colinux-devel

[coLinux-devel] coLinux 0.5.3-pre1

[Dan A. <da...@gm...>]

Hello,

After two days of intensive coding, I'm releasing this pre-release. The 
reason for is that although it appears to be working, I'm not finished 
cleaning it up, so I might missed some real issues. 

The files for this release on sourceforge were added to the 0.5.2 
release in order to avoid attention from stray users :)

Here are the list of changes:

* Version 0.5.3-pre1 AKA "flamed chicken"
  * Fixed a BSOD bug (thanks Richard Goodwin for the memory dump stacks)
  * No longer depending on the Microsoft DDK for the build (review 
    doc/building), (thanks Nir Perry for the encourgement)
  * The daemon launches the console by default when it is launched (can be 
    turned off with '-d', see '-h').
  * The daemon is killable just like a normal Windows process. Doing so stops 
    the Linux machines cleanly (from Windows' standpoint), so feel free 
    to close the window, end the task, CTRL-C, etc.
  * It no longer depends on the Microsoft DDK for the build.
  * An almost complete rewrite - we now use message passing instead of 
    hardwired ioctls(). Only the daemon communicates with the driver. The 
    console and the network daemon communicate with the daemon using a 
    named pipe.
  * Fixed the block device bug that fsck triggered (thanks Daniel Glockner)
  * Hopefully addressed the network corruption issue (thanks Daniel Glockner
     again)
  * Boot parameters are now passed to the kernel. Make sure you update your 
    XMLs with 'root=/dev/cobd0' if you still want to boot from the first 
    block device. 
  * The conet daemon is a separate process (shouldn't affect you).

	
-- 
Dan Aloni
da...@gm...

Re: [coLinux-devel] coLinux 0.5.3-pre1

[Thomas F. <tf...@no...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello Dan,

cool (pre)release. Sounds like all major problem are solved?! I compiled
your *falmed chicken* with success with linux. But sorry... I have
currently no windows here, so I could not test the binary so far.

Just one question to the colinux-community: This is about the tab-win32
"net"-driver. I have this tap-win32 driver in the colinux tree but it
looks like this driver is not build in colinux? Also in the binary-distro
this files come with other date/time. Is this the driver from the openvpn
projekt?
Did you build this drivers from windows or from linux?

So long and thanks for the chicken...
  Thomas



> Hello,
>
> After two days of intensive coding, I'm releasing this pre-release. The
> reason for is that although it appears to be working, I'm not finished
> cleaning it up, so I might missed some real issues.
>
> The files for this release on sourceforge were added to the 0.5.2
> release in order to avoid attention from stray users :)
>
> Here are the list of changes:
>
> * Version 0.5.3-pre1 AKA "flamed chicken"
>   * Fixed a BSOD bug (thanks Richard Goodwin for the memory dump stacks)
>   * No longer depending on the Microsoft DDK for the build (review
>     doc/building), (thanks Nir Perry for the encourgement)
>   * The daemon launches the console by default when it is launched (can be
>     turned off with '-d', see '-h').
>   * The daemon is killable just like a normal Windows process. Doing so
> stops
>     the Linux machines cleanly (from Windows' standpoint), so feel free
>     to close the window, end the task, CTRL-C, etc.
>   * It no longer depends on the Microsoft DDK for the build.
>   * An almost complete rewrite - we now use message passing instead of
>     hardwired ioctls(). Only the daemon communicates with the driver. The
>     console and the network daemon communicate with the daemon using a
>     named pipe.
>   * Fixed the block device bug that fsck triggered (thanks Daniel
> Glockner)
>   * Hopefully addressed the network corruption issue (thanks Daniel
> Glockner
>      again)
>   * Boot parameters are now passed to the kernel. Make sure you update
> your
>     XMLs with 'root=/dev/cobd0' if you still want to boot from the first
>     block device.
>   * The conet daemon is a separate process (shouldn't affect you).
>
>
> --
> Dan Aloni
> da...@gm...
>
>
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> coLinux-devel mailing list
> coL...@li...
> https://lists.sourceforge.net/lists/listinfo/colinux-devel
>


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.3 (GNU/Linux)

iD8DBQFAJYwinwJBIFTVIqwRAl83AJwOpNZhX80cI0sdwBMJ8jS/L7rkYQCfXwuK
TvcrjAZhTCKT4hWcCvJeUjM=
=RZLJ
-----END PGP SIGNATURE-----

Re: [coLinux-devel] coLinux 0.5.3-pre1

[Richard G. <ric...@ri...>]

Something appeared to be odd with the networking (at least the first few
times I tried).  If I started with -d, I could ping the cL session, but if I
let the console autostart, I can't.

AFter some more playing around, it seems like when I close the console
sometimes networking dies. (killing console with the X).

Anyone else?

----- Original Message ----- 
From: "Dan Aloni" <da...@gm...>
To: "Cooperative Linux Development" <col...@li...>
Sent: Saturday, February 07, 2004 6:19 PM
Subject: [coLinux-devel] coLinux 0.5.3-pre1


> Hello,
>
> After two days of intensive coding, I'm releasing this pre-release. The
> reason for is that although it appears to be working, I'm not finished
> cleaning it up, so I might missed some real issues.
>
> The files for this release on sourceforge were added to the 0.5.2
> release in order to avoid attention from stray users :)
>
> Here are the list of changes:
>
> * Version 0.5.3-pre1 AKA "flamed chicken"
>   * Fixed a BSOD bug (thanks Richard Goodwin for the memory dump stacks)
>   * No longer depending on the Microsoft DDK for the build (review
>     doc/building), (thanks Nir Perry for the encourgement)
>   * The daemon launches the console by default when it is launched (can be
>     turned off with '-d', see '-h').
>   * The daemon is killable just like a normal Windows process. Doing so
stops
>     the Linux machines cleanly (from Windows' standpoint), so feel free
>     to close the window, end the task, CTRL-C, etc.
>   * It no longer depends on the Microsoft DDK for the build.
>   * An almost complete rewrite - we now use message passing instead of
>     hardwired ioctls(). Only the daemon communicates with the driver. The
>     console and the network daemon communicate with the daemon using a
>     named pipe.
>   * Fixed the block device bug that fsck triggered (thanks Daniel
Glockner)
>   * Hopefully addressed the network corruption issue (thanks Daniel
Glockner
>      again)
>   * Boot parameters are now passed to the kernel. Make sure you update
your
>     XMLs with 'root=/dev/cobd0' if you still want to boot from the first
>     block device.
>   * The conet daemon is a separate process (shouldn't affect you).
>
>
> -- 
> Dan Aloni
> da...@gm...
>
>
> -------------------------------------------------------
> The SF.Net email is sponsored by EclipseCon 2004
> Premiere Conference on Open Tools Development and Integration
> See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
> http://www.eclipsecon.org/osdn
> _______________________________________________
> coLinux-devel mailing list
> coL...@li...
> https://lists.sourceforge.net/lists/listinfo/colinux-devel
>
>

Re: [coLinux-devel] coLinux 0.5.3-pre1

[Dan A. <da...@gm...>]

On Sat, Feb 07, 2004 at 07:23:19PM -0600, Richard Goodwin wrote:
> Something appeared to be odd with the networking (at least the first few
> times I tried).  If I started with -d, I could ping the cL session, but if I
> let the console autostart, I can't.
> 
> AFter some more playing around, it seems like when I close the console
> sometimes networking dies. (killing console with the X).
> 
> Anyone else?

I knew there was something I missed. I'll look into this, thanks.

What about the network packet corruption problem? Does anyone still 
experience it with 0.5.3-pre1?

-- 
Dan Aloni
da...@gm...

[PATCH] Re: [coLinux-devel] coLinux 0.5.3-pre1

[Dan A. <da...@gm...>]

On Sat, Feb 07, 2004 at 07:23:19PM -0600, Richard Goodwin wrote:
> Something appeared to be odd with the networking (at least the first few
> times I tried).  If I started with -d, I could ping the cL session, but if I
> let the console autostart, I can't.
> 
> AFter some more playing around, it seems like when I close the console
> sometimes networking dies. (killing console with the X).

Yep, there was a bug in the daemon, this should fix it:

--- colinux-20040208/src/colinux/user/daemon.c	2004-02-08 01:20:19.000000000 +0200
+++ colinux-main/src/colinux/user/daemon.c	2004-02-08 08:32:12.000000000 +0200
@@ -556,7 +556,7 @@
 		message.message.type = CO_MESSAGE_TYPE_OTHER;
 		message.message.size = sizeof(message.switchm);
 		message.switchm.type = CO_SWITCH_MESSAGE_FREE_RULE;
-		message.switchm.destination = CO_MODULE_CONET0;
+		message.switchm.destination = module->id;
 		
 		co_message_switch_dup_message(&module->daemon->message_switch, &message.message);
 		co_message_switch_free_rule(&module->daemon->message_switch, module->id);


-- 
Dan Aloni
da...@gm...