Anybody has any idea why I would have lots of CRC errors in different ATM ports, in different VIPS and different routers ! (7500 and 7200 platforms) The CRC errors were so much CPU blew up to 100%, and IP traffic was practically down. Strange thing is packets per second in interfaces was well above average (average packet size was 56 bytes, where I usually have a 400 bytes average size). Something with ATM switches ??
I suspect you are exceeding your scr (cir in atm) , some cells are are being dropped by the atm switch in the cloud then a packet is recieved by the router, since the atm layer did not see the end of the packet when it was converting back into a packet from cells it creates a packet that is a giant, since it did not know that some cells were lost in the cloud and a crc counter would increment
We already have traffing shaping in our ATM sub-interfaces. Anyway, traffic shaping is only good for outgoing traffic. CRC errors are due to incoming packets. Somehow we were receiving lots of cells with EOM=1, hence when the packet was assembled it contained CRC errors, probably due to missing cells which were lost in transit. And I'm not talking about some CRCs, thousands of them. Traffic was practically down due to this. Our ATM switches were not discarding cells.
Could have been a DDOS ?
Maybe our upstreams provider's ATM switches were discarding cells ? Anyone has experience with this ?
What is the outbound rate on the PVC on the remote end , this may also need to be traffic shaped , Another thing to do would be to call your provider and have them check the settings in the cloud , it would not be the first time they misconfigured something , possibly they were making changes or a switch rebooted and it took a default setting that is much lower .
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...