Linksys, Netgear, ect. Webbase configurations for switches or layer 2 devices.
1 post • Page 1 of 1
Hello Jens, Thank you for participating in the Small Business support community. My name is Nico Muselle from Cisco Sofia SBSC. These log messages mean that some device in your network is flooding broadcast UPnP messages over the switches in such a way that the switch(es) doesn (don ) have time to handle all of them, thus an overflow occurs. Going back to the moment these messages started to appear, has anything changed on the network, has any device been added at that moment that could have caused this UPnP flooding ? The easiest way to solve this is if you know where the traffic is generated and configure this particular device in such a way that it stops flooding the network with UPnP packets. If you don know what device causes it, the best thing to do is to monitor the network traffic for UPnP messages using Wireshark and determine there what is the source IP of these packets, to be able to put a halt to it. This way, you could also decide what is the destination multicast IP address and eventually implement an ACL on the source port to block this traffic if you cannot turn it of on the device causing it. Hope this helped, Best regards, Nico MuselleSr. Network Engineer - CCNA
Hello Nico, the change is simple: weve changed the backbone from Summit to Cisco But the network is large and it could also be another cause... Now the message appears every few minutes:11-Apr-2011 16:49:56 %UPNP-W-DBOVERFLOW: UPnP retransmission DB overflow.11-Apr-2011 16:51:20 %UPNP-W-DBOVERFLOW: UPnP retransmission DB overflow.11-Apr-2011 16:51:25 %UPNP-W-DBOVERFLOW: UPnP retransmission DB overflow. Ive monitored the connection to one Linksys and the six SSDP-packets before the syslog packet (DB-Overflow) are all from the same switch! No. Time Source Destination Protocol Info
Hi Jens, Thank you for the log. As you indicated, you have a large network with a lot of devices. The SRW device has an internal memory pool that keeps UPnP messages for retransmission. The %UPNP-W-DBOVERFLOW: UPnP retransmission DB overflow messages indicate that this memory pool has been exhausted.This scenario might occur in networks with lots of UPnP devices and many UPnP control points each requesting UPnP reports from the SRW device.For example if you have a network with many PCs running Win XP and UPnP UI installed, the device might get many queries and utilize 100% of thismemory pool.As far as we can see, the behavior is as designed - overflow occurs when too many messages are received within a short period of time.There is no memory leak, and after the en-queued messages were handled, new requests are accepted by the device.To summarize, these log messages are normal and do not adversely affect the performance of the switch. The switch should continue to filter and forward frames as configured.However, you could verify whether all this UPnP traffic is necessary on the network and eventually reconfigure the hosts generating it.Please let me know if you have any other questions.Best regards, Nico MuselleSr. Network Engineer - CCNA