iRule or something for 'timing' load balancing

You approach is not leverageing load balancing redundancy, if one of your server goes down then the service is down. It also lets the good server on a break while it can participate to the job (7pm to 7am).

Another approach is if one server is performing well and the second does not then you would rather set the first to work all the day 24x7. The second would work 12x7 in off peak hours. Here you are losing load balancing in peak hours so if server 1 goes down the service will go down.

A third approach which seems more robust and easiest to setup: use priority group activatio with the value "less than 1". server 1 will be your primary server and will handle all traffic all the time, but if it goes down server 2 takes over. This is a typical active/passive scenario, no load balancing but redundancy guaranteed.

With that said, if you chose the timing approach What you could do is to associate a regular monitor to each member plus an external monitor to each member you want to control, this external monitor will check the current time and make the member up or down depending on the local time. This involves some scripting. More on this here: https://devcentral.f5.com/articles/ltm-external-monitors-the-basics

If you have the courage, you can chose a load balancing method (Dynamic Ratio load balancing) wich checks the sever performance using WMI or SNMP: https://support.f5.com/kb/en-us/products/big-ip_ltm/manuals/product/ltm_configuration_guide_10_0_0/ltm_appendixb_monitor_considerations.html1172375