Post a Comment
It's nice to see that OpenBSD 4.0 supports some hardware that I was hoping to get in 2.6.18 Linux kernel. I thought VIA 8237A chipset will get into today's release. It's 5 lines of code that make existing drivers identify new piece of hardware and this patch was in -mm for 2 months already. I'm seriously disappointed.
Because I'm considering migration to BSD (probably FreeBSD) I must ask about one thing. What's the state of sound support in BSDs? I'm most worried about sound because my Audigy SE is barely working with ALSA (no input recording, no midi hardware synth :/).
The above information is incorrect. Alan Cox submitted support for the VIA chipset and it already in 2.6.18 kernel from rc-7 onwards and will be supported by FC6 being released next month.
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.gi...
http://fedoraproject.org/wiki/Core/Schedule.
The Linux kernel supports more architectures and drivers that any other operating system in the world according to http://www.kroah.com/log/linux/ols_2006_keynote.html
Sorry, I din't find it in the changelog. I was searching for something along the lines of sata_via or 8237a, there was no mention of it. I'm glad that it's there, thanks. Too bad 2.6.18 won't make it into Edgy Eft :/
The driver issue with Linux nowadays seems to be WiFi support and form what I've read BSD are far ahead.
RE[2]: VIA 8237A chipset
So far OpenBSD has been the subject at hand, perhaps we should stick to it? Besides, OpenBSD has greater support for wireless cards than the Linux kernel does, and they're even open source instead of Binary HALs and daemons to play monkey in the middle with the hardware and operating system.
The Linux kernel supports more architectures and drivers that any other operating system in the world according to http://www.kroah.com/log/linux/ols_2006_keynote.html
Yeah. <irony>You just have to find out what patchset, or third party kernel branch, and find some userland that works properly with it.</irony> I have nothing against Linux, but Linux portability is nothing like portability of NetBSD or OpenBSD. E.g. in NetBSD one can build the system (kernel + userlands) for arbitrary platforms with a single command, from a single source tree.
The wonders of computers can usually work this stuff out based on the number of items. Most sites I've bought online provide at the very least a guide to what it should cost, without leaving you in the leech.
I will email them and ask them, charging back whatever it costs isn't good enough because the cost could far outway the cost of the product which I have seen in some cases.
E.g buy a t-shirt on thinkgeek costs $25USD for delivery which isn't worth it for international orders.
That sounds like a bad excuse. There is no need for a microsoft.com or even web 2.0. Just make it look a little bit more proffesional. It wont hurt, it will only make life easier for users as the corporate world might take OpenBSD more seriously.
It's almost like your afraid of having a non-geekish professional looking site.
If it really bothers you just browse the site in links.
I know this isn't exactly what you asked, but I think it's a good answer:
8.23 - Why do the OpenBSD web pages not conform to HTML4/XHTML?
The present web pages have been carefully crafted to work on a wide variety of actual browsers going back to browser versions 4.0 and later. We do not want to make these older pages conform to HTML4 or XHTML until we're sure that they will also work with older browsers; it's just not a priority. <em>We welcome new contributors, but suggest you work on writing code, or on documenting new aspects of the system, not on tweaking the existing web pages to conform to newer standards.</em>
Not to be a smart-ass, but if it bothers you so much, why don't you volunteer to "spruce-up" the OpenBSD website for them? They take volunteers of all kinds, and I'm sure the main participants are way too busy working on OpenBSD to worry much about the looks of their site.
Besides, the people that are going to use OpenBSD will use it irregardless of the looks of the site.
(P.S. I pre-ordered my 4.0 copy! I'm trying to help support open source as best I can, and right now, purchasing the CD versions of the operating systems I use is the best I can do with my work schedule)
had OpenBSD installed on it one day and I've just upgraded along the way. OpenBSD is rock solid, and good hardware helps. I use Ubuntu on my desktops and laptops, but OpenBSD on servers where ever it makes sense, and I'm allowed.
I met Theo at a BSD can in Ottawa, ON, Canada a couple of years back, and even though I had not even a message on the OpenBSD mailing lists, forget about a patch submitted, he was very happy to chat with me and pose for an obligatory photo. Back then he gave a talk on NX (no execute) support that he had pushed for in new model x86 CPUs. Theo and OpenBSD have made such significant contributions to Open Source and computing in general that it really amazes me it isn't more popular. (I am secretly happy about this as I love the base system and server focus over the desktop. Stability gives me warm fuzzies inside.) OpenBSD fills some very important cracks in the computing industry and we should support their efforts, as we all benefit from their work.
I've heard and read lots of stuff about Theo, but meeting him in person was infinitely better. I argue that OpenBSD is kinda like that, looks rough and intimidating at first, but is a pleasure to work with as well as being very stable and dedicated to doing things right. Remember work isn't always fun, but it helps when it's productive and the end goal is of course to move ahead.
I've bought install cds myself (for display purposes and stickers), and try to advocate donations from companies. I don't think I have ever actually used an install cd I bought. heh
Thank you Theo and OpenBSD!
PS I've not written any mailing list because the documentation and mailing list archives have always held the answers for me pertaining to the OpenBSD system. Maybe I'm lucky, maybe it's the documentation...
Try: http://ports.openbsd.nu/
And for the driver: http://www.openbsd.org/cgi-bin/man.cgi?query=wpi&apropos=0&sektion=...
Over the past few days I have run OpenBSD on my desktop machine (amd64).
I ran across a few problems:
1. In the console, Alt+Left and Alt+Right acted as Left and Right respectively; however, when I showed this on #openbsd, a channel member looked through the relevant source code, found the problem and submitted the diff file to the appropriate place... I think it's great OpenBSD has people like that. (Other OSs could have people who do that, I've just never seen it done through IRC). Also, I got some good advice from bsdforums.org
2. It seems impossible to transfer data to or from FTP servers from a machine with Packet Filter (OpenBSD's firewall) without allowing all out connections for high ports. This isn't much of a problem because it's unlikely anything would try to send data through those ports from the machine. It becomes less of a problem if you specify which user can send from those ports.
However, iptables is able to handle this by reading the PORT command (this is in the FTP protocol) and determining which port is going to be used for the data port.
BTW, I'm only considering passive FTP here.
3. A few programs I regularly use (mpd, ncmpc) haven't been built for 3.9 but are in current (and hence will be in 4.0). I could follow -current rather than -stable but it's not recommended (though I figure now that the ports tree has been locked, it should be fine).
When I hadn't used OpenBSD and was considering doing so, I heard that their community was the harshest out there. While OpenBSD's community is harsh (mailing lists, IRC), I think it's for two reasons:
1. People don't want to spend time answering questions which can be solved by looking in the docs (man pages, FAQ, mailing list archives, Google).
2. By not holding peoples hands through setting up, maintaining and configuring OpenBSD, it forces users to learn how to research properly, which in the long run is best... IMO.
"However, iptables is able to handle this by reading the PORT command (this is in the FTP protocol) and determining which port is going to be used for the data port. "
On a workstation I don't see why this matters in any way.
On a firewall you'd use ftp-proxy for this. Well, you could probably use ftp-proxy on a workstation too but why bother?
--- On a workstation I don't see why this matters in any way. ---
Despite how it's just a workstation, doesn't mean I'm going to want it much less secure than that of a server. I would prefer to have one outgoing port open for FTP rather than 20,000.
--- On a firewall you'd use ftp-proxy for this. Well, you could probably use ftp-proxy on a workstation too but why bother? ---
I know ftp-proxy would be used for a firewall, to allow machines behind the firewall to use FTP properly; however, ftp-proxy doesn't allow the actual machine with PF (be that a firewall machine, or a workstation with PF) to access FTP properly.
I tried redirecting packets from 127.0.0.1 port 21 to 127.0.0.1 port 8021 (the port on which ftp-proxy listens) but this never worked.
I know it's not much of a problem, I'd just prefer to only have the neccessary ports open.
As to why I'm running PF on a workstation, I'm going to uni in a week and they only allow one computer connected to their network, meaning no firewall machine... and since I need a firewall, it must be on the workstation.
if your university only allows 1 machine connected, just set up a local lan and mask it properly from your univ's network? i can't imagine your univ's admin checking each dorm room and counting all the appliances that can be networked.
if you insist on having a single workstation doing everything, and insist on having a 'secure' way of doing ftp, you're indeed bound to use the ftp-proxy locally (i never tried this, but i'm very sure it's perfectly possible to do)
I have considered using two machines and masking it, but I'd rather not risk it.
I have tried running ftp-proxy locally but this doesn't seem to work (at least the method I tried):
rdr proto tcp from 127.0.0.1 port ftp -> 127.0.0.1 port 8201 (IIRC)
If you have idea how it could work locally, I'd be grateful if you told me, either through OSNews or replying to my post on misc (though it doesn't seem to have gone through yet).
I think you can download some floppies, I think they are three, and install from floppies or you can burn a cd40.iso to a CD (it should be a very small size).
Start the installation and then choose ``u":
(I)nstall, (U)pgrade or (S)hell?
that should do it.
That's what I remember.
Remember to check the docs with 4.0, such as the file INSTALL.i386 ...



Audigy SE doesn't use EMU10k1 chipset, it's CA0106 and according to ALSA docs: "Digital/Analog input does not work yet. Needs more development work."