Hi, I have several OpenVPN dedicated hardware servers on different datacenters with hundreds of users connected to each. Recently I've changed the client-connect and client-disconnect scripts to point to new scripts that can take up to 2 seconds more to run than the previous ones took (the previous script ran in about 500 miliseconds).
After that, a handful of the users started complaining about packet loss (too much breaking on voip calls, lag on ssh connections, etc) on different servers (different countries, different ISPs). I can see indeed the counters of UDP packet receive errors with: # netstat -csu (the network cards are good, no bad frames) Is it possible that the time the client-connect and client-disconnect scripts take to run can affect the packet loss of OpenVPN? Is there any known bad practice of what not to do when writing client-connect scripts? Are there other internal causes on OpenVPN can cause packet drops? Users connect through UDP, I'm running OpenVPN openvpn-2.3.6-1.el6.x86_64 on CentOS 6. Thanks in advance, -- Vinicius Mello ------------------------------------------------------------------------------ Presto, an open source distributed SQL query engine for big data, initially developed by Facebook, enables you to easily query your data on Hadoop in a more interactive manner. Teradata is also now providing full enterprise support for Presto. Download a free open source copy now. http://pubads.g.doubleclick.net/gampad/clk?id=250295911&iu=/4140 _______________________________________________ Openvpn-users mailing list Openvpn-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openvpn-users
