1

Maximum packet length reached?

view story
linux-howto

http://serverfault.com – Clients are having issues connecting to perforce revision server I have set up for them to use. It appears that they have too many checked out/added files that need to be put into the system and they are getting packet reseet issues. Specifically the TCP connection is resetting when its trying to communicate with the server. I made sure the MTU on the eth0 was set to 1492, but I am at a loss on what else could be done to help resolve this problem. I have csf firewall installed on the server and was searching for any kind of settings that could be tweaked for TCP packet limit but could find (HowTos)

dayu's picture
Submitted by dayu on 01/16/2013

Stories similar to Maximum packet length reached?

I am currently in the process of modifying how internet reaches my front end servers and I am having a strange issue with one of the existing stacks.

There is one physical firewall which is above the server, if I use this, I start recieving packet loss across the entire subnet, if I use wireshark I can see arp requests but I still get packet loss.

Very high packet loss in burst 39 weeks 36 min ago

We sometimes have 90%+ more packet loss on our server, but it does not always appends. Right now it works perfectly, but just half an hour ago, it had just that problem.

Our service provider is telling us to go in a recovery system to test if this is really a hardware problem and not software on our side.

Hi

Hi,I have a netbook I use as a small media server with a single nfs share running Ubuntu 12.04 server and my desktop running Arch. Both are connected to a Virgin Media Super Hub with ethernet cable. Over the last couple of days videos started to freeze when playing over in vlc on nfs.

Xen PV packet loss 29 weeks 4 days ago

I'm having some serious issues with packetloss with our home server. This server is a somewhat old (P4-era) machine running Debian Squeeze and Xen 4.0. There are two domUs running on it (both also Debian Squeeze), one gateway and a fileserver.

I am having Clear OS Community Edition running as router with 2 network interfaces.

Interface eth0 is connected to LAN with ip 192.168.122.2 and interface eth0 with ip 192.168.2.2 is having internet connection.

On the Client machine in LAN i have added proper gateway using

route add default gw 192.168.122.2 eth0

and i am able to ping gatway. but when making ping request to google.com,i am fac

For past days experienced in periods a very slow server and several crashes where the server becomes almost 100% unreachable - (Packet loss 70-90% from all location) and it is now network related as i have confirmed from the DC. A reboot fixes the problem usually temporary but the problem usually arises again within a few hours.

So, may i know how to investigate it.

I have to implement a pseudo TCP handshake that leads into a file transfer. I have no problem doing the file transfer: simple send-acknowledge. What I can't wrap my head around is how to do the three way handshake assuming there will be packet loss during one of the 3 steps.

For instance: My client sends a SYN packet.

UDP packet loss when cache enable 36 weeks 20 hours ago

UDP packet loss when we enable the cache.Here i am using MCF5485 cold fire processor in that Access Control Registers(bits 5-6,CM) i.e,00 Cache able,write-through.01 Cache able,copy back.10 Cache-inhibited, precise.11 Cache inhibited,imprecise. when we set 00 or 01 mode(i.e Enable mode) in signal.h file we are getting packet loss. packet length is 1300 and No of packets are 30,000.