Optimizing server performance for handling large amount of NTP packet efficiently

[fxxkputin]

I have a ntp server, I found that when number of packet become large, the server seems overloaded which many packet is being dropped. So could you tell me how to tune sysctl.conf in order to optimize its ability for the server to handle large amount of UDP packets

[penguinpee]

Did you try searching? The information you provide is not very detailed. So, it be difficult to judge as to what causes your server to overload.

Depending on the root cause of the overload you experience there may be different solutions. Two (random) suggestions for further reading:

Sysctl configuration for high performance NTP server

If the server has enough RAM, you may want to reduce the chance of swapping pages to disk. This is controlled by the vm.swappiness setting. A lower setting reduces the swapping activity. Default is afaik 60, My server uses a value of 20, but you can set lower values if you want to. vm.swappiness = 20 Another important performance setting is in the firewall where you have to switch off connection tracking for NTP packets. Assuming your server uses iptables, add the following lines to your iptab…

"io timeout" in CSV logs

Seems more stable now, i’ve dropped server speed configuration to 768kbit, this is a little shameful but it keeps server available. Also, tunned some parameters at sysctl in server side. Like: net.netfilter.nf_conntrack_udp_timeout = 120 net.netfilter.nf_conntrack_udp_timeout_stream = 720 net.netfilter.nf_conntrack_buckets = 48256 net.netfilter.nf_conntrack_max = 262144 net.nf_conntrack_max = 262144 net.core.netdev_max_backlog = 350000 net.core.rmem_max = 38414400 net.core.rmem_default =…

[HAHAHAHA]

What’s the specifications, NTP server software configuration, and how much traffic you were seeing which caused packet drop?

Nobody can tell what’s going on without knowing any details.

[fxxkputin]

packet drop increase sharply when bandwidth more than 20mbit/s(real bandwidth, not net speed). after optimizing settings the problem still exist, so i think is OS problem(I previously use slackware due to its reputation of lightweight which i think it may increase performance.) However NTP performance actually decrease compared to mainstream OS like Ubuntu because after reinstalling the server with Ubuntu the problem disappear completely.