Le 17 sept. 2013 à 00:32, Olivier Mascia <[email protected]> a écrit :
>> I have been using 2.01 for about 2 years. >> Just upgraded to 2.10. >> This an amd64 full install. >> >> I’m seeing high-cpu usage (which was in the past < 1 or 2%) and I can >> further verify that /usr/local/bin/check_reload is eating one full core. >> There is no noticeable impact thanks to this being a multi-core system. >> What should I further check to narrow down the issue? > > I have since read that check_reload wouldn’t be the issue in itself, but the > consequence of some other problem. > The question is: which one? > What to check for? > > The purpose of upgrading from 2.0 to 2.1 was to activate our IPv6 native > connectivity from our provider (EDPNET in Belgium). > > For what it looks like, our IPv6 setup on this box, also doing NATed IPv4 is > quite correct, as seen from inside LAN workstations, both PC and Macs, that > can access IPv6 servers on the internet without issue - let’s cite > http://ipv6.google.com, but also http://ipv6.whatismyv6.com. > > Though, we lack past experience with pfSense IPv6 and some detail might well > be the trigger to this issue. > EDPNET ask for using DHCPv6, which we did. I dropped the Npt, using on the LAN interface the static subnet routed to us by the ISP through DHCPv6 WAN and it works nicely, high-cpu usage of check_reload disappeared. > There is a secondary issue, which might be linked to this one (don’t know). > On the Dashboard page, the OpenVPN widget keeps saying « [error] unable to > contact daemon » and this for all our 3 OpenVPN server definitions. Yet > those 3 VPN setups work without issue (can connect to them). This strange OpenVPN dashboard issue, is still present, but again, despite these errors those OpenVPN tunnels and remote access setups do work fine. __ Olivier Mascia integral.be _______________________________________________ List mailing list [email protected] http://lists.pfsense.org/mailman/listinfo/list
