Menu
▾
▴
colinux-devel — Cooperative Linux development mailing list
You can subscribe to this list here.
| 2004 |
Jan
(64) |
Feb
(530) |
Mar
(266) |
Apr
(580) |
May
(360) |
Jun
(161) |
Jul
(185) |
Aug
(164) |
Sep
(123) |
Oct
(160) |
Nov
(59) |
Dec
(84) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2005 |
Jan
(156) |
Feb
(95) |
Mar
(124) |
Apr
(81) |
May
(79) |
Jun
(179) |
Jul
(35) |
Aug
(64) |
Sep
(56) |
Oct
(57) |
Nov
(18) |
Dec
(41) |
| 2006 |
Jan
(65) |
Feb
(37) |
Mar
(59) |
Apr
(73) |
May
(65) |
Jun
(27) |
Jul
(54) |
Aug
(76) |
Sep
(103) |
Oct
(23) |
Nov
(45) |
Dec
(29) |
| 2007 |
Jan
(41) |
Feb
(47) |
Mar
(61) |
Apr
(24) |
May
(14) |
Jun
(6) |
Jul
(23) |
Aug
(30) |
Sep
(16) |
Oct
(9) |
Nov
(53) |
Dec
(36) |
| 2008 |
Jan
(19) |
Feb
(49) |
Mar
(74) |
Apr
(21) |
May
(24) |
Jun
(5) |
Jul
(9) |
Aug
(53) |
Sep
(26) |
Oct
(23) |
Nov
(32) |
Dec
(19) |
| 2009 |
Jan
(47) |
Feb
(49) |
Mar
(39) |
Apr
(61) |
May
(28) |
Jun
(19) |
Jul
(12) |
Aug
(10) |
Sep
(31) |
Oct
(16) |
Nov
(60) |
Dec
(26) |
| 2010 |
Jan
(17) |
Feb
(9) |
Mar
(32) |
Apr
(11) |
May
(24) |
Jun
(33) |
Jul
(5) |
Aug
(2) |
Sep
(7) |
Oct
(8) |
Nov
(17) |
Dec
(7) |
| 2011 |
Jan
(12) |
Feb
(16) |
Mar
(2) |
Apr
(12) |
May
(5) |
Jun
(10) |
Jul
(3) |
Aug
(3) |
Sep
(2) |
Oct
(1) |
Nov
(17) |
Dec
(1) |
| 2012 |
Jan
(9) |
Feb
(9) |
Mar
(8) |
Apr
(4) |
May
(2) |
Jun
(1) |
Jul
(4) |
Aug
(8) |
Sep
(11) |
Oct
(1) |
Nov
(2) |
Dec
(2) |
| 2013 |
Jan
|
Feb
(7) |
Mar
(4) |
Apr
(10) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(5) |
Dec
(3) |
| 2016 |
Jan
|
Feb
(2) |
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
| 2023 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
|
From: Richard G. <ric...@ri...> - 2004-02-05 17:09:53
|
Interesting.. I actually have a dual port adapter, plus the TAP, plus a bluetooth network "adapter", plus one more... I wonder why bridging doesn't show up for me. Tony, ICS IS missing for you right? BTW, the easiest answer to all this is to run a proxy on the Windows host, and most of the linux stuff will communicate through an http proxy. However, this crashes the current daemon (Dan is working on it, memory problem) so I'm still toying with the bridging stuff in the meantime. Ian, Thanks for the smackdown on the networking info...In my haste of making up numbers, I broke some rules ;-) ----- Original Message ----- From: "Tony Hoyle" <tm...@no...> To: "Richard Goodwin" <ric...@ri...> Cc: "Ian C. Blenke" <ia...@bl...>; "Cooperative Linux Development" <col...@li...> Sent: Thursday, February 05, 2004 11:03 AM Subject: Re: [coLinux-devel] TCP Forwarding? Or General Networking Questions..? > Richard Goodwin wrote: > > Ian, > > > > It's not authentication... apparently XP removes the bridging and ICS > > options when in a domain. They're simply not there. > > > Bridging is there... I'm in a domain at the moment and right clicking on > the adapter has a 'bridge connections' option. > > It only appears if you have >1 network adapter I think. > > Tony > > > |
|
From: Tony H. <tm...@no...> - 2004-02-05 17:03:55
|
Richard Goodwin wrote: > Ian, > > It's not authentication... apparently XP removes the bridging and ICS > options when in a domain. They're simply not there. > Bridging is there... I'm in a domain at the moment and right clicking on the adapter has a 'bridge connections' option. It only appears if you have >1 network adapter I think. Tony |
|
From: Ian C. B. <ia...@bl...> - 2004-02-05 17:01:28
|
On Thu, Feb 05, 2004 at 10:49:32AM -0600, Richard Goodwin wrote: > Ian, > > It's not authentication... apparently XP removes the bridging and ICS > options when in a domain. They're simply not there. Oh?! Interesting! That's a first for me. Has anyone else experienced this before? - Ian |
|
From: Richard G. <ric...@ri...> - 2004-02-05 16:49:37
|
Ian, It's not authentication... apparently XP removes the bridging and ICS options when in a domain. They're simply not there. ----- Original Message ----- From: "Ian C. Blenke" <ia...@bl...> To: "Richard Goodwin" <ric...@ri...> Cc: "Cooperative Linux Development" <col...@li...> Sent: Thursday, February 05, 2004 10:29 AM Subject: Re: [coLinux-devel] TCP Forwarding? Or General Networking Questions..? > On Thu, Feb 05, 2004 at 10:00:04AM -0600, Richard Goodwin wrote: > > Ok, > > > > So on my work machine, where I'd most dearly love to run cL, I am unable to > > Bridge or enable ICS because we are members of a domain. So, I need to I > > guess use the IP forwarding of XP or figure out some other solution. > > Why does bridging or ICS affect your authentication? I'm confused. > > > Here's the setup: > > > > Local Area Connection 2: "corporate" network, gets DHCP and DNS addresses > > from network. For arguments sake, let's say it's IP is 199.199.199.199, and > > the gateway is 199.199.199.1, and subnet mask is 255.255.255.252 (which it > > actually is) > > That's impossible. If your IP is 199.199.199.199 with a 255.255.255.252 > mask, your IP is the broadcast (all ones) address on the 199.199.199.196 > network. In that case, I would think it far more likely that your IP > would be 199.199.199.198, with a default route through your router at > 199.199.199.197. The 199.199.199.1 IP would be on a completely different > netblock, unreachable directly. > > You have a host on a /30 segment? Bridging probably is right out for you > then.. but why does ICS block your SMB connectivity? > > If you meant a 255.255.252.0 netblock, then bridging would make sense. > I'll bet this is what you have, as a /22 block with 1024 hosts is far > more common on Enterprise networks. > > > TAP: You know what this is ;-) Currently set to 192.168.0.1, and the gateway > > set to the IP of the LAC#2 (don't know if that's right or not) > > You don't want a gateway on that interface. You only want one "default gateway" on a > box (where packets not destined for any local static routes are sent). > > On your Linux image, you would set your default gateway 192.168.0.1. > > Think of gateway as the "gateway of last resort". If a packet is > destined for somewhere your machine doesn't know how to get to, it hands > it off to the gateway for it to deliver the traffic for you. > > > eth0 in cL: currently set to 192.168.0.40, and gateway set to TAP IP. > > That is correct. > > > I did the IPEnableRouter = 1 in the registry and rebooted, and I can ping > > TAP and LAC#2 from cL, but can't ping past LAC#2. Am I missing something > > obvious? > > Yes. Your network beyond your 199.199.199.x gateway does not have a route > to get back to the 192.168.0.0/24 network. Packets go out, but your enterprise > network does not know how to give them back. > > There are two solutions to this problem: > > 1. You need to enable ICS. All traffic will then appear to be coming from your > LAC#2. ICS is glorified Network Address Translation (NAT). > 2. Convince your enterprise network guys to add a route to your little private > virtual segment on the routers. > > - Ian C. Blenke <ia...@bl...> > > > > ------------------------------------------------------- > 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 > > |
|
From: Ian C. B. <ia...@bl...> - 2004-02-05 16:29:58
|
On Thu, Feb 05, 2004 at 10:00:04AM -0600, Richard Goodwin wrote: > Ok, > > So on my work machine, where I'd most dearly love to run cL, I am unable to > Bridge or enable ICS because we are members of a domain. So, I need to I > guess use the IP forwarding of XP or figure out some other solution. Why does bridging or ICS affect your authentication? I'm confused. > Here's the setup: > > Local Area Connection 2: "corporate" network, gets DHCP and DNS addresses > from network. For arguments sake, let's say it's IP is 199.199.199.199, and > the gateway is 199.199.199.1, and subnet mask is 255.255.255.252 (which it > actually is) That's impossible. If your IP is 199.199.199.199 with a 255.255.255.252 mask, your IP is the broadcast (all ones) address on the 199.199.199.196 network. In that case, I would think it far more likely that your IP would be 199.199.199.198, with a default route through your router at 199.199.199.197. The 199.199.199.1 IP would be on a completely different netblock, unreachable directly. You have a host on a /30 segment? Bridging probably is right out for you then.. but why does ICS block your SMB connectivity? If you meant a 255.255.252.0 netblock, then bridging would make sense. I'll bet this is what you have, as a /22 block with 1024 hosts is far more common on Enterprise networks. > TAP: You know what this is ;-) Currently set to 192.168.0.1, and the gateway > set to the IP of the LAC#2 (don't know if that's right or not) You don't want a gateway on that interface. You only want one "default gateway" on a box (where packets not destined for any local static routes are sent). On your Linux image, you would set your default gateway 192.168.0.1. Think of gateway as the "gateway of last resort". If a packet is destined for somewhere your machine doesn't know how to get to, it hands it off to the gateway for it to deliver the traffic for you. > eth0 in cL: currently set to 192.168.0.40, and gateway set to TAP IP. That is correct. > I did the IPEnableRouter = 1 in the registry and rebooted, and I can ping > TAP and LAC#2 from cL, but can't ping past LAC#2. Am I missing something > obvious? Yes. Your network beyond your 199.199.199.x gateway does not have a route to get back to the 192.168.0.0/24 network. Packets go out, but your enterprise network does not know how to give them back. There are two solutions to this problem: 1. You need to enable ICS. All traffic will then appear to be coming from your LAC#2. ICS is glorified Network Address Translation (NAT). 2. Convince your enterprise network guys to add a route to your little private virtual segment on the routers. - Ian C. Blenke <ia...@bl...> |
|
From: Richard G. <ric...@ri...> - 2004-02-05 16:00:03
|
Ok, So on my work machine, where I'd most dearly love to run cL, I am unable to Bridge or enable ICS because we are members of a domain. So, I need to I guess use the IP forwarding of XP or figure out some other solution. Here's the setup: Local Area Connection 2: "corporate" network, gets DHCP and DNS addresses from network. For arguments sake, let's say it's IP is 199.199.199.199, and the gateway is 199.199.199.1, and subnet mask is 255.255.255.252 (which it actually is) TAP: You know what this is ;-) Currently set to 192.168.0.1, and the gateway set to the IP of the LAC#2 (don't know if that's right or not) eth0 in cL: currently set to 192.168.0.40, and gateway set to TAP IP. I did the IPEnableRouter = 1 in the registry and rebooted, and I can ping TAP and LAC#2 from cL, but can't ping past LAC#2. Am I missing something obvious? Thanks! |
|
From: Dan A. <da...@gm...> - 2004-02-04 22:41:52
|
On Thu, Feb 05, 2004 at 12:32:34AM +0200, Dan Aloni wrote: > Would one of you like me to maintain precompiled binaries of a cygwin > cross build tools for Linux, as .deb and/or .rpm files, so that the > others won't have to mess with setting up this build environment? No, what I actually wanted to write there: would one of you like to maintain it? -- Dan Aloni da...@gm... |
|
From: Dan A. <da...@gm...> - 2004-02-04 22:32:47
|
Good evening, I've successfully managed to build and test linux.sys using the GNU toolchain. I used ReactOS's helper.mk, pointed by Steven Edwards to generated the rather complex command line sequence that achieves this goal (look at the patch below). Thanks for Nir Perry for encouraging me to try this method further than I thought. Next coLinux release can be built on a Linux box entirely, using the GNU toolchain. However, w32api needs to be patched and built from source, because there is a mistake in one of the kernel imports defined there (I'm mailing the author). Would one of you like me to maintain precompiled binaries of a cygwin cross build tools for Linux, as .deb and/or .rpm files, so that the others won't have to mess with setting up this build environment? --- colinux-20040131/src/colinux/os/winnt/build/Makefile 2004-01-31 18:57:08.000000000 +0200 +++ colinux-20040204/src/colinux/os/winnt/build/Makefile 2004-02-05 00:11:27.000000000 +0200 @@ -21,47 +21,30 @@ DRIVER_TARGET := $(BUILD_PATH)/linux.sys -WINDIR=$(WINUSER)@$(WINBOX):$(WINREMOTEDIR) -WINBOX_SCP=scp -i ~/.ssh/locallan -WINBOX_SSH=ssh -X -l $(WINUSER) -i ~/.ssh/locallan $(WINBOX) -WINLINK=$(WINBOX_SSH) /cygdrive/c/NTDDK/bin/link.exe - -WINLINK_FLAGS= \ - /machine:ix86 \ - /STACK:262144,4096 \ - /MERGE:_PAGE=PAGE \ - /MERGE:_TEXT=.text \ - /SECTION:INIT,d \ - /OPT:REF \ - /OPT:ICF \ - /IGNORE:4001,4037,4039,4044,4065,4070,4078,4087,4089,4198 \ - /INCREMENTAL:NO \ - /FULLBUILD \ - /FORCE:MULTIPLE \ - /NOCOMMENT \ - /release \ - /NODEFAULTLIB \ - /debug:FULL \ - /debugtype:cv \ - /version:5.00 \ - /osversion:5.00 \ - /optidata \ - /driver \ - /align:0x1000 \ - /filealign:0x1000 \ - /subsystem:native,5.00 \ - /base:0x10000 \ - /entry:DriverEntry@8 \ - -WINLINK_LIBS= \ - C:\\\\NTDDK\\\\libfre\\\\i386\\\\ntoskrnl.lib \ - C:\\\\NTDDK\\\\libfre\\\\i386\\\\hal.lib \ - C:\\\\NTDDK\\\\libfre\\\\i386\\\\wmilib.lib \ - -$(DRIVER_TARGET): $(BUILD_PATH)/driver.o - $(WINBOX_SCP) $^ $(WINDIR) - $(WINLINK) /out:$(notdir $@) $(WINLINK_FLAGS) $(notdir $^) $(WINLINK_LIBS) - $(WINBOX_SCP) $(WINDIR)$(notdir $@) $@ +CLEAN_FILES += $(BUILD_PATH)/driver.base.tmp +$(BUILD_PATH)/driver.base.tmp: $(BUILD_PATH)/driver.o + i686-pc-cygwin-gcc -Wl,--base-file,$@ \ + -Wl,--entry,_DriverEntry@8 \ + -nostartfiles -nostdlib \ + -o junk.tmp $^ -lntoskrnl -lhal + rm -f junk.tmp + +CLEAN_FILES += $(BUILD_PATH)/driver.base.exp +$(BUILD_PATH)/driver.base.exp: $(BUILD_PATH)/driver.base.tmp + i686-pc-cygwin-dlltool --dllname linux.sys \ + --base-file $^ \ + --output-exp $@ + +$(DRIVER_TARGET): $(BUILD_PATH)/driver.o $(BUILD_PATH)/driver.base.exp + i686-pc-cygwin-gcc \ + -Wl,--subsystem,native \ + -Wl,--image-base,0x10000 \ + -Wl,--file-alignment,0x1000 \ + -Wl,--section-alignment,0x1000 \ + -Wl,--entry,_DriverEntry@8 \ + -Wl,$(word 2,$^) \ + -mdll -nostartfiles -nostdlib \ + -o $@ $(word 1,$^) -lntoskrnl -lhal LOCAL_TARGET += $(DRIVER_TARGET) LOCAL_FILES = \ @@ -74,16 +57,7 @@ colinux: $(LOCAL_FILES) -parts: $(LOCAL_FILES) - -upload: $(LOCAL_FILES) - $(WINBOX_SCP) $^ $(WINDIR) - -upload_console: $(BUILD_ROOT)/colinux/os/current/user/console/colinux-console.exe - $(WINBOX_SCP) $^ $(WINDIR) - driver: $(BUILD_PATH)/driver.o - CLEAN_FILES := $(CLEAN_FILES) \ $(LOCAL_TARGETS) Here's the patch against w32api: --- w32api-2.4/lib/ddk/ntoskrnl.def 2003-02-09 15:26:01.000000000 +0200 +++ w32api-2.4-patched/lib/ddk/ntoskrnl.def 2004-02-04 23:40:02.000000000 +0200 @@ -596,8 +596,8 @@ KeSetTargetProcessorDpc@8 ;KeSetTimeIncrement @KeSetTimeUpdateNotifyRoutine@4 -KeSetTimer@12 -KeSetTimerEx@16 +KeSetTimer@16 +KeSetTimerEx@20 ;KeStackAttachProcess KeSynchronizeExecution@12 ;KeTerminateThread -- Dan Aloni da...@gm... |
|
From: Szalai F. <sz...@ei...> - 2004-02-04 20:09:16
|
On Wed, 2004-02-04 at 19:16, andre wrote: > you could also install Debain in a harddisk partition and move the partition > with > dd if=/dev/hda2 of=/mnt/windows/colinux/harddisk.img here may come the 'debootstap' to install the base system to specified directory as root. -- Regards Feri |
|
From: Dan A. <da...@gm...> - 2004-02-04 19:31:56
|
On Wed, Feb 04, 2004 at 01:58:41PM -0300, Paulo Silva wrote: > Hello Dan, > > could you provide some information on how the Debian > root image was created? I would like to use a smaller > filesystem image (1 GB is kind of big, at least to my > notebook :-) so any directions on how to do that are > welcome. I took an existing Debian root file system from User Mode Linux's sourceforge page and made slight modifications for it to work. That image's initial size was about 60MB. -- Dan Aloni da...@gm... |
|
From: andre <avb...@gm...> - 2004-02-04 19:00:03
|
On Wednesday 04 February 2004 20:44, John LeSueur wrote: > >ps. install another editor besides vi because windows closes the > >colinux-console window when you press ESC. Atleast it does that on my > >computer. > > Or you could install sshd and then you don't have to worry about the > console at all, unless networking isn't working... > > John But than you have to have ssh installed on windows to. Also it makes editing you network settings a bit hard when you don't have a connection. Neither of those two are big problems but it maybe better to solve if possible the behaviour of colinux-console |
|
From: Ballard J. <sac...@ho...> - 2004-02-04 18:57:08
|
From: "andre" <avb...@gm...> > > ps. install another editor besides vi because windows closes the > colinux-console window when you press ESC. Atleast it does that on my > computer. Do an "apt-get install telnetd" and then from your main system you can use telnet. You'll probally want to set "TERM=vt100" and do a "setterm -initialize" afterwords. Works good and is faster. |
|
From: John L. <jo...@su...> - 2004-02-04 18:49:53
|
>ps. install another editor besides vi because windows closes the >colinux-console window when you press ESC. Atleast it does that on my >computer. > > > Or you could install sshd and then you don't have to worry about the console at all, unless networking isn't working... John |
|
From: andre <avb...@gm...> - 2004-02-04 18:18:21
|
Don't know how he did it with Debian but generic it is something like this mount -t vfat /devhda /mnt/windows dd if=/dev/zero of=/mnt/windows/colinux/new.img count=X mkfs.ext3 /mnt/windows/colinux/new.img mount -t ext3 /mnt/windows/colinux/new.img /mnt/image -o loop Than install colinux kernel and other software in /mnt/image you could also install Debain in a harddisk partition and move the partition with dd if=/dev/hda2 of=/mnt/windows/colinux/harddisk.img Or you can do a lin4win installation and just start the resulting linuxsys.img instead of Debian-3.0r0.ext3.1gb and hope it works (It did with Mandrake 9.2) ps. install another editor besides vi because windows closes the colinux-console window when you press ESC. Atleast it does that on my computer. |
|
From: <pau...@ya...> - 2004-02-04 16:58:49
|
Hello Dan, could you provide some information on how the Debian root image was created? I would like to use a smaller filesystem image (1 GB is kind of big, at least to my notebook :-) so any directions on how to do that are welcome. Thank you Paulo Silva ______________________________________________________________________ Yahoo! GeoCities: 15MB de espaço grátis para criar seu web site! http://br.geocities.yahoo.com/ |
|
From: Ian L. <Ian...@mq...> - 2004-02-04 09:12:52
|
Thanks Dan - I'll wait for you. ----- Original Message ----- >From: "Dan Aloni" <da...@gm...> >To: "Ian Latter" <Ian...@mq...> >Subject: Re: [coLinux-devel] colinux-daemon debug info? or kernel sizes? >Date: Wed, 04 Feb 2004 08:11:09 +0200 > > On Wed, Feb 04, 2004 at 03:30:29PM +1000, Ian Latter wrote: > > > Maybe what would help is a debug version of the daemon, where > > it doesn't close down the monitor on exit (?) .. this would mean > > rebooting the machine between tests, but at this stage its better > > than not being able to get any info at all. Alternatively, a version > > of daemon that writes console output to a text file (like a "dmesg" > > style output) would also be very useful. > > I am planning to rewrite some parts of the daemon so it would work in a > different manner by returning to userspace in its main loop. It would > allow such printings to take place, and for other things to improve (the > console will do IPC against the daemon instead of connecting to the > driver, which should improve stablity). > > -- > 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 > -- Ian Latter Internet and Networking Security Officer Macquarie University Meet me at the Australian Unix and open systems User Group (AUUG) Security Symposium; 2004 http://www.auug.org.au/events/2004/security/ |
|
From: Ian L. <Ian...@mq...> - 2004-02-04 09:02:41
|
Sounds like great stuff. If you can see any reason for the kernel not booting lemme know. My attempt to put a delay/loop in the co_terminate failed .. it may well be too late by then. And I don't have the time to dik around with it too much, as I'm in the middle of a Tender. ----- Original Message ----- >From: "Ballard Jonathan" <sac...@ho...> >To: "Ian Latter" <Ian...@mq...> >Subject: Re:[coLinux-devel] a few .h files not found >Date: Tue, 03 Feb 2004 23:48:52 -0800 > > I do not have a standalone linux machine available to do the full compile. > I am resourceful enough to extract the information from the distributions > and piece it together to make it work. Only a few files slowed me down. > > I have started to peruse the software. And, I have started to update some > of the technology. Already, I have a version that compiles without the DDK > distribution. With a little research into the reversed engineered > technology, I trimmed a branch off and rewrote the assembly. > > One hurdle to overcome... or maybe it was the tollbooth on the bridge. It > was time to clean out the trolls. > > > From: "Ian Latter" <Ian...@mq...> > > > Ok, I can finally contribute some help back on this topic :) > > > > I don't have any problems compiling the 2.4.24 kernel with the > > 20040131 colinux patch. > > -- Ian Latter Internet and Networking Security Officer Macquarie University Meet me at the Australian Unix and open systems User Group (AUUG) Security Symposium; 2004 http://www.auug.org.au/events/2004/security/ |
|
From: Ballard J. <sac...@ho...> - 2004-02-04 07:31:04
|
I do not have a standalone linux machine available to do the full compile. I am resourceful enough to extract the information from the distributions and piece it together to make it work. Only a few files slowed me down. I have started to peruse the software. And, I have started to update some of the technology. Already, I have a version that compiles without the DDK distribution. With a little research into the reversed engineered technology, I trimmed a branch off and rewrote the assembly. One hurdle to overcome... or maybe it was the tollbooth on the bridge. It was time to clean out the trolls. From: "Ian Latter" <Ian...@mq...> > Ok, I can finally contribute some help back on this topic :) > > I don't have any problems compiling the 2.4.24 kernel with the > 20040131 colinux patch. |
|
From: Ian L. <Ian...@mq...> - 2004-02-04 06:46:21
|
Ok, I can finally contribute some help back on this topic :)
I don't have any problems compiling the 2.4.24 kernel with the
20040131 colinux patch.
These are the things that you need to have/do;
1. You've got to use GCC v3 (3.2.2 is what I currently use, 3.2.3
is what Dan uses). RedHat 9 is fine for this.
2. Unpack kernel
cd /usr/src
tar xvfz ~/linux-2.4.24.tar.gz
ln -s linux-2.4.24 linux
3. Apply colinux kernel patch
cd /usr/src/linux
cat ~/colinux-20040131/patch/linux | patch -p1
4. Build the kernel
cd /usr/src/linux
make mrproper
cp -f ~/colinux-20040131/conf/linux-config ./.config
make oldconfig
make dep
make vmlinux
This should all work. Then you may or may not have success
with that kernel ... either way, it should build fine/error free.
----- Original Message -----
>From: "Ballard Jonathan" <sac...@ho...>
>To: <col...@li...>
>Subject: [coLinux-devel] a few .h files not found
>Date: Tue, 03 Feb 2004 08:02:21 -0800
>
> I tried to compile some of the source and the compiler did not find some header files.
> I installed the colinux-20040131 source and the linux-2.4.24 source.
>
>
> linux/cooperative.h
> -- i looked but did not find
>
> asm/page.h
> -- I moved ./asm-i386 to ./asm and it works
>
> linux/autoconf.h
> -- i think i have to generate this by hand for a new setup
>
>
>
--
Ian Latter
Internet and Networking Security Officer
Macquarie University
Meet me at the Australian Unix and open systems
User Group (AUUG) Security Symposium; 2004
http://www.auug.org.au/events/2004/security/
|
|
From: Ballard J. <sac...@ho...> - 2004-02-04 06:30:55
|
I tried to compile some of the source and the compiler did not find some = header files. I installed the colinux-20040131 source and the linux-2.4.24 source. linux/cooperative.h -- i looked but did not find asm/page.h -- I moved ./asm-i386 to ./asm and it works linux/autoconf.h -- i think i have to generate this by hand for a new setup |
|
From: Dan A. <da...@gm...> - 2004-02-04 06:11:24
|
On Wed, Feb 04, 2004 at 03:30:29PM +1000, Ian Latter wrote: > Maybe what would help is a debug version of the daemon, where > it doesn't close down the monitor on exit (?) .. this would mean > rebooting the machine between tests, but at this stage its better > than not being able to get any info at all. Alternatively, a version > of daemon that writes console output to a text file (like a "dmesg" > style output) would also be very useful. I am planning to rewrite some parts of the daemon so it would work in a different manner by returning to userspace in its main loop. It would allow such printings to take place, and for other things to improve (the console will do IPC against the daemon instead of connecting to the driver, which should improve stablity). -- Dan Aloni da...@gm... |
|
From: Ian L. <Ian...@mq...> - 2004-02-04 05:45:55
|
> What happens here is probably that Linux boots, and then exists for > some reason from one of the co_terminate() calls, maybe due to a panic. > > At this stage, you can trying writing a script that attaches the console > just when the daemon starts booting Linux in order to catch it. My initial > intention was that the daemon itself would be given a switch that would > instruct it to launch a console when the boot starts. Using two dos windows to do the same .. I line the two commands up .. hit enter on the first, click the second and hit enter ... my timing is right, but the whole process is too quick. The console-monitor only manages to draw its frame before the daemon starts to shut things down. Maybe what would help is a debug version of the daemon, where it doesn't close down the monitor on exit (?) .. this would mean rebooting the machine between tests, but at this stage its better than not being able to get any info at all. Alternatively, a version of daemon that writes console output to a text file (like a "dmesg" style output) would also be very useful. Hmm .. if it calls co_terminate then perhaps I could get this function to sleep for a minute within the linux kernel itself ... might help. This I could have a look at. -- Ian Latter Internet and Networking Security Officer Macquarie University Meet me at the Australian Unix and open systems User Group (AUUG) Security Symposium; 2004 http://www.auug.org.au/events/2004/security/ |
|
From: Richard G. <ric...@ri...> - 2004-02-04 04:00:06
|
FWIW I apt-get install'ed kde3.1.5 full from scratch and didn't have any issues (that I could see). ----- Original Message ----- From: "John LeSueur" <jo...@su...> To: <col...@li...> Sent: Tuesday, February 03, 2004 6:40 PM Subject: Re: [coLinux-devel] Strange problem with Win-TAP and coLinux > I've seen very similar problems when using apt-get to install large > files(like kernel-source-2.4.24) > The download completes, then dpkg fails because the md5sum doesn't > match. Sounds like corrupted packets getting passed around, which makes > me think it's the tap drivers, or the connection between colinux and the > tap driver... > > It seems to be only associated with large transfers, and not network > load. But at least apache works. and php. and postgres. > > John > > > > > Ian Latter wrote: > > >Hello Vlad, > > > > I haven't seen the issues that you've reported - but I've seen some- > >thing else that's not right in the networking. > > > > The only application included in the CHAOS file system is setiathome, > >and it fails to acquire a new workunit from the Berkeley servers. What > >we see here is an ESTablished connection to their web server > >(which suggests that networking "works") but we never retrieve any > >data (not even up to the first Mbyte). Instead the connection hangs > >with a send-q of about 350 or so. > > > > We're testing this over two PCs (XP Pro SP1) with Bridging (mini mac) > >as the connection/networking mechanism. > > > > I think it was the other Ian who mentioned a much later TAP driver > >from the openVPN project; maybe its worth trying that out .. I might > >give that ago sometime in the next couple of days - see if it helps. > > > > > > > > > >----- Original Message ----- > > > > > >>From: "Vlad Doubrov" <vl...@w-...> > >>To: <col...@li...> > >>Subject: Re: [coLinux-devel] Strange problem with Win-TAP and coLinux > >>Date: Tue, 03 Feb 2004 15:05:03 +0300 > >> > >>Dan Aloni wrote: > >> > >> > >> > >>>>The same happens when I try to transfer that file via SMB (smbd and > >>>>smbclient is installed on coLinux) - except that the file is getting > >>>>corrupted after "successful" download. > >>>> > >>>>Small files (less than 1 MB in size) get transferred very well. > >>>> > >>>>So, is it a bug in Win-TAP driver or coLinux? Any workarounds possible? > >>>>Or may be it's my misunderstood? > >>>> > >>>> > >>>My guess is to try to using ifconfig to change eth0's MAC address. > >>> > >>> > >>It doesn't help. According to SSH docs, "corrupted MAC on input" message > >>means that some packet was corrupted during data transfer, it doesn't > >>mean anything to ethernet MAC addresses. And as I already wrote, when I > >>try to transfer the same large file via samba it gets corrupted the > >>same way. > >> > >>I'm completely stuck out as I couldn't transfer anything larger than 1Mb > >>to the virtual linux. Does this problem appear in your enviroment? Or > >>perhaps some workarounds are possible? > >> > >>-- vlad > >> > >> > >> > >> > >> > >>------------------------------------------------------- > >>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 > >> > >> > >> > > > >-- > >Ian Latter > >Internet and Networking Security Officer > >Macquarie University > > > > Meet me at the Australian Unix and open systems > > User Group (AUUG) Security Symposium; 2004 > > http://www.auug.org.au/events/2004/security/ > > > > > > > >------------------------------------------------------- > >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 > > > > > > > > ------------------------------------------------------- > 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 > > |
|
From: Gregor M. <mi...@ra...> - 2004-02-04 01:46:00
|
Hi, thanks for your answers. The file system error message has gone now. But the network is not working yet. I got the following messages at boot time: modprobe: Can't open dependencies file /lib/modules/2.2.24/modules.dep (no such file or directory) [...] Configuring network interfaces: Ignoring unknown interfache lo=lo. I renamed the 2.4.18-27um folder to 2.4.24. But that didn't solve the problem. (Now I get different error messages) May it be that without the loopback interface the network can't work properly? > Read the mailing list archives :-) Where can I find the archives? The link on the coLinux Homepage is dead. http://sourceforge.net/mailarchive/forum.php?forum_id=37408 ERROR Forum not found Cheers! Gregor John LeSueur wrote: > You'll have to edit the scripts that check the root file system on boot. > the easiest way to do this is to follow the instructi on to remount the > drive read/write, then edit the /etc/init.d/checkroot.sh. Find the line > that looks like: > rootcheck=yes > and change it to > rootcheck=no > > Then reboot. > It should boot normally. > As far as the ethernet interface goes, just configure the tap device in > windows to have the ip address 192.168.0.1. Or you can turn on internet > connection sharing on your real ethernet interface, and it will do this > for you automatically as well as allow your colinux install to contact > the real world. |
|
From: Ian L. <Ian...@mq...> - 2004-02-04 00:56:00
|
Actually thats a good point. We did run the CHAOS web server on the two machines (TinyHTTPD from ACME) and it ran fine, including the WAV file that we serve -- but still nothing over a Mbyte in any of that content. I did have problems with apt-get on the Debian file sysetm, but the errors looked like Gateway Timeout errors, which our transparent caching box does dish out -- so I can't be certain of the root-cause of that one. ----- Original Message ----- >From: "John LeSueur" <jo...@su...> >To: <col...@li...> >Subject: Re: [coLinux-devel] Strange problem with Win-TAP and coLinux >Date: Tue, 03 Feb 2004 17:40:30 -0700 > > I've seen very similar problems when using apt-get to install large > files(like kernel-source-2.4.24) > The download completes, then dpkg fails because the md5sum doesn't > match. Sounds like corrupted packets getting passed around, which makes > me think it's the tap drivers, or the connection between colinux and the > tap driver... > > It seems to be only associated with large transfers, and not network > load. But at least apache works. and php. and postgres. > > John > > > > > Ian Latter wrote: > > >Hello Vlad, > > > > I haven't seen the issues that you've reported - but I've seen some- > >thing else that's not right in the networking. > > > > The only application included in the CHAOS file system is setiathome, > >and it fails to acquire a new workunit from the Berkeley servers. What > >we see here is an ESTablished connection to their web server > >(which suggests that networking "works") but we never retrieve any > >data (not even up to the first Mbyte). Instead the connection hangs > >with a send-q of about 350 or so. > > > > We're testing this over two PCs (XP Pro SP1) with Bridging (mini mac) > >as the connection/networking mechanism. > > > > I think it was the other Ian who mentioned a much later TAP driver > >from the openVPN project; maybe its worth trying that out .. I might > >give that ago sometime in the next couple of days - see if it helps. > > > > > > > > > >----- Original Message ----- > > > > > >>From: "Vlad Doubrov" <vl...@w-...> > >>To: <col...@li...> > >>Subject: Re: [coLinux-devel] Strange problem with Win-TAP and coLinux > >>Date: Tue, 03 Feb 2004 15:05:03 +0300 > >> > >>Dan Aloni wrote: > >> > >> > >> > >>>>The same happens when I try to transfer that file via SMB (smbd and > >>>>smbclient is installed on coLinux) - except that the file is getting > >>>>corrupted after "successful" download. > >>>> > >>>>Small files (less than 1 MB in size) get transferred very well. > >>>> > >>>>So, is it a bug in Win-TAP driver or coLinux? Any workarounds possible? > >>>>Or may be it's my misunderstood? > >>>> > >>>> > >>>My guess is to try to using ifconfig to change eth0's MAC address. > >>> > >>> > >>It doesn't help. According to SSH docs, "corrupted MAC on input" message > >>means that some packet was corrupted during data transfer, it doesn't > >>mean anything to ethernet MAC addresses. And as I already wrote, when I > >>try to transfer the same large file via samba it gets corrupted the > >>same way. > >> > >>I'm completely stuck out as I couldn't transfer anything larger than 1Mb > >>to the virtual linux. Does this problem appear in your enviroment? Or > >>perhaps some workarounds are possible? > >> > >>-- vlad > >> > >> > >> > >> > >> > >>------------------------------------------------------- > >>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 > >> > >> > >> > > > >-- > >Ian Latter > >Internet and Networking Security Officer > >Macquarie University > > > > Meet me at the Australian Unix and open systems > > User Group (AUUG) Security Symposium; 2004 > > http://www.auug.org.au/events/2004/security/ > > > > > > > >------------------------------------------------------- > >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 > > > > > > > > ------------------------------------------------------- > 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 > -- Ian Latter Internet and Networking Security Officer Macquarie University Meet me at the Australian Unix and open systems User Group (AUUG) Security Symposium; 2004 http://www.auug.org.au/events/2004/security/ |
81 messages has been excluded from this view by a project administrator.
