We have recently had an issue on the network, we have a cisco 3560X switch which connects to our 6509e core at 1 gig, and then also to a 100meg wan link.
We had qos enabled on the switch, priority queue and mls qos trust dscp command was set on the ports that connect to the core but not on the wan port.
The WAN port was never more then 10% utilised.
We had issues wherby an application what runs over the 100meg wan link kept experiencing lost packets, from doing the trace, and lots of tcp retransmissions going on.
As soon as we disabled QOS on the switch the issue for the application is now fixed.
My question is, why was this happening, there were no drops on the interface to the wan or core!
Do you think we need to enable the qos on the wan link and priority queue also?
the engineer who found the issue said it may be because qos wasnt on the wan port and only to the core, or it could be because you are going from a 1 gig link to a 100 meg link! even though no congestion!
any ideas? and what would be best practice for this, we simply just want to prioritize the voice over the wan marked with EF, and everything else best effort. Is the best effot queue big enough maybe ?
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
From what you're describing, you may be describing a software defect. You note packets were lost, but no drop counters incremented, yet when QoS was disabled, drops went away.
By default, when you enable QoS on a 3560X it allocates buffers to 4 egress queue (vs. allocating all buffers to just one queue when QoS is disabled). QoS default buffers allocation often increases the drop rate because of reduced buffers per queue.
Although deactivation of QoS seems to point at the 3560X, its common to have issues with WAN Ethernet hand-offs when CIR isn't full link bandwidth. It's also possible, your WAN Ethernet hand-off deals with marked packets differently, and as enabling QoS, also by default, resets L3 ToS, you might have an issue there too.
As what you might do, too little information. In your other post on this subject, I asked about drops, and you noted none seen on the network device, but you didn't mention the trace results and/or TCP retransmissions. Perhaps you didn't know that at the time you answered my earlier question, but it's an example of how information can lead into a whole new direction.
With QoS, there's often a huge number of "could be"s.
Oh, I guess your earlier question "Can qos cause issues on a LAN", has been answered. Now you see why I asked whether you had yet tried your switch with QoS disabled.
We are pleased to announce availability of Beta software for 16.6.3. 16.6.3 will be the second rebuild on the 16.6 release train targeted towards Catalyst 9500/9400/9300/3850/3650 switching platforms. We are looking for early feedback from custome...