Comments on: What’s coming in 2.0

By: Chris Buechler

Chris Buechler — Wed, 03 Mar 2010 21:36:50 +0000

Francis: that’s not theoretically, or otherwise, possible because of the way the Internet works. Post to the forum or mailing list for more in depth discussion

By: Francis

Francis — Mon, 01 Mar 2010 14:43:24 +0000

hello there, my question deals with aggregation and is asked with total respect to the programmers as I know nothing about programming. That being said is this possible and work.

when a request to the internet reaches pfsense, pfsense uses 1 wan to get the size of the page or whatever from the page server. Then pfsense splits the reply in half and requests a half from each of the wan’s, gets it and sends the page to the requesting computer.

since everything dealing with the internet deals in packets theoreticlly this is possible, but is it practical in real life? is it something that cam be programmed?

Thanks for the info

By: Chris Buechler

Chris Buechler — Thu, 18 Dec 2008 03:07:18 +0000

Al: the load balancer in 1.3/2.0 has already been replaced with relayd and does what you mentioned.

By: Al

Al — Thu, 18 Dec 2008 02:43:29 +0000

Totally love this firewall! Rock solid. Multi-wan is awesome. Carp is great. Failover is a dream!! Can’t say enough positive as a firewall.

BUT – Also agree with above RE: Load Balancer monitoring…

I have not been able to use the built-in load balancer in PFSense because it lacks customized monitors. I know there are solid open-source tcp monitor packages around, was hoping to see this added. Better yet, a little http get monitor with text string evaluation would do it… Many times a web server is UP on tcp, but down for HTTP GET on 80…

Adding this would mean many of us could ditch two boxes entirely (Master LB and Failover LB) and use PFSense for the whole thing…

What a dream that would be!

By: Chris Buechler

Chris Buechler — Sat, 20 Sep 2008 16:58:06 +0000

We have no plans to implement a “clientless” (which is marketing BS) or “browser-based” (also marketing BS) SSL VPN for two reasons.

1) there isn’t a good open source one.
2) Reasons explained here:
http://article.gmane.org/gmane.comp.security.firewalls.pfsense.support/14336/

By: Fred Stephani

Fred Stephani — Sat, 20 Sep 2008 15:16:30 +0000

Are there any plans to implement a browser based SSL VPN solution in 1.3?

You are all doing a great job with this project, I am a huge fan of pfSense.

By: A.I.

A.I. — Tue, 12 Aug 2008 03:31:57 +0000

Tried the alpha-alpha release. Could not establish an IPsec VPN with older 1.2 version. PPTP server seems to always check the radius option even though I repeatedly un-check it. Interface looks great! The rate limiter option could use a download/upload perspective instead of the src/dst address. Do we need to add 2 rules there, one for upstream, one downstream? Overall it looks mighty fine!!

By: Chris Buechler

Chris Buechler — Thu, 19 Jun 2008 22:20:55 +0000

JBanks: we’re not secretive about a road map, we don’t have a formal one. The primary reason is we have no system to easily do so – this is being covered as part of the git conversion, Redmine which will replace cvstrac gives us facilities to put together development road maps. Look for one after the git conversion is completed.

By: JBanks

JBanks — Thu, 19 Jun 2008 06:49:40 +0000

Is there a reason the pfSense group/project is so secretive about the next product roadmap – seriously? I see so many of the SAME questions by numerous people and its always the same. Most other projects have “some” sort of ETA; organized ones anyhow. For some people, this product is what they based their IT decisions on – Vyatta or pfSense? Untangle or pfSense? SmoothWall or pfSense? etc etc etc…

By: Chris Buechler

Chris Buechler — Thu, 12 Jun 2008 04:21:40 +0000

I’ll get a development update post up probably this weekend. It works, it’s been working for a while, snapshots are building, we’ll probably let it out to the general public soon. We don’t have time to deal with the onslaught of bug reports right now (the majority of which end up being misconfiguration, but take significant time to investigate), and a number of changes are in process at the moment.