> On 14 Jan 2019, at 18:44, Dave Cottlehuber <[email protected]> wrote:
> 
> 
> 
> 
>> On Mon, 14 Jan 2019, at 17:15, Pete French wrote:
>> So, until the middle of this afternoon I was, doing my load
>> balancing using> relayd from ports and PF. My own fault for not checking, 
>> but I
>> upgraded> one of the firewall pair to 12 and then discovered that the
>> relayd port is> no 
>> Am now puzzling over solutions to this - I dont really want to stay on> 11 
>> forevere. Moving to OpenBSD to get their PF and relayd is a bit of> an 
>> uncomfortable idea as we gain a lot from having one OS
>> everywhere that> people know, so does anyone have any suggestions ?
>> 
>> PF round robin is not good enough for this as I have some dynamic
>> problems> which indicate when a node is up or down. Relayd will check
>> these, but the> basic PF wil not as far as I know.
>> 
>> What do other people do ?
> 
> haproxy does proper failover and allows custom health checks either via
> URL or real world traffic of external scripts. Traefik has lots of
> container oriented features.
> Dave

There’s also the very venerable (hence reliable) HTTP proxy/load balancer, 
Apache Traffic Manager, https://trafficserver.apache.org

- Mark
_______________________________________________
[email protected] mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "[email protected]"

Reply via email to