|
* What precisely does the InActConn figures show?
Any connection that is not in the ESTABLISHED state.
Most of these will be closed, just waiting to time out -
they are kept around in case any stray packets show up.
You can observe the state of connections using ipvsadm -Lcn
HI Horms,
Thanks for the response. OK. I can see a lot of connections in the
FIN_WAIT state. The cluster is now lightly-loaded - but despite that the
number of connections in the FIN_WAIT state seems too high? Is this
something to worry about or is it normal?
When the cluster comes under a lot of load the InActConn numbers
grow very quickly and jump to over 2000. Is there a resource penalty
for FIN_WAIT connections? - the simple answer may be there is not enough
capacity in the cluster to handle the higher loads.
Here is a snapshot under light load:
IP Virtual Server version 1.0.10 (size=65536)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 66.98.x.y:80 rr
-> 66.98.x.y:80 Tunnel 1 7 1095
-> 67.15.x.y:80 Tunnel 1 5 1101
-> 66.98.x.y:80 Tunnel 1 3 1099
-> 207.44.x.y:80 Tunnel 1 4 1092
-> 207.44.x.y:80 Tunnel 1 5 1098
-> 69.57.x.y:80 Tunnel 1 4 1096
-> 67.15.x.y:80 Tunnel 1 7 1095
-> 67.15.x.y:80 Tunnel 1 20 1091
-> 207.44.x.y:80 Tunnel 1 5 1104
-> 66.98.x.y:80 Tunnel 1 8 1093
-> 67.15.x.y:80 Tunnel 1 13 1086
-> 67.15.x.y:80 Tunnel 1 6 1095
-> 69.57.x.y:80 Tunnel 1 5 1111
-> 69.57.x.y:80 Tunnel 1 6 1100
-> 67.15.x.y:80 Tunnel 1 9 1112
-> 69.57.x.y:80 Tunnel 1 5 1106
-> 67.15.x.y:80 Tunnel 1 10 1086
-> 67.15.x.y:80 Tunnel 1 17 1082
-> 67.15.x.y:80 Tunnel 1 13 1094
-> 67.15.x.y:80 Tunnel 1 9 1109
-> 207.44.x.y:80 Tunnel 1 5 1081
-> 207.44.x.y:80 Tunnel 1 5 1098
-> 66.98.x.y:80 Tunnel 1 3 1108
-> 207.44.x.y:80 Tunnel 0 0 0 # busted node
-> 207.44.x.y:80 Tunnel 1 4 1089
Kind regards,
Nige
| 