vThunders setup Health check for 443
jferr
Member ✭
Hello,
Just getting my feet wet with this as we have few sets of A10 vThunders on 4.1.4-GR1-P6 Build 90 and looking to setup a Health check so that if 443 is not responsive that A10 knows to pass traffic to the other server and not the one that is down on 443.
I created a Health Monitor for port 443 HTTPS and have defaults set. Is how I am thinking this works correct? Or what other setting do I need to make to have the A10 stop using this server for traffic and use the other once since 443 is down on one of them?
We have a need where we have a 2 server pool for 443 and those servers have a few things running. If one of the servers suddenly has an issue with the service that is using 443, i'd like the A10 to be able to say, "hey look this isn't working for the last 30 seconds lets force traffic to the other server and not this one until we detect its back online" We have them setup in Round robin at the moment with no health checks.
Any help is appreciated!
Comments
Your understanding is correct, and we need to configure L7 health check to provide the failover intelligence. By default, a SLB server health check is "ping" and defined ports such as TCP 443 are checked by "3-way TCP handshake." Configuring L7 health check provides more confidence that the service is up and able to process traffic requests or that the service is down, and we must failover.
If the server resource hosts multiple services on the same port, unique health checks may be configured at the service-group level for a more granular approach.
If you notice sessions getting stuck on a down service port (server is responding, but the port configured i.,e 443 is down) you can use this:
slb template tcp test re-select-if-server-down