I have to agree with the parent. The pfSense team's attitude seems to be that it must be the users fault for mis-configuring it. They tell you to go read the documentaion (which is frankly rubbish, especially on traffic shaping) or some useless forum posts.
The documentation is rubbish, I agree. Part of the documentation issue is that it's excruciatingly difficult to document a changing system. A large amount of the documentation that was created during the alpha and beta phases of development were rendered useless after various changes were made to either idiot-proof the system or eliminate bugs. I can't comment on the forum posts, although I'm well aware of yours. I never once thought that it was perfect, I've never claimed it was complete, I've never claimed it was "easy", I did however go out of my way to create code that worked out of the box for most people. Is it configurable...not so much, that's part of the "I never claimed it was complete". I'm probably the person that understands our shaper code the most, I also am the least likely to answer questions on it cause I'm tired of it, it's a part of the codebase I wish I'd never touched. I regret that we didn't leave the m0n0 code in there, it's old, sucks ass, and is nearly useless, but at least it's configurable.
PS. It usually is a misconfiguration issue and for the cases where it makes sense for us to put a sanity check in for the misconfiguration, or add a note to a field to clarify it's use, we do.
"why don't you submit patches?"
Because he found a more cost-effective solution from his point of view.
Great. So instead of helping out when the developers can't replicate an issue, he moves on to another project and then complains about it. Fine by me, open source only works when there's a community behind it. Those who complain about features that don't work quite right or don't work the way they want them to need to remember exactly how much they paid for it and exactly what the developers owe them. Nothing.
"This is an open source project"
So what? That only means that he *can* submit patches, not that he *must*.
So? I'm failing to see a point here. This is STILL a free project, he still hasn't paid a dime for it. He has a couple options, help the developers pinpoint the bug, submit a patch, or move on. He's chosen the latter path, fine by me. That's how this works.
"It's also a one dot oh release."
Which usually means "the first functionally complete, bug-free as-far-as-we-know release". You know, they themselves publish this release as "The pfSense team is excited to bring you our first ever real release!" . Which is quite far to be the current situation.
Actually, no, this means that it's as good as we can make it. Realize that software has bugs in it, some are known, some require changes that are large enough to not warrant fixing in a given release (changes can also introduce bugs). Being a 1.0 release means that up until this point, considering it's only ever been posted as a beta, a LOT of people won't have touched it.
I feel the main developers suffer from quite a strong "featurittis" that focuses them so much towards their new "HEAD" than to polish their current 1.0. Oh! and a bit of reality-negating attitude. Usually an "I can replicate that on my environment" (or even an "I don't want to take the time to test it now") tends to become "You must be wrong: it surely works properly" which in turn tends to mean that not so easy to catch bugs (like those on NAT and balancing, for instance) just go through untouched since months ago.
We spent a YEAR polishing 1.0. Some things were pushed to HEAD out of necessity as previously mentioned, some bugs are just too damn big to squash easily - for those we either provided workarounds where applicable or removed the feature entirely (and people whine). Lots of work has been done in HEAD sure...why? Because stiffling development to fix the trickle of bug reports on RELENG_1 would mean a loss of developers. Our last hackathon was spent testing the codebase and working on bugs, not on developing new features. Reporting bugs without enough information to duplicate the bug ends up with no fixes - if it can't be duplicated or we don't have enough information to trawl the code path for the bug, it's not going to get fixed.
This has been gone over numerous times in both the pfsense forums and the mailing lists. The short answer is hardware support and that bsdinstaller is only available on freebsd and dragonfly at this time.
Yeah, especially when the studio is owned by a parent company that has other subsidiaries that produce said DVD players. Think Sony here...screw it, if it won't play in my player, I'll be first in line on the following class action lawsuit. I refuse to buy another DVD player when the one I currently have plays every other movie I want to watch.
No kidding. So let's see, this makes obsolete:
$1200 laptop (no removable DVD drive)
$1400 laptop (ditto)
$800 PC (unless my manufacturer replaces the drive w/ one that LOOKS like it belongs in the machine - all aftermarket hardware is attached USB and firewire)
$800 PC (ditto)
$100 DVD burner
$250 DVD player
I'm in for a $4500 check. Oh and I guess all my old DVDs (some of which I still haven't watched!) need to be replaced as well as presumably data DVDs that I've burned. Since my time has no value (according to my employer) I won't bill for the replacement time.
You don't think NASA's engineers are smart enough to calculate the pressure at mach 10 and build accordingly?
I guess that all depends on if they've decided on the same type of units this time. If so, then yes. But if half the team uses Metric and half the team doesn't, that could be one interesting plane!:)
And I can buy a 2.4Ghz 10Watt amp from Canada and exceed the TX power on 2.4Ghz restrictions the FCC has in place, but I guess that would be my own damn fault wouldn't it? Oh wait, that's right, it's also my fault that I changed the code on a well written driver to up the output power, or change the frequency cause I'm sick and tired of my phone stomping on my network.
As for the "bits of licensed code in the driver"...um, we're talking about firmware, not drivers (at least not OS level drivers). But maybe you were replying to another poster, not the original post.
I have a co-worker that once had an emergency and had his doctor paged. The idiot had decided to take a golfing trip three states away and was PISSED that he had to return. Teaches him to leave the area when he's on call:)
I want whatever you're smoking. I think you're mixing up your terms.
A Checkpoint/NOKIA solution may be software, but it does NOT run on your web server, it runs between your network ingress point (a router of some sort) and the rest of your network. Well, put simply. IE. it runs on another (dedicated)box.
As for something closer to a hardware solution, vs. the software running on a general purpose OS (but still dedicated hardware) check out NetScreen.
FWIW, I can't recommend either vendor, I think they both suck pretty bad.
So that would be a random WEP key they default to right? Encryption is pointless if the key is the same across the board. Note, I'm talking out my ass here, I know nothing about the Airport, but enough about wireless to know that the WEP key still needs to be changed! Or are they doing EAP/LEAP/someother fucked up method of logging into the AP and changing WEP keys every 5 minutes?
No, see clearly it's a ploy to get your address from the warranty card you sent in. "Look Joe, another sucker registered his MP3 player today - let's send a C&D letter to him!"
Seriously though, it's just a way to make money from both ends, even if they lose, they win. And if they win, they have your name and address (you did want that $50 rebate right?)
Right, the place I work at is still on NT4 workstation while our servers are a mix of NT4 and 2k. We're just now starting to roll out XP company-wide. Granted, this is desktop, not server, but the reasoning behind such a large lag on the desktop is that we have over 600 applications company-wide that need to be re-certified when the OS changes. Servers tend to run a smaller software set so require less testing. Regardless, up until the last 6 months or so, NT4 was the defacto server build, 2k only got installed if the application required it.
--Dox
Re:And still perl is a port now and java builds
on
FreeBSD 5.1 Released
·
· Score: 1
Proof in Pudding: Think of heavy iron appliances with various free operating systems in it. I can think of two for FreeBSD. The godly Juniper routers and the F5 BigIP. These are serious pieces of networking equipment and they chose FreeBSD for a reason - its far more pleasant to deal with commercially, its fast stable and coherent and the license permits modifications without divulging them to the world.
Hrm, actually, the BigIP versions 4.0-4.5 run on BSDi. F5's 3DNS version 3.x ran on FreeBSD, but they migrated it to BSDi in version 4.0. ISMan might run on FreeBSD though - it comes as an ISO.
So this isn't really a thing like gnutella. It's an enterprise product.
10-50 users makes this an enterprise product??? I work in a SMALL enterprise that has 15,000 employees, the IS department alone is over 1000 of those, and the IS Support staff (the ones that handle all the user complaints - on the order of 100 employees worldwide) currently use an internal IRC server as our enterprise IM software (Sametime) didn't exist when they put out the IRC solution. We'll probably end up moving to use Sametimes meeting features instead of IRC in the future. My personal sametime address book has 40 or so people that I regularly communicate with at work; 10-50 wouldn't even begin to cut it. Now, I can see this being used to communicate with personal friends (assuming it can utilize HTTP proxy servers to get outside our firewalls).
--Dox
5. I personally think it's down to what game you are planning to write. If you want to do Doom III in.NET, you'll probably have to wait till computers with 5GHz processors and 2GB memory are common. Then again, everybody will agree with me that certain games just have to be written the hardcore C/C++ way. So what's the point of the question?
So, uhhh this would be like next year right? Sounds reasonable to me seeing as it's gonna take that long at least to write the code:)
--Dox
Sadly, will have to stick to FreeBSD till nVidia opensources the Geforce drivers.
Why? I'm running OpenBSD with a Geforce 2 and X, works fine. Are you trying to run dual head (Twinview I think is the _only_ reason you'd need nvidia drivers)?
--Dox
And in fact a large number of corporations block Flash for security reasons. I for one work in an environment where anything inside the OBJECT tag is stripped and removed. This was done initially to remove ActiveX for security reasons, however in light of more recent security issues surrounding flash and other activex plugins it seems it was a good tactic.
There are _way_ to many sites that use flash and don't have other ways in.
Imagine playing Quake against a guy that only had to move his mouse.451mm to pull a 180 (nm the fact that he wouldn't be able to aim!). Try hitting that!
Slashdot Mirror
Great, we thank you for donating our sparc64 build systems.
This has been gone over numerous times in both the pfsense forums and the mailing lists. The short answer is hardware support and that bsdinstaller is only available on freebsd and dragonfly at this time.
Maybe you could convince the author of this:6 1
http://www.vmware.com/vmtn/appliances/directory/3
to update it to release.
curses interface? Are you sure you are looking at pfSense?
Yeah, especially when the studio is owned by a parent company that has other subsidiaries that produce said DVD players. Think Sony here...screw it, if it won't play in my player, I'll be first in line on the following class action lawsuit. I refuse to buy another DVD player when the one I currently have plays every other movie I want to watch.
No kidding. So let's see, this makes obsolete:
$1200 laptop (no removable DVD drive)
$1400 laptop (ditto)
$800 PC (unless my manufacturer replaces the drive w/ one that LOOKS like it belongs in the machine - all aftermarket hardware is attached USB and firewire)
$800 PC (ditto)
$100 DVD burner
$250 DVD player
I'm in for a $4500 check. Oh and I guess all my old DVDs (some of which I still haven't watched!) need to be replaced as well as presumably data DVDs that I've burned. Since my time has no value (according to my employer) I won't bill for the replacement time.
--Dox
Ummm, no it's not, I just tried, they answered! Of course they told me I had the wrong number, can't figure out why though.
--Dox
And I can buy a 2.4Ghz 10Watt amp from Canada and exceed the TX power on 2.4Ghz restrictions the FCC has in place, but I guess that would be my own damn fault wouldn't it? Oh wait, that's right, it's also my fault that I changed the code on a well written driver to up the output power, or change the frequency cause I'm sick and tired of my phone stomping on my network.
As for the "bits of licensed code in the driver"...um, we're talking about firmware, not drivers (at least not OS level drivers). But maybe you were replying to another poster, not the original post.
--Dox
I have a co-worker that once had an emergency and had his doctor paged. The idiot had decided to take a golfing trip three states away and was PISSED that he had to return. Teaches him to leave the area when he's on call :)
--Dox
I want whatever you're smoking. I think you're mixing up your terms.
A Checkpoint/NOKIA solution may be software, but it does NOT run on your web server, it runs between your network ingress point (a router of some sort) and the rest of your network. Well, put simply. IE. it runs on another (dedicated)box.
As for something closer to a hardware solution, vs. the software running on a general purpose OS (but still dedicated hardware) check out NetScreen.
FWIW, I can't recommend either vendor, I think they both suck pretty bad.
--Dox
So that would be a random WEP key they default to right? Encryption is pointless if the key is the same across the board. Note, I'm talking out my ass here, I know nothing about the Airport, but enough about wireless to know that the WEP key still needs to be changed! Or are they doing EAP/LEAP/someother fucked up method of logging into the AP and changing WEP keys every 5 minutes?
--Dox
No, see clearly it's a ploy to get your address from the warranty card you sent in. "Look Joe, another sucker registered his MP3 player today - let's send a C&D letter to him!"
Seriously though, it's just a way to make money from both ends, even if they lose, they win. And if they win, they have your name and address (you did want that $50 rebate right?)
--Dox
Ahh, but it's OK with Sun, they have a well known history of this.
SunOS 5.8 == Solaris 2.8 == Solaris 8
So really, which one is it?
--Dox
Right, the place I work at is still on NT4 workstation while our servers are a mix of NT4 and 2k. We're just now starting to roll out XP company-wide. Granted, this is desktop, not server, but the reasoning behind such a large lag on the desktop is that we have over 600 applications company-wide that need to be re-certified when the OS changes. Servers tend to run a smaller software set so require less testing. Regardless, up until the last 6 months or so, NT4 was the defacto server build, 2k only got installed if the application required it.
--Dox
Hrm, actually, the BigIP versions 4.0-4.5 run on BSDi. F5's 3DNS version 3.x ran on FreeBSD, but they migrated it to BSDi in version 4.0. ISMan might run on FreeBSD though - it comes as an ISO.
--Dox
--Dox
5. I personally think it's down to what game you are planning to write. If you want to do Doom III in .NET, you'll probably have to wait till computers with 5GHz processors and 2GB memory are common. Then again, everybody will agree with me that certain games just have to be written the hardcore C/C++ way. So what's the point of the question?
:)
--Dox
So, uhhh this would be like next year right? Sounds reasonable to me seeing as it's gonna take that long at least to write the code
Sadly, will have to stick to FreeBSD till nVidia opensources the Geforce drivers. Why? I'm running OpenBSD with a Geforce 2 and X, works fine. Are you trying to run dual head (Twinview I think is the _only_ reason you'd need nvidia drivers)? --Dox
All 0's worked too ;) It was a stupid math trick they did to validate numbers (something along the lines of a modulus 7).
--Dox
And in fact a large number of corporations block Flash for security reasons. I for one work in an environment where anything inside the OBJECT tag is stripped and removed. This was done initially to remove ActiveX for security reasons, however in light of more recent security issues surrounding flash and other activex plugins it seems it was a good tactic.
There are _way_ to many sites that use flash and don't have other ways in.
--Dox
Neat, I didn't know automobile was spelled with a 'v'!
--Dox
The possible words were:
car cars automovile
Imagine playing Quake against a guy that only had to move his mouse .451mm to pull a 180 (nm the fact that he wouldn't be able to aim!). Try hitting that!