Ace loadbalancer alerting (4710 standalone unit)
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-26-2013 05:10 AM
Hello
I wandered if someone can help me with this query.
If i set up a VIP with rate-limiting, i want to be able to send the ops a team an alert to say the threshold has been reached, is this possible by syslog or snmp traps? Or what is the best way to achieve this?
Thanks
- Labels:
-
Application Networking
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-26-2013 07:48 PM
Hi,
not sure of VIP rate-miting. but if it related rserver rate-limiting then when a rserver reach its limit its status change to "MAX-CONN" and a syslog is generated
%ACE-4-442003: Real Server real_server_name in serverfarm sfarm_name changed state to new state
you can configure this to be forwarded to your external syslog server
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-27-2013 08:31 AM
Hi
Thanks Gaurav, below is the two configs rserver or VIP rate limiting, it similar config and hoping it does generate a log, guess i will have to test this out and find out. Do you have to configure any additional or it will just send the syslog message if syslog server is configured?
Also once the threhold is reached, all other traffic is discarded? How does the server reset it status to recieve traffic again?
rserver based rate limiting
serverfarm host api-farm
rserver api-server
rate-limit connection 300
Vserver based rate limiting
parameter-map type connection api-map
rate-limit connection 300
!
policy-map multi-match vlanx-vips
class VIP80
connection advanced-options api-map
Thanks
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
06-29-2013 02:00 AM
My mistake, i confused rate-limit with conn-limit and provided that option.
unfortunately when a rserver reacheds to its rate limit, its states do not change so the above mentioned syslog option will not work.
only way of verifiying is to mointor the " out-of-rotation count " under show serverfarm detail. this counter will increase whenver limit is reached. I think we do not have any syslog or snmp trap option to notify whether limit is reached or not.
