eth0: transmit timed out

DISCUSSION Arne Kepp 2 participants 4 messages Feb 25, 2008 KVM

Share & tools

Arne Kepp

author

2008-02-25 00:48 UTC #0 lore raw

Hi,

I am running KVM 61 on CentOS 5.1 x86_64 (on a Dell PowerEdge 2950 III). I start my virtual machine as follows:
qemu-kvm -hda /dev/kvm_sdb/geust_root -m 1024 -net tap -net nic,vlan=0,macaddr=00:16:3e:00:00:02 -daemonize -nographic -smp 2

The problem is that when the guest (also CentOS 5.1 x86_64) utilizes the network heavily (scp to a machine on the same switch, around 12 mbyte/s) the network on the virtual machine just drops out.

I can no longer ping the virtual machine, and any existing SSH sessions die. If I connect to the virtual machine using VNC everything looks okay (eth0 is up, routes are okay), but I still cannot connect to anything. All other virtual machines on the same host continue to work as if nothing has happened. The same problem occurred in -58 and -59.

/var/log/messages on the guest says:
NETDEV WATCHDOG: eth0: transmit timed out
eth0: link up, 100Mbps, full-duplex, lpa 0x05E1

Any suggestions (permanent fixes or temporary workarounds, I'll take what you got) would be much appreciated.

Thanks for all the great work :)
-Arne

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

david ahern

2008-02-25 02:51 UTC #1 lore raw

I presume you are using the default rtl8139 nic. Correct?

What does 'ethtool -S eth0' show when the network locks up? Many months ago adding 'noapic' to the (guest) kernel boot parameters helped, but that option is not currently helping with my RHEL4 networking issues.

A thread for this issue is at:
http://sourceforge.net/tracker/index.php?func=detail&aid=1802082&group_id=180599&atid=893831

david

Quoted 27 lines · Arne Kepp

Hi,
I am running KVM 61 on CentOS 5.1 x86_64 (on a Dell PowerEdge 2950 III). I start my virtual machine as follows:
qemu-kvm -hda /dev/kvm_sdb/geust_root -m 1024 -net tap -net nic,vlan=0,macaddr=00:16:3e:00:00:02 -daemonize -nographic -smp 2
The problem is that when the guest (also CentOS 5.1 x86_64) utilizes the network heavily (scp to a machine on the same switch, around 12 mbyte/s) the network on the virtual machine just drops out.
I can no longer ping the virtual machine, and any existing SSH sessions die. If I connect to the virtual machine using VNC everything looks okay (eth0 is up, routes are okay), but I still cannot connect to anything. All other virtual machines on the same host continue to work as if nothing has happened. The same problem occurred in -58 and -59.
/var/log/messages on the guest says:
NETDEV WATCHDOG: eth0: transmit timed out
eth0: link up, 100Mbps, full-duplex, lpa 0x05E1
Any suggestions (permanent fixes or temporary workarounds, I'll take what you got) would be much appreciated.
Thanks for all the great work :)
-Arne

Quoted 9 lines

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
kvm-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Arne Kepp

author

2008-02-25 04:07 UTC #2 lore raw

Thanks David,

Yes, I am using rtl8139 and it is running 2.6.18-53.1.13.el5 (Red Hat kernel, CentOS).

ethtool -S eth0 on the guest says, when locked up:
NIC statistics:
early_rx: 0
tx_buf_mapped: 0
tx_timeouts: 4
rx_lost_in_ring: 0

It appears that 'noapic' did the trick, which is great, thanks! I am somewhat worried I'll be missing out on some SMP performance? But that's secondary.

Please let me know if there is anything I can do to gather any more info to help debug the problem.

-Arne

Quoted 61 lines · david ahern

I presume you are using the default rtl8139 nic. Correct?
What does 'ethtool -S eth0' show when the network locks up? Many months ago adding 'noapic' to the (guest) kernel boot parameters helped, but that option is not currently helping with my RHEL4 networking issues.
A thread for this issue is at:
http://sourceforge.net/tracker/index.php?func=detail&aid=1802082&group_id=180599&atid=893831
david
Quoted 27 lines · Arne Kepp
Hi,
I am running KVM 61 on CentOS 5.1 x86_64 (on a Dell PowerEdge 2950 III). I start my virtual machine as follows:
qemu-kvm -hda /dev/kvm_sdb/geust_root -m 1024 -net tap -net nic,vlan=0,macaddr=00:16:3e:00:00:02 -daemonize -nographic -smp 2
The problem is that when the guest (also CentOS 5.1 x86_64) utilizes the network heavily (scp to a machine on the same switch, around 12 mbyte/s) the network on the virtual machine just drops out.
I can no longer ping the virtual machine, and any existing SSH sessions die. If I connect to the virtual machine using VNC everything looks okay (eth0 is up, routes are okay), but I still cannot connect to anything. All other virtual machines on the same host continue to work as if nothing has happened. The same problem occurred in -58 and -59.
/var/log/messages on the guest says:
NETDEV WATCHDOG: eth0: transmit timed out
eth0: link up, 100Mbps, full-duplex, lpa 0x05E1
Any suggestions (permanent fixes or temporary workarounds, I'll take what you got) would be much appreciated.
Thanks for all the great work :)
-Arne
Quoted 10 lines
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
kvm-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/kvm-devel
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
kvm-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/kvm-devel

Quoted 3 lines

!DSPAM:4038,47c22d43155652085621377!

david ahern

2008-02-25 04:39 UTC #3 lore raw

Those stats suggest you are using the 8139too nic driver. You should switch to the 8139cp driver.

david

Quoted 16 lines · Arne Kepp

Thanks David,
Yes, I am using rtl8139 and it is running 2.6.18-53.1.13.el5 (Red Hat kernel, CentOS).
ethtool -S eth0 on the guest says, when locked up:
NIC statistics:
early_rx: 0
tx_buf_mapped: 0
tx_timeouts: 4
rx_lost_in_ring: 0
It appears that 'noapic' did the trick, which is great, thanks! I am somewhat worried I'll be missing out on some SMP performance? But that's secondary.

Quoted 77 lines

Please let me know if there is anything I can do to gather any more info to help debug the problem.
-Arne
Quoted 69 lines · david ahern
I presume you are using the default rtl8139 nic. Correct?
What does 'ethtool -S eth0' show when the network locks up? Many months ago
adding 'noapic' to the (guest) kernel boot parameters helped, but that option is
not currently helping with my RHEL4 networking issues.
A thread for this issue is at:
http://sourceforge.net/tracker/index.php?func=detail&aid=1802082&group_id=180599&atid=893831
david
Quoted 30 lines · Arne Kepp
Hi,
I am running KVM 61 on CentOS 5.1 x86_64 (on a Dell PowerEdge 2950 III). I start my virtual machine as follows:
qemu-kvm -hda /dev/kvm_sdb/geust_root -m 1024 -net tap -net nic,vlan=0,macaddr=00:16:3e:00:00:02 -daemonize -nographic -smp 2
The problem is that when the guest (also CentOS 5.1 x86_64) utilizes the network heavily (scp to a machine on the same switch, around 12 mbyte/s) the network on the virtual machine just drops out.
I can no longer ping the virtual machine, and any existing SSH sessions die. If I connect to the virtual machine using VNC everything looks okay (eth0 is up, routes are okay), but I still cannot connect to anything. All other virtual machines on the same host continue to work as if nothing has happened. The same problem occurred in -58 and -59.
/var/log/messages on the guest says:
NETDEV WATCHDOG: eth0: transmit timed out
eth0: link up, 100Mbps, full-duplex, lpa 0x05E1
Any suggestions (permanent fixes or temporary workarounds, I'll take what you got) would be much appreciated.
Thanks for all the great work :)
-Arne
-------------------------------------------------------------------------
Quoted 9 lines
This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
kvm-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/kvm-devel
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
kvm-devel mailing list [email protected]
https://lists.sourceforge.net/lists/listinfo/kvm-devel
!DSPAM:4038,47c22d43155652085621377!