Hello,
Seem to have found another problem with VMware snapshots and *NIX.
The previous issue with Solaris was fixed with patch ESX-6431040.
I now have found a similiar problem with Linux 2.6 kernels.
I am specifically running RHEL 4 WS/AS.
When a snapshot is taken from the VM it loses network connectivity and the below is found in dmesg and messages file:
******
e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
TDH
NETDEV WATCHDOG: eth0: transmit timed out
e1000: eth0: e1000_watchdog: NIC Link is Up 1000 Mbps Full Duplex
***
When the snapshot is committed a similiar error is posted.
Anyone else experiencing the same problem and maybe have a fix?
Thanks,
drew
Does this happen with the vmxnet NIC too?
We are seeing similar problems with 2.6.x kernels and snapshots.
e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
Tx Queue
NETDEV WATCHDOG: eth1: transmit timed out
e1000: eth1: e1000_watchdog: NIC Link is Up 1000 Mbps Full Duplex
We are running Debian Etch as 64 bit VMs.
\# uname -a
Linux paprika 2.6.18-4-amd64 #1 SMP Mon Mar 26 11:36:53 CEST 2007 x86_64 GNU/Linux
vmxnet is unfortunatly not available with 64 bit linux.
Cheers,
René
I've just tried the latest e1000 driver (7.6.5), and it does the same thing.
I believe to be experiencing this same issue, did anyone ever find a working solution?
Thank you
Are you taking a snapshot of the disk only or a snapshot of memory? Linux is finicky when it comes to memory snapshots... There is an option to take a snapshot and DON'T enable the memory option ![]()
I believe we are currently taking snapshots if the memory as well. Should I not do that with my Linux VM's?
