Slave router not taking over and IPv6 problem (bugs after todays update)
I've just updated my slave router and it now runs:
built on Sat Jun 2 14:57:29 EDT 2012
Everything was looking fine, so after some time I've upgraded my master router. While upgrade went fine I had short network breakage because slaver router switched to master as it normally should (and was doing). It just came up with this error:
[03-Jun-2012 17:18:43 UTC] PHP Fatal error: Call to undefined function return_gateway_groups_array() in /etc/inc/interfaces.inc on line 4774
So now I have two routers but, failover doesn't work :-(
Additionaly after update my IPv6 gateway is permanently offline. Nothing in logs except:
apinger: ALARM: WANGWv6(my_ipv6_here) *** down ***
I was previously running 2.1-DEVELOPMENT from May 1st and everything was working ok.
I hope these issues will be fixed ASAP. I'll look deeper into IPv6 problem tomorrow.
That error message is limited to the rc.banner (menu) script and fixed since.
The backup router should drop back really. It seems to do so for us.
Are complaining that both members are master?
No. When I'm for example rebooting master router usually slave will immediately pick up that master has failed and take over keeping my network up. After todays update when master router goes down slave router just shows this error and doesn't take over (fully or partially, can't tell exactly). My network is down until master router goes up again.
I've just noticed also that my OpenVPN no longer works. I'm using radius to authenticate users and while radius accepts users and sends accept request back router still says "authentication failure" and this appears in OpenVPN log:
WARNING: Failed running command (–auth-user-pass-verify): external program exited with
error status: 255
Going back to old version right now. Current snapshot is totally broken for me.
thanks, we'll have to look into it. We somehow broke PHP causing pfSense to fail in spectacular ways.
Reverting this change made most of things working again:
Yeah, we found that out too. We fixed the underlying issue.
Great. When this fix will be available in snapshot?
I hope so, the next?