At Thu, 24 Oct 2013 15:38:11 +0200,
Svante Signell wrote:
Well, the question is quite simple: what happens when the sender
provides faked ports, e.g. pointing to other proc/auth servers? That's
where having to explain how the patch is working would possibly even
work out the security
At Thu, 10 Oct 2013 18:08:20 +0200,
Justus Winter wrote:
* console-client/console.c (main): Replace epilogue with console_exit.
---
console-client/console.c |3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/console-client/console.c b/console-client/console.c
index
At Thu, 5 Sep 2013 16:57:41 +0200,
Justus Winter wrote:
I made two rather small and (as I thought) straight forward changes to
gnumach to keep track of a tasks father task and to make this
information available.
What happens when the parent task is destroyed? Are the children
destroyed with
At Tue, 30 Jul 2013 11:59:18 +0200,
Justus Winter wrote:
fsys_get_children returns any active translators bound to nodes of the
receiving filesystem as an argz vector containing file names relative
to the root of the receiving translator.
What if the caller is chrooted? The filenames should
At Tue, 30 Jul 2013 22:44:22 +0200,
Richard Braun wrote:
On Tue, Jul 30, 2013 at 12:20:32PM +0200, Neal H. Walfield wrote:
fsys_get_children returns any active translators bound to nodes of the
receiving filesystem as an argz vector containing file names relative
to the root
-dev
Uploaders: Jeff Bailey jbai...@raspberryginger.com,
Neal H. Walfield n...@debian.org, Michael Banck mba...@debian.org,
Samuel Thibault sthiba...@debian.org
I just noticed this. I'm not a DD any more. And, Jeff hasn't been
active in years. It might be appropriate to prune these entires
At Thu, 4 Jul 2013 17:06:09 +0200,
Thomas Schwinge wrote:
I wonder: if MAP_STACK is set, would it even be
reasonable for mmap to ignore the supplied length, and instead use the
one proper value, 0x20?
I think that this is only acceptable if the length exceeds the
supplied length.
This
For this patch, the amount of data dynamically allocated is not large, a
few strings of size much less than PATH_MAX of 4096 bytes. According to
the manpage for malloc the default MMAP_THRESHOLD is at 128kB.
Alternately asprintf could be used instead of malloc+snprintf, but
asprintf is not
At Tue, 17 Jun 2008 22:49:16 +0200,
Arthur de Jong wrote:
Hello list (I'm not subscribed so please keep me in Cc).
I'm the maintainer of nss-ldapd. I saw that the package was recently
built for GNU Hurd. I haven't tested the package on Hurd but there is
one (not very critical) thing that
At Wed, 18 Jun 2008 12:20:10 +0200 (CEST),
Arthur de Jong wrote:
One question you should consider is: why do you need this information?
[...]
I agree with your point in general and think there are better ways to
do access control.
nss-ldapd is an NSS module that does lookups in an LDAP
At Wed, 18 Jun 2008 12:41:48 +0200,
Neal H. Walfield wrote:
At Wed, 18 Jun 2008 12:20:10 +0200 (CEST),
Arthur de Jong wrote:
One question you should consider is: why do you need this information?
[...]
I agree with your point in general and think there are better ways to
do access
At Wed, 18 Jun 2008 12:56:53 +0100,
Samuel Thibault wrote:
Well, I guess he doesn't have a running Hurd system.
Actually I guess we could easily add SO_PASSCRED to pflocal sockets, by
using auth_user_authenticate/auth_server_authenticate indeed.
The problem is which credentials: remember a
At Mon, 20 Aug 2007 02:30:41 -0400,
Michael Casadevall wrote:
1: I think it should be raised in gnumach itself, not just in the
Debian
version.
I agree. The value should also be dymanically changeable by the
user if possible (I am not familiar enough with the code in question
to know
At Wed, 1 Aug 2007 18:30:20 +0200,
[EMAIL PROTECTED] wrote:
On Wed, Jul 25, 2007 at 09:17:49PM -0400, Michael Casadevall wrote:
Hurdng - the project of porting hurd translators to another
microkernel beside mach such as L4.
That is not fully correct. The original port to L4 was simply
At Sun, 10 Jun 2007 12:15:20 +0200,
Samuel Thibault [EMAIL PROTECTED] wrote:
Neal H. Walfield, le Sat 09 Jun 2007 00:29:38 +0200, a écrit :
The theory is that we don't trust the server to honor the timeout: it
may be malicious and trick the client into waiting forever.
Or it may
At Sat, 9 Jun 2007 01:30:49 +0800,
Samuel Thibault [EMAIL PROTECTED] wrote:
Actually, on the Hurd, a timeout of 0 probably doesn't make sense (since
we at least need to give back cpu to the server). What I'd propose is
the attached patch (not tested), that rounds up the timeout value, and
in
At Sat, 25 Nov 2006 14:34:14 +0100,
Michael Banck wrote:
As setting the stack size is optional (it is only done if
pthread_attr_setstacksize() is available in the first place), mono
shouldn't abort if it fails, either.
That's right.
For more discussion see
Package: Hurd
/dev/{null,zero,full} should not be run as root: they don't need root
access to operate correctly. Because they are translators and the
node is owned by root, the file system starts them with root
permission. Ideally, they should be run with no user ids but as an
interim measure,
At Thu, 9 Jun 2005 10:35:55 +0200,
Oleksandr Shneyder wrote:
But unfortunately I have found new problem - I try to compile my own old
simple qt project to test how qt work. This is very simple project without
processes or threads. With no-threaded qt version all work just fine, but
with
At Tue, 7 Jun 2005 20:18:48 +0200 (CEST),
Santiago Vila wrote:
After an apt-get upgrade today, my Hurd didn't boot properly anymore,
giving this error message:
panic: kmem_suballoc
and rebooting afterwards.
I was able to fix this by reducing the value of uppermem in menu.lst
(the
At Thu, 26 May 2005 12:08:03 +0100,
Colin Watson wrote:
I've fixed all the Hurd build problems in openssh 1:4.0p1-1, which I'll
be uploading to experimental shortly. However, I still can't get sshd to
work, and it's beginning to look like a bug in glibc's Hurd support.
sshd's debug log looks
At Tue, 17 May 2005 23:57:24 +0200,
Michael Banck wrote:
3. There is a pretty severe issue with pthread_attr_setstacksize()
from libgthread resulting in nautilus to abort. I hacked around that
in glib for now, Neal might be able to look into that later on.
Our current libpthread doesn't
At Wed, 9 Mar 2005 18:29:41 +0100,
Michael Banck wrote:
On Wed, Mar 09, 2005 at 05:06:22PM +0100, Samuel Thibault wrote:
Michael Banck, le mer 09 mar 2005 17:58:05 +0100, a dit :
3. Running the Hurd console in a tight endless loop results in mayhem if
the user introduces errors the
Extensibility
Neal H. Walfield
- Interactions in a Multiserver Operating System: The Importance of a
good RPC Framework
Marcus Brinkmann
- L4/Hurd driver model
Peter 'p2' De Schrijver
- GRUB 2
Marco Gerards
- Debian GNU/Hurd
Michael Banck
Thanks,
Neal
--
To UNSUBSCRIBE, email to [EMAIL
I am getting errors with the trial build of the K8 set which was built
using the latest archives.
These messages were copied by hand.
When booting.
hd0: dma_intr: status=0x51 { DriveRady SeekCompleteError }
hd0: dma_intr: error 0x84 { BadCRC DriveStatusError }
LBAsect=4054335,
I just remember there being issues about constructors not being called
if -lpthread was not on the link line. I think perl did this but I
forget.
At Sun, 2 Nov 2003 19:53:31 +0100 (CET),
Santiago Vila wrote:
And this is what happens when compiling python2.3:
gcc -DNDEBUG -g -O3 -Wall -Wstrict-prototypes -fPIC -fno-strict-aliasing
-DWITH_APPINIT=1 -DWITH_BLT=1 -I/usr/include/tcl8.4 -I/usr/X11R6/include -I.
__USE_LARGEFILE64 and __USE_XOPEN_EXTENDED are *not* user macros. You
want to use _FILE_OFFSET_BITS and _XOPEN_EXTENDED respectively.
However, these must be defined before any includes are done and (at
least the former) should be done across all files. I suggest that
they be added to the CFLAGS
but modern *BSDs use a kernel-space Linux approach. are you sure
um-pppd is still maintained?
Last update was on June 19, 2003 [1]. I attempted to submit my
patches to Brian several times, however, I never got any type of
response.
[1] http://www.awfulhak.org/~brian/
Johannes Rohr [EMAIL PROTECTED] writes:
should be enough, not? Again, I only see 'translator died'.
How can I gather further info?
Try setting an active translator (i.e. `settrans -a') so that all
output goes to your terminal instead of /dev/null.
Second: I should probably recompile the
I have typescripts of dumpe2fs and e2fsck available, if anyone would
like to examine them, please let me know.
Sure.
After the failure to send email to you, I've uploaded the logs to
http://home.arcor.de/j.rohr/hurd-logs/hurd-fs-crash-logs.tar.bz2
I have emailed the administrator
I have typescripts of dumpe2fs and e2fsck available, if anyone would
like to examine them, please let me know.
Sure.
I have added what appears to be [1] the latest version of your
translation to [2]. In the future please respect my mail
Mail-Copies-To header when replying to me.
[1] http://std-err.de/hurd-install-guide/german
[2] http://walfield.org/papers/hurd-installation-guide/german.
i have translate the Hurd-Installation-Guide from Neal H. Walfield
to german so this post is writing in german, sorry :)
Thanks, but if you want it to be integrated it, you will need to
translate the texinfo file.
Can you please send the output of `tune2fs /dev/partition' where
This should be `tune2fs -l /dev/partition'
Can you please send the output of `tune2fs /dev/partition' where
partition is the partition on which your GNU/Hurd installation
resides.
If our only alternatives are
1) no ssh
2) ssh with no security
Wrong, which just proves that you have not read this thread: we are
arguing about entropy; ssh is only a side argument.
1) no ssh
2) ssh with no security
you have advocated (2), right? It is that statement which I am
arguing against.
No, I have advocated against including a unsecure random translator.
You are forgetting the third alternative, making ssh use its own
random pool. Assuming
Why do I feel like repeating this old mantra: Bad security is worse
than no security.
Sez you. Many disagree. Especially for a system in development, with
already has bad security.
I think that we can all accept that there are currently a variety of
security holes in the Hurd. The type
Your argument is absurd. Network security considerations are different
than local system security considerations. We have control over who can
have an account on our systems. We don't necessarily have control over
who has access to our IP ports.
These are all excellent reasons to
OK folk, what is causing the problem?
This is not an ext2fs bug; something is causing a Mach panic. The
reason that your file system has been corrupted is that the data was
not synch'ed to disk (when Mach panics, it takes the whole system
down). The correct solution is to fix Mach.
ext2fs should be quite robust: Even pulling the plug at any time should not
corrupt the filesystem beyond what e2fsck can repair.
Let us assume that ext2fs writes a block of metadata to disk. In the
kernel, in the middle of the DMA operation, the kernel panics. There
is no guarantee that
Disk hardware guarantees that a sector write can always be completed
even if the power goes out partway through. That means that writing a
single sector *is* always atomic.
The size of a single sector does not necessarily equal the size of of
a disk block.
I support the addition of ssh, *even* with a weak random.
ssh is not being excluded; it is in the archive, if you want it, you
just have to set it up yourself. By helping the user with this
horrible kludge--essentially installing pregenerated host keys--we are
creating a false sense of
Remove `-lc' from the link line.
Thanks. Is this long-term the right solution? If yes, I'll prepare the
patch to hint/gnu.sh and send it upstream.
I do not know what the problem is but it would seem to be something to
do with symbol resolution. I had hoped that Roland might have been
able to comment on it during the last
At that point, I'm not sure how I can do a backtrace, because session is
disconnected.
Then you should attach via the console.
If you are losing network access then I doubt time is the problem; it
is likely pfinet.
1) /sbin should be added to every users path on i386-gnu systems. The
concept of a binary that is completely unusable for regular users is
almost unheard of for us. (The only few that come to mind is init,
fdisk, and grub).
Why would a normal user not want to run fdisk or grub? Think
You can get up as much as 2gb if I heard other reports correctly. I am
using 1.5gb today. The exact limit is not known. It might depend on where
libraries are loaded.
BTW, the higher this value is the lower is the number and size of files you
can have read/write at the same time. Every
http://www.pick.ucam.org/~mcv21/hurd.html
Old and out of date.
http://www.synack.net/~bbraun/hurd.html
Ancient and out of date.
I just would like to know which guide and mainly download source is most
up-to-date and useable for me.
Please follow the prescribed way. You can find
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
My slides are now online[1].
[1] http://web.walfield.org/papers/better-best-effort-20021026/
apt-get build-dep hurd. You should read the documentation.
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
Now that the Hurd 0.3 is out
It is not. No release has been made.
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
On Mon, Jul 22, 2002 at 10:32:15PM +0200, Andreas Rottmann wrote:
That's a more interesting project IMO (makeing debugging and the like
much easier presumably). However, If you are interested in a HURD on
top of UN*X, I'd like to suggest a different approach: The HURD will
(hopefully) be
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
When I try to boot using the module statements from the installation
instructions (I did copy and paste), it always hangs up. The message
where it hangs up is:
start /hurd/ext2fs.static:
right after loading the modules.
This is pasted from my menu.lst:
--- snip ---
title
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
What am I doing wrong here?
I do not know, however, I think it might have something to do with not
passing --prefix= to configure.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
This is what showed
on the screen directly attached to the Hurd machine:
memory_object_data_request(0x0, 0x0, 0x43000, 0x1000, 0x3) failed,
268435459
When the kernel requests data from a manager and the manager dies
before it can send a response, the kernel emits this message.
... And a
I think that this patch will solve your problem:
I have committed this.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Running dselect or apt-get very often gets my hurd box on it's knees.
When it's 'Reading Package Lists...' , the system hangs, after giving:
ext2fs.static: ../../ext2fs/pager.c:396: file_pager_write_page:
Assertion `block' failed.
Can you send us the output of `tune2fs -l /dev/PART'
I don't know where you got the idea you should use `-T typed'.
-T device is what you need for the store name you are using.
We use -T typed so that root can be set using either the device type,
e.g. device:hd0s2 or the part type, e.g. part:2:device:hd0.
The typos are -T typed and ${root}
These are not typos. There is something wrong with the way oskit-mach
is parsing the arguments.
When you attach gdb the kernel, set a breakpoint at
gdb_break_stub. Then start you kernel (pressing 'c'). Say you want
to debug a function in the kernel where you don't have break point,
start your compiled gdb-break program on the console. Now you should
hit the gdb_break_stub
cp should not copy passive translator settings.
Why not? And what is the right way to copy them?
cp should not copy passive translator settings.
Why not? And what is the right way to copy them?
Because cp is supposed to copy the data of the file; it should read
the data.
For example, on Linux, if you do
mknod /dev/foo ...
cp /dev/foo /tmp/bar
then /tmp/bar is *not* a
Oh, cp -R is different. I'm of two minds about what the Right Thing
is for the cp -R case.
The way to copy it, of course, is to fetch the translator entry and
set it on the copy.
I am not clear what you mean here.
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
Oh, cp -R is different. I'm of two minds about what the Right Thing
is for the cp -R case.
The way to copy it, of course, is to fetch the translator entry and
set it on the copy.
I am not clear what you mean here.
Open the node with O_NOTRANS.
Fetch the translator spec
On Tue, Feb 26, 2002 at 08:16:24PM +0100, Marco Parrone wrote:
Hi all.
Emacs21 for hurd-i386 is linked to xlibs, libpng, libjpeg, libtiff,
libXaw3d (and possibly others), but the package don't depend on them.
I think it is a bug. The i386 (linux) package don't have this problem.
Ok,
On Sun, Feb 24, 2002 at 01:14:44PM -0500, B. Douglas Hilton wrote:
Hi, I followed your instructions and I got the thing to boot like it
is supposed to! I had to make the asm(cli) modification, and in
order to build kernel-ide and kernel-ide+scsi I edited the
top level makefile and did
But the reason why they depend on sysvinit might be many, and I don't have
an overview of what the common reasons are. That I ported sysvinit at all
was a kludge for Debian GNU/Hurd by itself: The Hurd doesn't really
need it.
In that case, the Debian Hurd package could perhaps provide
I'm new to Hurd and I've just installed it. I used the scripts from
debian.org for installation.
Those instructions are a bit old (and the procedure they describe is
no longer supported). Can you please try the updated documentation
found here [1]?
[1] http://www.gnu.org/software/hurd
I didn't say it was difficult. It's easy. It's also silly. Why
should I worry all the time about such adaptation?
How many languages do you want me to support?
C++ is an *incompatible* extension of C. Extension yes, compatible,
no.
Just a weeks ago, you were arguing on one of the
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
I intend to build the H3 images in the next few days.
Are there any issues I should be aware of?
This could well be the last set before ABI is implemented.
We need a new emacs21 package. I just compiled it and it
worked, however, I then managed to mess up the .deb file. I will redo
the
There was a discussion about some bugs in processing empty file names
but while reading it i didn't notice the special nature of these file
names.
We were referring to the file inside of a symlink (i.e. what
happens when we look up a symlink whose target is ; `ln -s foo').
Here you are
Reducing memory from 768meg to 512meg did the trick tho.
GNU Mach is known to work with up to 640 MB of ram and fail with more
than 768 MB.
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
Has someone made a floppy image or three that can format a HURD partition
and start an ftp install, sort of like FreeBSD does?
There is the cd image, but no floppies -- at least not yet.
In the GNU Hurd Hardware Compatibility Guide
(http://www.urbanophile.com/arenn/hacking/hurd/hurd-hardware.html) I
read that the Debian GNU Mach package does NOT include the NCR
drivers..
I have an NCR scsi controller, so where can I found the drivers?
(maybe one CD)
1) Use the updated HCL
Can i get a vi editor for GNU/HURD?
Yes, use apt.
Can i configure X for Hurd?
Yes, read the install guide and the FAQ.
Are you saying that this stuff is still missing from the kernel? I certainly
hope not ;-) Userlevel stack allocation is handled within NGPT already.
Looking at the threading interface posted on the Hurd website, I don't think
it will take much to port NGPT over to Hurd.
Are you
When i try to mount the CD with
settrans -a /cdrom /hurd/isofs /dev/hd2
i get the following errors:
/hurd/isofs: Could not find valid superblock
settrans: /hurd/isofs : Translator died
Did you actually make the /dev/hd2 device?
nano editor is not working well, it is getting loaded,
but
Has anyone got ppp running on hurd. I am a newbie and I tried
compiling the source which I got from the ftp site from the sid tree.
when I ran ./configure it said that hurd system was not supported. I
would appreciate some pointers to docs/sites where it is given in
detail.
The one ppp
This is not something that can be done this way in Unix, so a program would
not expect it to work the way it is written above, so in the Unix world this
is not a useful feature to have.
Try this:
[EMAIL PROTECTED]:~ (0)$ mkfifo foo
[EMAIL PROTECTED]:~ (0)$ while /bin/true; do
You are suggesting to make Yet Another Personal Hurd Site but this is
exactly what I was arguing against.
This is not what he is suggesting.
My point was that people should
contribute to some central site more than to their own sites.
I can't contribute myself now because I am a newcomer. I
kernel /boot/gnumach.gz -s root=device:hd0s6
module /hurd/ext2fs.static --multiboot-command-line=${kernel-command-line}
module /lib/ld.so.1 /hurd/exec $(exec-task=task-create)
I've got stuck in
Well, you need to actually use the entire command line in -- not just
the first 70 characters.
Welcome to the GNU Hurd
===
Welcome to the GNU Hurd. This email is automatically sent at the
begining of each month to the help-hurd@gnu.org and
debian-hurd@lists.debian.org mailing lists. This message is intended
for a quick
:
2001-11-24 Neal H Walfield [EMAIL PROTECTED]
* pager.c (find_block): Take a new argument, CREATE. Pass
this to ext2_getblk instead of assuming that the callee never
wants the indicated block block to be allocated.
(file_pager_read_page): Conform to the new
1. Has anybody thought about partial delegation of networking? Does
that make sense at all?
I have some vague ideas. I think it does make sense, but making it
work seems hard. But if someone prods me, then I can describe my
vague ideas.
Prod, prod.
3. Is there a reasonable way to
I don't know if that works, because now I'm not abble to just use wget or
lynx.
This is what you are getting when you use diald or when you connect
using straight ppp?
Try starting with something a bit more simple, for instance, pinging
your gateway. Then, move on to pinging a host on the net
This behavior is normal: the login shell runs with no uids and no
gids, however, this does not means that it has absolutely no
permission on the system (at the very minimum, it has access to the
CPU, memory and a tty, thus, why not disk space too). In a default
installation of GNU, the S_IUSEUNK
I've try with other labels than 'ondemand' and with other devices
(original /dev/cuaa1 - /dev/ttyS0, /dev/tty00, /dev/tty01)
but I always got the same errors.
Try /dev/com0.
So, can I connect my GNU/Hurd box to the net with an ISP ?
Connecting should not be a problem. However, once you
But I have a little probleme: I can't make a device /dev/tun0 (or N)
with ./MAKEDEV tun0 in /dev like it's said in the ppp man pages.
Well, the man pages are BSD centric. In the Hurd, pfinet will
automatically create the tunnel device for you. Just run
settrans -acpfg
1 - 100 of 349 matches
Mail list logo
