if_pppoe problems with php-fpm causing loops. (resolved)
-
@stephenw10 said in if_pppoe problems with php-fpm causing loops. (resolved):
Ok I think we have a fix for this, we are testing it internally. It should be in the next 25.11 build.
BTW. Anything new in this matter?:
https://redmine.pfsense.org/issues/16229
-
Not yet. The issue is more complex that initially thought. Enabling the pppoe encapsulation there prevents it crashing out but it never matches any traffic.
-
@stephenw10 said in if_pppoe problems with php-fpm causing loops. (resolved):
Not yet. The issue is more complex that initially thought. Enabling the pppoe encapsulation there prevents it crashing out but it never matches any traffic.
Thanks for the Info & explanation!
-
@stephenw10 said in if_pppoe problems with php-fpm causing loops. (resolved):
Ok I think we have a fix for this, we are testing it internally. It should be in the next 25.11 build.
Thank you very much for this. Is there a possibility of a 2.8.2 for this?
I think j I found the patch here, so I can manually add to system patches I guess.
https://redmine.pfsense.org/projects/pfsense/repository/2/revisions/63566d0b473c6399376de6b97e0e06b9a6fb2550
Also on the MPD thing, it has its own build in tcpmssfix, which I learnt about when examining the scrub changes made for if_pppoe, as you guys replaced that feature with scrub for if_pppoe. From what I can tell it behaves absolutely fine on if_pppoe, but when doing some tests on iperf3, I was seeing some weird results with mpd on the negotiated mss whilst its tcpmssfix was enabled. Which I think is why if_pppoe fixed my fragmentation issues. This becomes irrelevant anyway as if_pppoe is the future. :)
Since its not a kernel change, perhaps it can be added to system patches package for CE? I have added patch manually, but it will be useful to others.
-
Yup might do that if the next CE release looks too far out.
-
@stephenw10 Had a brief ISP outage and if_pppoe auto recovered. :)
