Wizard error

zcache

Hi all,
I got this message on GUI first page (after login page), anybody face with this??

snapshort pfSense-2.1-DEVELOPMENT-4g-i386-nanobsd-20120610-1718.img
alix3d2 system board

zcache

try with new snapshort, problem still remain. any idea??

pfSense-2.1-DEVELOPMENT-4g-i386-nanobsd-20120611-2204.img
Alix 3d2 system board

jimp

Do you have any packages installed that might be altering functions.inc? A clean functions.inc from the repo does not contain two definitions for that function, and that second line number is near the end of the file.

https://github.com/bsdperimeter/pfsense/blob/master/etc/inc/functions.inc

zcache

no any package installed in it, pure clean install not from update. after i login in gui mode it will show wizard page and look like redirect to other page and that error show up. will capture in vdo after moment in new snapshot

//==== Add VDO with snapshort pfSense-2.1-DEVELOPMENT-4g-i386-nanobsd-20120612-0515.img ====//
http://youtu.be/GBDyTgtgmd8

PS. on case of no ip for wan will take more long waiting time…

phil.davis

I just flashed a CF with the latest image and loaded it on an Alix 2D3:
2.1-BETA0 (i386)
built on Tue Jun 12 15:14:41 EDT 2012
FreeBSD 8.3-RELEASE-p3

There were no problems using the initial wizard or logging out and in afterwards.
/etc/inc/functions.inc only has 1 instance of get_menu_messages

I guess functions.inc has been edited somehow on the OPs system and a copy-and-paste done accidentally of get_menu_messages?

zcache

no body idea here???
by the way, today i tried to install release 2.01 and make an upgrade to 2.1 beta. all working correct and perfect…. no any error message..
:)

yon

you should use pfSense-Full-Update-2.1-DEVELOPMENT-i386-20120612-2233 or later version.

jimp

There have been exactly two changes to functions.inc in the past two years, and none of them were anywhere near the second line you received the error about.

Something definitely happened to just your system, and then your browser probably cached the error. This was never a problem for anyone else as far as I'm aware.

zcache

@jimp:

There have been exactly two changes to functions.inc in the past two years, and none of them were anywhere near the second line you received the error about.

Something definitely happened to just your system, and then your browser probably cached the error. This was never a problem for anyone else as far as I'm aware.

I think so that cause nobody said about that before :) I will try in many more… thank for share :))

jharr

zcache isn't crazy, I'm getting the exact same error message on a freshly baked CF card.

PHP's "cannot redeclare" errors will send you on a wild goose chase. What's really going on is that "functions.inc" is getting included twice and PHP's error reporting is less than stellar. Without a stack trace, this isn't obvious, but it's bit me hard enough times to make me notice the pattern.

I'm running "pfSense-2.1-DEVELOPMENT-1g-i386-nanobsd-20120627-1609.img.gz" on an ALIX 2d2 and this has been happening to me for the past 3 or 4 builds I've tried. What's strange is that the 20120627-1609 OVA image doesn't have any of those errors in it.

I "fixed" mine with a console sledge hammer:


mount -o rw /
cd /usr/local/www
sed -i -bak 's|require(|require_once(|' `grep -rl 'require(' .`
mount -o ro /

I'll see if I can dig into the problem this weekend to narrow down what's happening, but I'm wondering why the builds aren't turning out exactly the same for portable code portions like PHP.

Discussing the merits of the *_once functions might also be prudent, but probably not best suited for this thread. I'll post my findings this weekend and then maybe open that can of worms.

phil.davis

get_menu_messages is the first proper function in functions.inc - so if functions.inc is getting require'd twice somehow, then that would be the first hit on a double declaration.
I just fired up a fresh 2G CF card on an Alix 2D3:
2.1-BETA0 (i386)
built on Mon Jun 25 16:02:04 EDT 2012
The wizard ran fine.
zcache had a 4G CF, jharr has a 1G CF, I have a 2G.
Surely it can't be an issue with the 1G and 4G images but not with the 2G!
I have an Alix 2D3 (3 ethernet ports - LAN, WAN, OPT1), the others have Alix systems with 2 or 1 ethernet port - how can the way the wizard selects the default hardware for LAN and WAN be much different so it does some weird sequence that induces this on on 1 or 2 ethernet nanobsd systems?
Has anyone else seen this, and if so, on what hardware and software combination?
Does it happen every time you start the wizard on a fresh install on your hardware? (reproducible or hit and miss?)
Does it happen right at the start of the wizard, just after login? (From the posts I think yes)