Bug #3209
closed
editing VLAN tag on OPT interface changes LAN interface instead
100%
Description
Changing the VLAN tag on my [renamed] OPT interface from "3" to "202" (to accommodate a change my upstream made) actually changed the VLAN tag on the [renamed] LAN interface instead.
I verified this in /conf/config.xml on the console, and fixed it by editing /conf/config.xml and rebooting.
(On a related note, the Developer Console appears to be non-functional now. Is there another way to reload config from the console?)
I'm going to re-test now just to make sure this isn't a PEBKAC issue.
Updated by Adam Thompson about 11 years ago
Sadly, I've just confirmed this... and also confirmed that IT DOESN'T HAPPEN EVERY SINGLE TIME! WTF?!?
I changed OPT1/RD_LOM_DIST from 202 back to 3 - no problem.
I changed OPT1/RD_LOM_DIST from 3 back to 202 - no problem.
I then changed OPT1/RD_LOM_DIST from 3 to 203 - and blew away my LAN/HOMEPRIVATE interface.
In config.xml, my LAN interface got changed from lagg0_vlan4 to lagg0_vlan203 while OPT1 remained at lagg0_vlan202.
Things I'm doing differently: all my interfaces are VLAN subifs of lagg0.
I can attach my (working) config.xml if needed - once I figure out how to strip passwords. Or I can send it directly to someone if needed, as-is.
Updated by Shahid Sheikh about 11 years ago
I am seeing this too. My workaround has been to fix the assignment from the Interfaces: Assign network ports: Interface assignments tab. As far as I can tell, its always the last two entries that get incorrect assignments.
As far as it not happening every single time - to me it was happening every time when creating new VLANs.
And if you are not paying attention and reboot the firewall then the firewall doesn't come up because of interface assignment mismatch. It sits at bootup at configuring interfaces.
Updated by Adam Thompson about 11 years ago
Not verified (and I'm not going to break this firewall again just to check it), but I also believe that at some point, editing one of the VLAN subinterfaces changed the LAG members of the parent interface; during debugging, I noticed that one reason my LAG wasn't coming back up was that only em0 was a member. I am 100% certain that em1 was also a member when I started testing, as I had verified LACP partner status on both links from the switch.
Updated by Phillip Davis almost 9 years ago
This looks like it has been just a plain old dumb bug - https://github.com/pfsense/pfsense/pull/2601
Updated by Phillip Davis almost 9 years ago
- Status changed from New to Feedback
- % Done changed from 0 to 100
Applied in changeset 1c32fb7e988fae89cf2da474778f39bcdd8a8656.
Updated by Chris Buechler almost 9 years ago
- Status changed from Feedback to Resolved
- Target version set to 2.3
- Affected Version changed from 2.1 to All
Thanks Phil! never happened to hit that circumstance in trying to replicate this, but that's clearly the issue.