Forum Discussion
CirrusNode Connection Fails By Pool
Coming from the Cisco ACE Load Balancer, I frequently check the status of the nodes members (real servers) by using the following command:
ACE/C15 show server farm SERVERFARMNAME
serverfarm : SERVERFARMNAME, type: HOST
total rservers : 5
----------connections-----------
real weight state current total failures
---+---------------------+------+------------+----------+----------+---------
rserver: SERVER0
192.168.1.100:0 8 OPERATIONAL 506 0 0
rserver: SERVER1
192.168.1.101:0 8 OPERATIONAL 519 0 0
rserver: SERVER2
192.168.1.102:0 8 OPERATIONAL 501 0 0
rserver: SERVER3
192.168.1.103:0 8 OPERATIONAL 511 0 0
rserver: SERVER4
192.168.1.104:0 8 OPERATIONAL 490 0 42
This gives me a quick glance at the status of each of the servers in the server farm. More importantly, I can see the errors (no SYN response or reset from the server).
Is there an equavalent command to "show serverfarm SERVERFARMNAME" on F5, "show ltm pool SERVERFARMNAME" doesn't appear to have a column for failures.
Thank you,
MJ
2 Replies
gsharriAltostratus
LTM does not track per-node/pool member connection failures so you will not see any connection failed stats for individual nodes.
A health monitor assigned to node/pool member will generate a log msg and snmp trap when marking an object up or down.
Failures can also be seen at the layer 4 profile level for a virtual server:
tmsh show ltm profile tcp profile_name
Or other errors in tmm stats:
tmsh show sys tmm-traffic
TechTNimbostratus
Hi MJ, If you are looking out for a reset from the server , you can try running tcpdump on cli and see the packet capture output if reset is coming from the server end or not.
SYNTAX: tcpdump -nni 0.0 host x.x.x.x
