.

Adjust the Speed and Duplex settings with no material affect. Also adjusted the CheckSum location processing with no change. The attached .PNG file shows the network setting after each speed/duplex change. The traffic is being generated by a single file (137MB) being copied from the virtual machine to another server in the blade chassis.

n

A second virtual windows 2003 server is also located on the host and has normal network throughput.

Any suggestions would be appreciated. Have run this by the Networking group at VMware Tech Support and they are not sure what is happening with this.

Thanks!!

[cwatkins01]

I am also having this exact same problem. This is my first time installing SCO OS6 in a virtualized environment. I have also modifed the .vmx file and changed the nic from "Flexible" to "E1000" and removed/re-added the LAN adapter in SCO. Ping times are slower than normal and FTP is extremely slow.

Pinging SCO OS6 VM (10.10.16.197) from another SCO box (10.10.16.97, not virtualized, on the same switch)

PING 10.10.16.197 (10.10.16.197): 56 data bytes

64 bytes from 10.10.16.197 (10.10.16.197): icmp_seq=0 ttl=64 time=14.168 ms

64 bytes from 10.10.16.197 (10.10.16.197): icmp_seq=1 ttl=64 time=9.578 ms

64 bytes from 10.10.16.197 (10.10.16.197): icmp_seq=2 ttl=64 time=9.812 ms

64 bytes from 10.10.16.197 (10.10.16.197): icmp_seq=3 ttl=64 time=3.414 ms

--- 10.10.16.197 ping statistics ---

4 packets transmitted, 4 packets received, 0% packet loss

round-trip min/avg/max = 3.414/9.243/14.168 ms

Pinging SCO OS6 VM (10.10.16.198) from SCO OS6 VM (10.10.16.197) on the same ESXi host

PING 10.10.16.198 (10.10.16.198): 56 data bytes

64 bytes from 10.10.16.198 (10.10.16.198): icmp_seq=0 ttl=64 time=43.754 ms

64 bytes from 10.10.16.198 (10.10.16.198): icmp_seq=1 ttl=64 time=10.144 ms

64 bytes from 10.10.16.198 (10.10.16.198): icmp_seq=2 ttl=64 time=17.051 ms

64 bytes from 10.10.16.198 (10.10.16.198): icmp_seq=3 ttl=64 time=9.208 ms

--- 10.10.16.198 ping statistics ---

4 packets transmitted, 4 packets received, 0% packet loss

round-trip min/avg/max = 9.208/20.039/43.754 ms

FTP’ing 10MB file from another SCO box (10.10.16.97), not virtualized, on the same switch to SCO OS6 VM (10.10.16.197)

ftp> get test1.txt

local: test1.txt remote: test1.txt

200 PORT command successful.

150 Opening BINARY mode data connection for test1.txt (10111878 bytes).

226 Transfer complete.

10111878 bytes received in 4.5 seconds (2.2e+03 Kbytes/s)

[LeeSolt]

We spent a lot of time with SCO on this and they have identified it as a problem and will not be able to address until their VMware tools come out in the May timeframe. Not happy with that as it derailed our project.