Bug #4674
closed
invalid state table entries after WAN IP change
Added by Daniel Haid over 9 years ago.
Updated almost 5 years ago.
Description
This is similar to Bug #1629. I have a SIP client behind pfsense 2.2.2. When the WAN IP changes, there is a state table entry with the old IP, and the SIP client can not register anymore.
I suspect that this happens because the SIP client continuously generates packets, so that one could arrive between pfSense_kill_states and filter_configure_sync. My patch reverses the order of the two functions and seems to work, but I do not know whether this will introduce other problems.
Files
Interesting workaround! I will have to try this myself as we've had similar problems with SIP devices & Asterisk.
Can someone look into this? Sounds to me like the ordering here is indeed just wrong.
SIP behind pfSense with changing WAN IP address is with this bug impossible.
I must delete every day the old states.
- Category set to Rules / NAT
- Assignee set to Jim Pingle
- Target version set to 2.5.0
Looking at /etc/rc.newwanip it does appear to make more sense to configure before killing the old states.
- Status changed from New to Feedback
- % Done changed from 0 to 100
- Target version changed from 2.5.0 to 2.4.5
- Status changed from Feedback to Resolved
No feedback from OP or previous commenters, and at least with a quick test here it appears to be doing the right thing now compared to the previous code.
Can revisit if people still encounter issues.
Also available in: Atom
PDF