Accel-ppp services should not use all CPU cores to process requests.
After this commit accel-ppp services use all available CPU cores to process requests from the subscribers (establish/update session/etc).
During mass connection of sessions, this can lead to the fact that it utilizes all CPU, and for other services like FRR, there is not enough CPU time to process their own stable work.
It causes watchfrr cannot to communicate with the required daemons like zebra and after 90 seconds it kills all routing daemons.

vyos@vpp# cat /run/accel-pppd/pppoe.conf | grep thre
thread-count=2
[edit]
vyos@vpp# 
[edit]
vyos@vpp# run show system cpu 
CPU socket: 0
CPU Vendor:       AuthenticAMD
Model:            AMD Ryzen 5 3600 6-Core Processor
Cores:            2
Current MHz:      3593.246
[edit]
vyos@vpp#

It should be configurable option or use by default not all CPU's