2

I am running multiple instances of siege, so siege is reusing the ports as a results some of the requests are not going through. Is there a way where the different siege instances could use different port ranges? 

   HTTP/1.1 200   0.00 secs:     146 bytes ==> 
   HTTP/1.1 200   0.00 secs:     146 bytes ==> 
   HTTP/1.1 200   0.00 secs:     146 bytes ==> 
   HTTP/1.1 200   0.00 secs:     146 bytes ==> 
   HTTP/1.1 200   0.01 secs:     146 bytes ==> 
   HTTP/1.1 200   0.00 secs:     146 bytes ==> 
   HTTP/1.1 200   0.01 secs:     146 bytes ==> 
   [alert] socket: 671299328 select timed out: Connection timed out
   [alert] socket: 788797184 select timed out: Connection timed out
   [alert] socket: 721655552 select timed out: Connection timed out
   [alert] socket: 738440960 select timed out: Connection timed out
   HTTP/1.1 200   0.01 secs:     146 bytes ==> /
   HTTP/1.1 200   0.01 secs:     146 bytes ==> /
   [alert] socket: 822368000 select timed out: Connection timed out
   HTTP/1.1 200   0.01 secs:     146 bytes ==> /
   HTTP/1.1 200   0.01 secs:     146 bytes ==> /
   HTTP/1.1 200   0.01 secs:     146 bytes ==> /
Ria
  • 10,237
  • 3
  • 33
  • 60
user2952766
  • 21
  • 1
  • 2
  • What did you try so far? – L.Butz Nov 04 '13 at 14:27
  • Have you determined whether the problem is on the client or the server? It's not clear to me from either answer or the question which is the case. In my case there seems to be some evidence for each possibility. – iconoclast Jul 03 '14 at 14:43

2 Answers2

0

I see you have a lot of requests one after another, do you consider that you can have problems with KeepAlive.

On server sockets are opened for a little bit longer than connection alone. You can run out of ports quite qucik if KeepAlive is set to high value.

Mateusz
  • 2,287
  • 20
  • 28
0

You can set net.ipv4.tcp_tw_reuse = 1 net.ipv4.tcp_tw_recycle = 1 in /etc/sysctl file and run #sysctl -p to active it. Please have a try. Hope this can help you.

Queenie Wong
  • 123
  • 1
  • 4
  • 11