We have configured Multicast routing on 3550-EMI Core Switches as well as CGMP compatibility for our 3548XL L2 edge switches. Currently Multicast imaging works in most of the clients LAN locations. However at one particular site they are having massive packet loss reported by the client at which point the image process fails on all but one or two computers.
I am pretty confident that the network configuration is not the issue. The site we are having problems at was the demo site for the design and rollout of the rest of the networks edge sites so the configs are virtually identical. We have done everything from nuke the network configs and reconfigure from scratch to doing a diff against a working site and the broken sites configs and so far nothing jumps out at me. Today we are testing again and I have noticed that the Multicast stream from the imaging server is 15214 kbps via the sh ip mroute active command. From what I understand this tells me that my imaging server is dishing out a 15214 kbps multicast stream for the clients to connect to. This in turn translates to a 15+ mbps data stream. The system admin doing the imaging thinks he is sending a 1 mbps data stream and is arguing that the 15+ number is a byproduct of 15 client systems connecting to the image
I will attach a sanatised config of the 3550 EMI as well as one of the 3548XL edge switches. Any thought or ideas you might have would be greatly appreciated.
Thanks for your reply. I should have already replied to this but things have been busy. We were able to resolve the issue.
The information you provided had already been dealt with and should have been reflected in the attached configs. In this case we have 22 of 23 site that were functioning correctly when it comes to multicast imaging. Our TTL is set to 5 which exceeds our total routing hops in the LAN. We are also not using Ghost. I Point this out because it is key in understanding what our final solution was. We are using OS X's built in Netboot and imaging platform. We had run full multicast tests using VLC media player to provide a multicast video stream across our 32 VLANs on the site successfully. This combined with verification of our multicast configuration on our switches and core L3 switch had us convinced that this was not a networking issue.
In the end we were throwing out ideas and we decided to compare the IOS revisions on the 3550 EMI switches that were working compared to the one site that was not here is what we found.
This was a rude awakening. All of our normal multicast test (That TAC approved!) worked but our primary multicast application failed at an alarming rate. At the end of the day the only thing we can figure is that our Multicast tests using VLC media player were successful due to the applications buffering mechanism. When it came to the OS X imaging application it appears that it has a much less capable buffering and error correction capability.
While using the SEE version of the 12.2 code we saw now errors in debugging, interfaces or stream setup and join but once more than 3 clients joined the stream we would begin seeing huge packet loss on the client imaging applications.
As soon as we moved to the SEE2 version of the 12.2 IOS not only did we have 100% success but we were able to crank up our network and Client speeds to max recommended for 100 Mbps Ethernet and Full speed for G4 Flat screen Imacs.
I am taking this as a harsh reminder that regardless of testing and lack of errors, when in doubt upgrade or downgrade your IOS rev to test for issues in code.
Hi everyone, I would like to thank you in advance for any help you can provide a newcomer like myself!
Im studying the 100-105 book by Odom and am currently on the topic of Port security. I purchased a used 2960 and I'm trying to follow a...
While deploying a number of 18xx/2802/3802 model access points (APs), which run AP-COS as their operating platform. It can be observed on some occasions that while many of their access points were able to join the fabric WLC withou...
I am going to design and build an LAN network under a tunnel underground with long distance between the switches.
I will have 2 Catalyst switches and 8 Industrial IE3000, and they will be connected with fiber.
For now I am planning on use Layer-2 s...