2.0.2 ???

[ingmthompson]

@charliem:

I'd be happy to test the new ntpd aspects of 2.0.2 if any new 2.0.2 RCs happen to show up :)

There have been 2.0.2 RCs, they just weren't made public. Nor has the purported release made it to the download mirrors either.

Jim, when you talk about showstoppers, are you referring to 2.1 or 2.0.2 showstoppers? If issues with 2.0.2 are arising, then perhaps it's worthwhile putting an "RC" build of the current code out there and getting some testing feedback from users?

[jimp]

The last couple things were small fixes that were easy to confirm, but could have potentially been trouble in a release.

The testing we need to do for this is mostly things we need to confirm ourselves, making sure things work properly on specific hardware, install/upgrade tests and so on.

[tester_02]

This is a bad thread.  From reading this thread I would say there is a hidden agenda and totally opposite to the open source philosophy.
This is not saying that it is, but the responses from the dev's you can tell are not saying everything.

Please if you are not going to fully say what is going on, don't respond.  It's giving me bad vibes reading this, even though my personal belief is that it's just written out wrong.  My guess is that they were hoping for 2.1 to be released and not have to deal with 2.02.  2.1 is delayed and they are not sure if they even want to release 2.02.  Just my guess….
  Can we close this thread and move on.  Would have been better for people not to think a 2.02 was ever even coming (that's what I thought until this thread).

[ingmthompson]

@tester_02:

From reading this thread I would say there is a hidden agenda and totally opposite to the open source philosophy.

That's not true. The source code is still freely available from Github, and anybody with the means and time to compile the code themselves can have a copy of 2.0.2. I agree that this whole situation has a slightly odd sense to it, but considering that we're not paying for pfSense I'm not inclined to press the issue too hard, and I don't recommend anybody else do so either.

[jimp]

Read my explanation earlier in the thread. The builds are already tagged as release. We cannot do snapshots of such a build because clients would not recognize the "official" release when we post it. Yes it has taken longer than anticipated but it is just a point release. It contains some important fixes but nothing so critical as to warrant rushing it.

[jimp]

Had to make yet another new set of images again…

If you really want to test the images, they can be found here – but if you try them, do so knowing full well that you will need to manually do an upgrade to the official 2.0.2-RELEASE or subsequent image if this image is not the final one. If this is the actual release image, then no action will be needed. They are not yet signed, so the final hash values are also not available.

There is no way to use auto update to upgrade to these images, but if you use the console upgrade by URL function you can feed it the URL of a firmware update file that way.

[ingmthompson]

@jimp:

Had to make yet another new set of images again…

If you really want to test the images, they can be found here – but if you try them, do so knowing full well that you will need to manually do an upgrade to the official 2.0.2-RELEASE or subsequent image if this image is not the final one. If this is the actual release image, then no action will be needed. They are not yet signed, so the final hash values are also not available.

There is no way to use auto update to upgrade to these images, but if you use the console upgrade by URL function you can feed it the URL of a firmware update file that way.

Cheers Jim, much appreciated :) Going to give this a try now.

[chpalmer]

Just installed on one of my remote offices.  (no danger if goes down.)

MLPPP bonding 2 dsl connections.

Open VPN to primary data center.

Port forwards for backup email server, private company web server, and security camera's.

Upgraded just fine!

:)

[ingmthompson]

I'm having some issues with pfBlocker on this version. Is it just a matter of the package maintainers fixing it up for 2.0.2?

[phil.davis]

pfBlocker is working for me in 2.1-BETA0 as well as 2.0.1 - the same code, and I can't see any version-specific checks in the code. So it is a bit surprising that something happens with an in-between version like 2.0.2. What are the issues?
(if it's more than a quick issue and fix, then probably start a separate thread)

[ingmthompson]

@phil.davis:

pfBlocker is working for me in 2.1-BETA0 as well as 2.0.1 - the same code, and I can't see any version-specific checks in the code. So it is a bit surprising that something happens with an in-between version like 2.0.2. What are the issues?
(if it's more than a quick issue and fix, then probably start a separate thread)

Seems like a memory error similar to what you'd see when you forget to increase the number of firewall table entries, even though I did set that to about 2,000,000. I also don't see any rules when I visit the alias URL.

EDIT: The exact error message is below. Firewall table entries limit currently set to 10 million.

There were error(s) loading the rules: /tmp/rules.debug:22: cannot define table pfBlockerlist1: Cannot allocate memory
pfctl: Syntax error in config file: pf rules not loaded The line in question reads [22]: table <pfblockerlist1>persist file "/var/db/aliastables/pfBlockerlist1.txt"

I strongly suspect this commit to be the culprit.</pfblockerlist1>

[jimp]

Except the exact opposite would be true - that commit actually allowed the limits to be set, without that only the default values were being used.

Check pfctl -sm

[ingmthompson]

@jimp:

Except the exact opposite would be true - that commit actually allowed the limits to be set, without that only the default values were being used.

Check pfctl -sm

Here's the output:

$ pfctl -sm
states        hard limit   194000
src-nodes     hard limit   194000
frags         hard limit     5000
tables        hard limit     3000
table-entries hard limit   200000

This is with Advanced > Firewall/NAT set to this. Something's obviously up.

[jimp]

ok, found the problem. Had a couple other things come in overnight anyhow, will make a new set later today.

It's an easy fix if you want to do it manually…

diff --git a/etc/inc/filter.inc b/etc/inc/filter.inc
index 87401cc..d30ed24 100644
--- a/etc/inc/filter.inc
+++ b/etc/inc/filter.inc
@@ -257,7 +257,7 @@ function filter_configure_sync($delete_states_if_needed = true) {
        $limitrules .= "set limit tables {$maxtables}\n";
        /* User defined maximum table entries in Advanced menu. */
        if ($config['system']['maximumtableentries'] <> "" && is_numeric($config['system']['maximumtableentries']))
-               $rules .= "set limit table-entries {$config['system']['maximumtableentries']}\n";
+               $limitrules .= "set limit table-entries {$config['system']['maximumtableentries']}\n";

        if ($config['system']['optimization'] <> "") {
                $limitrules .= "set optimization {$config['system']['optimization']}\n";

[ingmthompson]

@jimp:

ok, found the problem. Had a couple other things come in overnight anyhow, will make a new set later today.

No worries, thanks for the prompt response :)

[ingmthompson]

@maverick_slo:

Hi!

Can someone on 2.0.2 run this command: mpd5 -v
And post output here?

Thanks a lot!

I'm getting a blank output.

[maverick_slo]

Hi!

Sorry to bother you all, I totally forgot I asket SAME question in other forum…
Sorry again, 2.0.2 uses 5.6 version of MPD5 package...

[ingmthompson]

Tested the new build today, it's working perfectly. Looking forward to an official release!

[Supermule]

NICE!!

[jimp]

Yes I meant to post about that new build last night. We had a couple issues come up (you can see the RELENG_2_0 commit log on github) that required some fixes, and I put in the limit fix for the table entries as well when I did that.