cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
2783
Views
0
Helpful
4
Replies

IP HTTP SLA / Proxy - I want to react on socket error

jpsenior_bell
Level 1
Level 1

The below configuration works when the proxy server can serve URLS. I get HTTP 200 OK in my statistics with this and things are good.

I want to specifically monitor and trap when the proxy itself goes down, and I have the below configured.

The return code always seems to be 'socket receive error', but I can't seem to generate a trap on this error.

ip sla monitor 127

type http operation get url http://www.google.com/ name-server 192.168.2.200 proxy http://192.168.1.100:80

frequency 60

ip sla monitor schedule 127 life forever start-time now

ip sla monitor reaction-configuration 127 react verifyError threshold-type immediate action-type trapOnly

ip sla monitor reaction-configuration 127 react connectionLoss threshold-type immediate action-type trapOnly

ip sla monitor reaction-configuration 127 react timeout threshold-type immediate action-type trapOnly

ip sla monitor reaction-configuration 127 react rtt threshold-type immediate action-type trapOnly

debug ip sla monitor trace 127 shows:

May 28 17:21:36.673 MST: IP SLA Monitor(127) Scheduler: Starting an operation

May 28 17:21:36.673 MST: IP SLA Monitor(127) http operation: Starting http operation

May 28 17:22:06.675 MST: IP SLA Monitor(127) Scheduler: Updating result

May 28 17:22:06.675 MST: IP SLA Monitor(127) http operation: Wait connection - Sock recv error

#sh ip sla monitor statistics 127

Round trip time milliseconds (RTT) Index 127

Latest RTT: 0 ms

Latest operation start time: 17:21:36.676 MST Wed May 28 2008

Latest operation return code: Socket receive error

Latest DNS RTT: 0 ms

Latest TCP Connection RTT: 0 ms

Latest HTTP Transaction RTT: 0 ms

Number of successes: 0

Number of failures: 13

Operation time to live: Forever

#sh ip sla monitor reaction-configuration 127

Entry number: 127

Reaction: verifyError

Threshold Type: Immediate

Threshold Count: 5

Threshold Count2: 5

Action Type: Trap only

Reaction: connectionLoss

Threshold Type: Immediate

Threshold Count: 5

Threshold Count2: 5

Action Type: Trap only

Reaction: timeout

Threshold Type: Immediate

Threshold Count: 5

Threshold Count2: 5

Action Type: Trap only

Reaction: rtt

Threshold Type: Immediate

Rising (milliseconds): 5000

Falling (milliseconds): 3000

Threshold Count: 5

Threshold Count2: 5

Action Type: Trap only

1 Accepted Solution

Accepted Solutions

To get traps, you need to config:

snmp-server enable traps rtr

or ipsla depending on what CLI your IOS supports. You will also have to send these traps to your NMS.

View solution in original post

4 Replies 4

Joe Clarke
Cisco Employee
Cisco Employee

A socket recv error is treated the same as a timeout in the code. So your timeout reaction should get tripped when this error occurs. If the trap is not triggering, you may have a bug. What version of IOS?

Thank you for your swift reply, jclarke.

I am running 12.2(31)SB8, service provider ssh.

I have reverified my configuration and it seems to be OK.

I am receiving:

rttMonCtrlOperTimeoutOccurred.160:= True (1)

in my SNMP management station.

I suppose I was expecting syslog messages for these failed SLA events. I do have ip sla monitor logging traps configured, but apparently all I get is syslog.

In any case, my alert is working, I was just blind to the results.

Thank you for your time.

To get traps, you need to config:

snmp-server enable traps rtr

or ipsla depending on what CLI your IOS supports. You will also have to send these traps to your NMS.

I don't think I explained this properly -

When I first did this post, I was looking toward syslog (show log) for verification of any failure events.

SNMP was always working for me, I just wasn't looking in the right place on my NMS.