Begin forwarded message:
Date: Mon, 13 Feb 2006 14:21:57 -0800 From: [EMAIL PROTECTED] To: [EMAIL PROTECTED] Subject: [Bugme-new] [Bug 6068] New: UDP packet loss and slow communication http://bugzilla.kernel.org/show_bug.cgi?id=6068 Summary: UDP packet loss and slow communication Kernel Version: 2.6.15.4 Status: NEW Severity: normal Owner: [EMAIL PROTECTED] Submitter: [EMAIL PROTECTED] Most recent kernel where this bug did not occur: 2.6.15.4 Distribution: Crux Hardware Environment: Any of our system. several is based on Asus motboard and P4 CPU or Tyan and P4 CPU. Software Environment: HLDS and standard routing Problem Description: Since version 2.6.13.x we have got problem with slow UDP communication. We have a lot of complaining from Steam/CS userers on our Game servers. And we also get strange communication error through our main Router. I'm not any programmer and can't found out more details because I'm not working with development. I only install systems and configure them. Steps to reproduce: Basic install an 2.6.15.4 kernel from Crux LiveCD version 2.1 with udev. http://jaeger.morpheus.net/linux/crux/files/crux-2.1-latest.iso One installation is and pure Router with extra tweak's for buffering. Installing HLDS (CS server) on this kernel version cause problem with a lot of choke and a lot of udp pacet loss.. And the latencey is not stabel in the local net. In general the latency is now more the 100% higher. If I go back and use 2.6.12.6 then the problem is not there. I'm sorry but I can't really define the problem more but I'm not a programmer. On HLDS (CS servers) running without netfilter (IPTABLES) so that is not an issue. The Main router use NETFILTER but the problem will still be therer if I turn IPTABLES off.. We only use netfilter for som blocking on the main router. I have checked several settings in the /proc/sys/net/ipv4/ and /proc/sys/net/ipv4/netfilter some of this value is changed in the mainrouter. echo "65535" >/proc/sys/net/ipv4/netfilter/ip_conntrack_max echo '1' > /proc/sys/net/ipv4/ip_forward sleep 1 /etc/rc.route # buffer echo "512" > /proc/sys/net/ipv4/neigh/default/gc_thresh1 echo "2048" > /proc/sys/net/ipv4/neigh/default/gc_thresh2 echo "4096" > /proc/sys/net/ipv4/neigh/default/gc_thresh3 echo "45" > /proc/sys/net/ipv4/netfilter/ip_conntrack_tcp_timeout_close_wait echo "600" > /proc/sys/net/ipv4/netfilter/ip_conntrack_tcp_timeout_established /sbin/modprobe ip_conntrack hashsize=65535 No special settings is made on the HLDS servers. Regards Charlie G Mentorez... ------- You are receiving this mail because: ------- You are on the CC list for the bug, or are watching someone who is. - To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html
