OpenBSD 5.1 Released

[kraftman]

I didn't say BSD or Solaris has better performance. You think Linux is perfect?
Some things are not fully compatible, like stability and performance.

You were saying Linux is not as reliable and that's what I was talking about here. The stability and performance is tunable and that's a good thing.

So tell me which hack was that?

The one that caused an application to be killed. I don't know what they did. Maybe it's the hack from your article?

True. Linux is next.

Not true. Linux is less bloated than Windows, BSD, Solaris. I'm not talking about kernel size, but about things like ability to run unmodified kernel in embedded systems.

"Mode 2 (which is new in 2.6) is certainly an improvement over modes 0 and 1 available in the older versions of the Linux kernel. However, mode 2 doesn't mean that memory will never be overcommitted. It just uses a different heuristic for guessing how much memory is safe to allow to be allocated."

And this quote is from Solaris devs? This is from the Linux kernel documentation:

http://www.mjmwired.net/kernel/Docum...mit-accounting

Don't overcommit. The total address space commit for the system is not permitted to exceed swap + a configurable percentage (default is 50) of physical RAM.

growing rapidly = getting bloated

Just a size of a package and number lines of code. You're running just a few percent of entire Linux.

[kraftman]

This shows, that FreeBSD is more stable than Linux (distributions).
Because there are much more Linux servers than FreeBSD, but FreeBSD has good ranks.

NetBSD is used by NASA: http://www.netbsd.org/gallery/resear...l#tcp_sat_nasa

That's funny, because it shows exactly opposite. The three the most reliable servers are running on Linux. Most of the 10 most reliable servers are Linux ones. Most of the servers in the first link are also running Linux. Your data is probably outdated (1999), because NASA is running Linux. However, they may also run other OS's. Linux is also used in CERN, Submarines and Army in general:

http://www.linuxjournal.com/article/7789
http://www.linuxjournal.com/article/6978
http://jaboutboul.blogspot.com/2008/...-frontier.html

[kraftman]

I also have Atheros based card and it works great without installing third party drivers on Windows. Also why would somebody use Windows without third party drivers.
Windows at least has more stable driver ABI, so drivers can easly be installed.

I have XP and I had to install drivers manually. That's true, but depending on third party members isn't a comfortable situation. Windows has very high market share, so they're doing what MS wants, but in the future it can change.

[LightBit]

You were saying Linux is not as reliable and that's what I was talking about here. The stability and performance is tunable and that's a good thing.

It defaults to 100% performance and it's not fully tunable.

The one that caused an application to be killed. I don't know what they did. Maybe it's the hack from your article?

Nobody knows that, maybe perl did something stupid?

Not true. Linux is less bloated than Windows, BSD, Solaris. I'm not talking about kernel size, but about things like ability to run unmodified kernel in embedded systems.

OpenBSD and NetBSD can both run on embedded systems without being recompiled with most of things disabled like Linux which also needs different userland.

And this quote is from Solaris devs? This is from the Linux kernel documentation:

http://www.mjmwired.net/kernel/Docum...mit-accounting

Another source: http://opsmonkey.blogspot.com/2007/0...vercommit.html
It probably doesn't overcommit all memory, but it can overcommit avilable memory.

Just a size of a package and number lines of code. You're running just a few percent of entire Linux.

That's bloat. On Arch Linux kernel is 16MB uncompressed + 37.7MB compressed modules. OpenBSD kernel 8.6MB uncompressed and it doesn' have modules. Not to mention userland.

[LightBit]

I have XP and I had to install drivers manually. That's true, but depending on third party members isn't a comfortable situation. Windows has very high market share, so they're doing what MS wants, but in the future it can change.

With Linux 2.5 there would be no way to get it work.

[LightBit]

That's funny, because it shows exactly opposite. The three the most reliable servers are running on Linux. Most of the 10 most reliable servers are Linux ones. Most of the servers in the first link are also running Linux. Your data is probably outdated (1999), because NASA is running Linux. However, they may also run other OS's. Linux is also used in CERN, Submarines and Army in general:

http://www.linuxjournal.com/article/7789
http://www.linuxjournal.com/article/6978
http://jaboutboul.blogspot.com/2008/...-frontier.html

Reliability of servers also depend on network connection and other stuffs. More FreeBSD servers are on high ranks than Linux servers percentually. There are many Linux servers much below 40 rank.

NASA also runs IRIX, it doesn't mean they don't run NetBSD. If you look popularity of os and where it is being used, you come to conclusion that Windows is the most reliable os on world.
You won't read news which says "Windows is being used by nuclear power plant controllers", because it is obvious.

[kraftman]

It defaults to 100% performance and it's not fully tunable.

No, it's not. Default is 0, so memory overcommit is not always being done. It's fully tunable. You have proven you don't know much about Linux and your basing on not trustworthy data that usually comes from Linux competitors.

Nobody knows that, maybe perl did something stupid?

Who cares? It was killed by Solaris even without OOM, am I right? It's better to kill some process or to crash?

OpenBSD and NetBSD can both run on embedded systems without being recompiled with most of things disabled like Linux which also needs different userland.

I'm interested in mobiles in this case. I don't care about user land, but about kernels.

Another source: http://opsmonkey.blogspot.com/2007/0...vercommit.html
It probably doesn't overcommit all memory, but it can overcommit avilable memory.

Don't you see this is from 2007 and don't you think that current Linux kernel documentation is better place to find information about what Linux does?

That's bloat. On Arch Linux kernel is 16MB uncompressed + 37.7MB compressed modules. OpenBSD kernel 8.6MB uncompressed and it doesn' have modules. Not to mention userland.

No, package size doesn't matter, because you're not running every file system and driver that package contains. Linux supports many more drivers and file systems than OpenBSD, so the package must be bigger. This should be obvious. Btw. I didn't know that not having modules is something good. I also didn't know having dozens of drivers and file systems is bad. What did you check? vmlinuz, perhaps?

[kraftman]

With Linux 2.5 there would be no way to get it work.

I'm sorry, but I don't understand what you're trying to say. Linux 2.5 is experimental branch.

[kraftman]

Reliability of servers also depend on network connection and other stuffs. More FreeBSD servers are on high ranks than Linux servers percentually. There are many Linux servers much below 40 rank.

It's nonsense. While there are more Linux servers in the first ten and while there are much more servers in the entire list it simply means Linux servers were more reliable, because they get into the list.

NASA also runs IRIX, it doesn't mean they don't run NetBSD. If you look popularity of os and where it is being used, you come to conclusion that Windows is the most reliable os on world.
You won't read news which says "Windows is being used by nuclear power plant controllers", because it is obvious.

Your last sentence proves the popularity isn't an only factor that matters. It's true that Linux servers runs the most critical workloads and popularity is just a bonus in this case.

[yogi_berra]

It's better to kill some process or to crash?

It's better to do neither on a Desktop. Windows doesn't kill anything, it enlarges the page file. Much better than OOM roulette.