All is well now.

Thanks

@Renato:

New PHP packages, 5.6.26_1 are being built right now. They will fix all these issues with extensions

Thank you so much!

Running the following commands resolves the problem:

pkg update pkg upgrade -f php-suhosin pkg upgrade

The patch essentially removes the extensions.ini file and replaces it with individual module files so that pfSense can explicitly control the php load order:

[2.3.3-DEVELOPMENT][root@pfSense.pf.lan]/~/: ls -l /usr/local/etc/php total 156 -rw-r--r--  1 root  wheel  26 Oct  7 07:35 ext-10-opcache.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-18-session.ini -rw-r--r--  1 root  wheel  20 Oct  7 07:35 ext-20-bcmath.ini -rw-r--r--  1 root  wheel  17 Oct  7 07:35 ext-20-bz2.ini -rw-r--r--  1 root  wheel  19 Oct  7 07:35 ext-20-ctype.ini -rw-r--r--  1 root  wheel  18 Oct  7 07:35 ext-20-curl.ini -rw-r--r--  1 root  wheel  17 Oct  7 07:35 ext-20-dom.ini -rw-r--r--  1 root  wheel  20 Oct  7 07:35 ext-20-filter.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-gettext.ini -rw-r--r--  1 root  wheel  18 Oct  7 07:35 ext-20-hash.ini -rw-r--r--  1 root  wheel  18 Oct  7 07:35 ext-20-json.ini -rw-r--r--  1 root  wheel  18 Oct  7 07:35 ext-20-ldap.ini -rw-r--r--  1 root  wheel  22 Oct  7 07:35 ext-20-mbstring.ini -rw-r--r--  1 root  wheel  20 Oct  7 07:35 ext-20-mcrypt.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-openssl.ini -rw-r--r--  1 root  wheel  19 Oct  7 07:35 ext-20-pcntl.ini -rw-r--r--  1 root  wheel  17 Oct  7 07:35 ext-20-pdo.ini -rw-r--r--  1 root  wheel  21 Oct  6 09:27 ext-20-pfSense.ini -rw-r--r--  1 root  wheel  19 Oct  7 07:35 ext-20-posix.ini -rw-r--r--  1 root  wheel  20 Oct  6 09:21 ext-20-radius.ini -rw-r--r--  1 root  wheel  22 Oct  7 07:35 ext-20-readline.ini -rw-r--r--  1 root  wheel  17 Oct  6 09:28 ext-20-rrd.ini -rw-r--r--  1 root  wheel  19 Oct  7 07:35 ext-20-shmop.ini -rw-r--r--  1 root  wheel  23 Oct  7 07:35 ext-20-simplexml.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-sockets.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-sqlite3.ini -rw-r--r--  1 root  wheel  18 Oct  6 09:21 ext-20-ssh2.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-sysvmsg.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-sysvsem.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-20-sysvshm.ini -rw-r--r--  1 root  wheel  23 Oct  7 07:35 ext-20-tokenizer.ini -rw-r--r--  1 root  wheel  25 Oct  6 09:20 ext-20-xdebug.ini -rw-r--r--  1 root  wheel  17 Oct  7 07:35 ext-20-xml.ini -rw-r--r--  1 root  wheel  23 Oct  7 07:35 ext-20-xmlwriter.ini -rw-r--r--  1 root  wheel  18 Oct  7 07:35 ext-20-zlib.ini -rw-r--r--  1 root  wheel  17 Oct  6 09:26 ext-20-zmq.ini -rw-r--r--  1 root  wheel  24 Oct  7 07:35 ext-30-pdo_sqlite.ini -rw-r--r--  1 root  wheel  21 Oct  7 07:35 ext-30-suhosin.ini -rw-r--r--  1 root  wheel  23 Oct  7 07:35 ext-30-xmlreader.ini

Perhaps, ultimately, some sort of notification from the backend conf file generation. Notification could display on the GUI and, if you have it set up, send a notification as well.

Yes, if the backend "fixes" something on-the-fly to avoid trouble, then it should make a notification so that someone will see it one day and follow it up.

Went to GitHub again and branches are now synced correctly. Thanks! :)

Can someone with access trigger a new snapshot generation task? Snapshot timings are out of sync (i386 vs amd64 - about 8hrs difference). I'd like to test the new changes on both architectures on my Intel P4 and Intel i5 systems.

Thanks again!

Edit: Ok now.

I figured it was harmless since everything appears to work. I believe it was caused by me trying to set a data limit on the kids. That didn't work and this appears to be the remnants of the attempt.

There is still an inconsistency, source hash key field is shown when it shouldn't.

PR for v2.4-dev (master) and v2.3.3-dev (RELENG_2_3):
https://github.com/pfsense/pfsense/pull/3174

PR for v2.3.2 (RELENG_2_3_2) - Clean up unused code:
https://github.com/pfsense/pfsense/pull/3175

Regards,
Jorge M. Oliveira

Edit: Important stuff has been dealt with. Thanks!

This could be implemented in RELENG_2_3_2 but is not important as RELEASE versions ignore PHP warnings and don't generate crash reports about it.

This is important to be applied in RELENG_2_3_2, so +1 from me.

Normalizing date formats is a very laudable idea. But please stay with ISO 8601 formats. No slashes.

2016-09-30

Should be caused by simultaneous events. A latent problem in pfSense is that it triggers the same events multiple times in certain cases.

This code could be silenced by adding a '@' before the unlink() instruction.

+1 for this feature!

I'd lean toward keeping the name/system/version/serial and such together and visible, and split the stats out to a separate widget (state table size, mbuf, cpu, mem, swap, disks). As long as the default config has it enabled so they show one right under the other, and upgraded configs get both, it seems like it might be a good place to split in two.

The core of this issue was the OpenSSL library problem that has been fixed. There is a separate issue where certain processes crashing can cause failures of the boot script as a whole, but it's not directly related. PHP itself was also crashing out, but neither PHP nor OpenVPN are crashing now like they were when the original problem happened.

It would be great if the process was a lot more robust, with exception handling everywhere to ensure errors are trapped and handled gracefully. That's a much larger task though, not one we're likely to see even in 2.4 at this point.

Also the secure smtp connection modes mutually exclusive fix has security implications (security bug).