Re: IPv6?

> Depending on how you are attacking this, it was already done or should be > done differently. how should it be done differently? > But it would also be interesting to implement /dev/random as a user space > server. Some user space random generator could be written as translator.

Re: Some questions...

> This change introduces new features, which are difficult to revert when > people start to use them. So I will not make the decision to apply this i am new to this: who actually has write CVS access, and who decides the patches and additions that make it into the CVS tree

Re: IPv6?

From: Marcus Brinkmann <[EMAIL PROTECTED]> Subject: Re: IPv6? Date: Tue, 12 Sep 2000 23:54:28 +0200 > Of course, for a user space driver, this is more difficult. You could have > an entropy device in gnumach, which provides raw entropy to the user space > driver. I agree. If you want to have a

Re: IPv6?

> Of course, for a user space driver, this is more difficult. You could have > an entropy device in gnumach, which provides raw entropy to the user space > driver. the first way to try is the above, since gathering the entropy is a different procedure then hashing it out to

Re: installing debian on hurd with dselect -> broken system

On Tue, Sep 12, 2000 at 09:55:31PM +0200, Oliver Kowalke wrote: > Hi, > > I've installed hurd from Marcus tar file. Chances are that you suffer from some old bug. I urgently need to get a new tar file done. Please always update the hurd package first and reboot when using the old tar file. > Al

Re: apt-get really works

On Tue, Sep 12, 2000 at 03:16:41PM -0700, Brent Fulgham wrote: > Looks like we may need either a revised AC_ macro or some different > configure logic to slam the hurd-i386 in there when we have a GNU > operating system. Done, see dpkg CVS or: http://cgi.debian.org/cgi-bin/bugreport.cgi?bug=39227

RE: apt-get really works

> You are correct, that did not work, however, my latest hack works > wonderfully... After configure runs, change the ARCHITECTURE= line > in include/config.h from: > > ARCHITECTURE=i386 > > to > > ARCHITECTURE=hurd-i386 > Hooray! Looks like we may need either a revised AC_ macro or some dif

experimental apt packages available

Hi, so, apt everyone? Please test them. ftp://alpha.gnu.org/gnu/hurd/debian [EMAIL PROTECTED]:/home/ftp/gnu/hurd/debian$ ls -l *apt* -rw-r--r--1 marcus ftp 297 Sep 12 18:03 apt_0.3.19.0.dsc -rw-r--r--1 marcus ftp627939 Sep 12 18:03 apt_0.3.19.0.tar.gz -rw-r--r--1

apt-get really works

Hi, > > Plus, the architecture issue... > > I got my changes into dpkg CVS, and apt archtable says it is mirrored from > dpkg, so we have a good chance :) You are correct, that did not work, however, my latest hack works wonderfully... After configure runs, change the ARCHITECTURE= line in inclu

Re: IPv6?

On Tue, Sep 12, 2000 at 09:42:16PM +0200, Niels Möller wrote: > Marcus Brinkmann <[EMAIL PROTECTED]> writes: > > > A GNU Mach char glue code was added by Okuji Yoshinori, with a few bug fixes > > by me. YOu can get it from alpha.gnu.org/gnu/hurd/marcus/ (actually in a > > subdirectory of it, I for

Re: apt-get works with a hack (msync)

On Tue, Sep 12, 2000 at 12:11:21PM -0700, Brent Fulgham wrote: > > I promise that your version of apt is broken. For example, it > > doesn't have the right architecture set (it will try to download > > i386 packages). See http://bugs.debian.org/39227. Brent was > > reporting some other > > stuff

Re: Some questions...

On Tue, Sep 12, 2000 at 02:06:33PM -0500, rt wrote: > > > This change introduces new features, which are difficult to revert when > > people start to use them. So I will not make the decision to apply this > > i am new to this: who actually has write CVS access, and who > decides the

Re: IPv6?

On Tue, Sep 12, 2000 at 02:04:09PM -0500, rt wrote: > > Depending on how you are attacking this, it was already done or should be > > done differently. > > how should it be done differently? > > > But it would also be interesting to implement /dev/random as a user space > > server. Some use

installing debian on hurd with dselect -> broken system

Hi, I've installed hurd from Marcus tar file. All works (booting ...) till I tried to install additional debian packages from ftp.de.debian.org via dselect. If I reboot hurd it hangs on mounting the hdd devices (someting like bad ...block - sorry I can't remeber the hole message). So I had to do t

Re: IPv6?

Marcus Brinkmann <[EMAIL PROTECTED]> writes: > A GNU Mach char glue code was added by Okuji Yoshinori, with a few bug fixes > by me. YOu can get it from alpha.gnu.org/gnu/hurd/marcus/ (actually in a > subdirectory of it, I forgot the name of the subdir). You also need > streamdev by Okuji. The tr

Re: apt-get works with a hack (msync)

Hi, > I experienced socket troubles with certain socket options were not > supported under the Hurd. I know Roland made some updates for this, > so perhaps these problems are just a bad memory now... I have done several `apt-get update's over both http and ftp. I also did an apt-get source. I

RE: apt-get works with a hack (msync)

> I promise that your version of apt is broken. For example, it > doesn't have the right architecture set (it will try to download > i386 packages). See http://bugs.debian.org/39227. Brent was > reporting some other > stuff I forgot > about, and we have to check if it can compile the GUI. > I e

Re: Some questions...

> patch all alone, and the Hurd authors are currently not doing much Hurd > stuff, so we have to wait for an official word on it. also, (sorry), is there a developers mailing list? is there a *need* for one? rt -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subje

Re: apt-get works with a hack (msync)

On Tue, Sep 12, 2000 at 11:22:27AM -0500, Neal H Walfield wrote: > I have been hacking at apt-get and have discovered that all that is required > to get it working under the hurd is one of two things: > > 1) The attached two line hack which gets rid of the msync support > which the hurd claim

Re: Some questions...

On Tue, Sep 12, 2000 at 01:02:51PM +0200, Julio Merino wrote: > Another thing... I've seen that the gnumach package now includes color > support :-)) But I don't see it in the gnumach cvs which I have > recently download. The file is still the same, without color. Why? Because color support really

Re: IPv6?

On Fri, Sep 08, 2000 at 01:05:26PM -0500, rt wrote: > > yes. i am currently bringing /dev/random over from > linux. Depending on how you are attacking this, it was already done or should be done differently. A GNU Mach char glue code was added by Okuji Yoshinori, with a few bug fixes

Re: root-device

Thanky dudes for supporting me that fast and helpful (I _really_ am amazed!) but I still have no real pointer to the resolution. I use "mke2fs v1.18" and my hurd is a "hurd_2803.deb" from (as well the other *.debs are). I think these should be quite up-to-date - but

Re: root-device

On Tue, 12 Sep 2000, Neal H Walfield wrote: > > What is the effect of this -s? I forgot to pass it, but everything seem to > > work fine anyway. > > -s boots to single user mode so that you can run native-install. But I ran native-install. (I must have, the system works.) --

Re: root-device

> What is the effect of this -s? I forgot to pass it, but everything seem to > work fine anyway. -s boots to single user mode so that you can run native-install. -- Neal H Walfield University of Massachusetts at Lowell [EMAIL PROTECTED] or [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL

Re: root-device

On Tue, 12 Sep 2000, Neal H Walfield wrote: > If this is your first boot, why are you not passing -s? (Not that that > would solve this problem) What is the effect of this -s? I forgot to pass it, but everything seem to work fine anyway. -- Nikla

Re: IPv6?

On Mon, Sep 11, 2000 at 04:29:16PM -0500, rt wrote: > > ipv6 is a special case because ipv4 and ipv6 are intimately related. I > > think you want the code to be shared so the server can dtrt for "v4 compat" > > IPv6 addresses. > > > most ip stacks are implemented like this, so i was planni

apt-get works with a hack (msync)

Hi Marcus et al, I have been hacking at apt-get and have discovered that all that is required to get it working under the hurd is one of two things: 1) The attached two line hack which gets rid of the msync support which the hurd claims to have but appears not to (at least not fully). N

Re: root-device

On Tue, Sep 12, 2000 at 03:13:23PM +0200, Florian Zimmermann wrote: > > > On Mon, 11 Sep 2000, Neal H Walfield wrote: > > > Hi, > > > > Sounds like your root partition is using features that the hurd does not > > support. Try passing the -O sparse_super to mke2fs. > > > > -Neal > > Thanks!

Re: root-device

> Sorry for speking some fussy. > Before gnumach.gz is loaded there are outputs which ide.devices have been > found, > therefore the > > hd0s1 hd0s2 > hd1s1 hd1s2 hd1s3 > > which I can see. > Then, as you said, gnumach is not able to read the root partition and prompts > for it. > I

Re: root-device

On Tue, 12 Sep 2000, Neal H Walfield wrote: > > Thanks! I did once again (mke2fs -O sparse_super -o hurd /dev/hdb3) > > -> doesn't work. > > Hurd printk's the following: > > > > hd0s1 hd0s2 > > hd1s1 hd1s2 hd1s3 > > > > where hd2s3 IS filled with /boot/gnumach.gz and the other files. > > I

Re: Problem solved! - sort of. (was Re: kernel halts during first boot)

--- Cowboy <[EMAIL PROTECTED]> wrote: > > >> From: Prabhu Ramachandran > <[EMAIL PROTECTED]> > >> > >> hi, > >> > >> > "Me" == Me <[EMAIL PROTECTED]> > writes: > >> > >> Me> I encountered this problem once. It was > when my pnp ne2000 > >> Me> was enabled. I think after scsi it tries

Re: root-device

> Thanks! I did once again (mke2fs -O sparse_super -o hurd /dev/hdb3) > -> doesn't work. > Hurd printk's the following: > > hd0s1 hd0s2 > hd1s1 hd1s2 hd1s3 > > where hd2s3 IS filled with /boot/gnumach.gz and the other files. I do not understand what you are trying to say. Where did it printk

Re: root-device

On Mon, 11 Sep 2000, Neal H Walfield wrote: > Hi, > > Sounds like your root partition is using features that the hurd does not > support. Try passing the -O sparse_super to mke2fs. > > -Neal Thanks! I did once again (mke2fs -O sparse_super -o hurd /dev/hdb3) -> doesn't work. Hurd printk's t

Re: Problem solved! - sort of. (was Re: kernel halts during first boot)

>> From: Prabhu Ramachandran <[EMAIL PROTECTED]> >> >> hi, >> >> > "Me" == Me <[EMAIL PROTECTED]> writes: >> >> Me> I encountered this problem once. It was when my pnp ne2000 >> Me> was enabled. I think after scsi it tries to detect the >> Me> network cards, if one is enabled

Some questions...

Hi all, I've get the base/gnumach and base/hurd new packages that I have found in the ftp site. Then, I've cleaned completly my hurd hard drive and I've grown it's partition. It was about 400 Mb before and now it is about 523 Mb (528 cylinders). First, gnumach gives me errors complaining about th

Problem solved! - sort of. (was Re: kernel halts during first boot)

hi, > "Me" == Me <[EMAIL PROTECTED]> writes: Me> I encountered this problem once. It was when my pnp ne2000 Me> was enabled. I think after scsi it tries to detect the Me> network cards, if one is enabled that doesn't match the 3com Me> drivers that are in the kernel it stop

Re: kernel halts during first boot

- Original Message - From: Prabhu Ramachandran <[EMAIL PROTECTED]> To: Stefan Rieken <[EMAIL PROTECTED]> Cc: Sent: Monday, September 11, 2000 7:53 PM Subject: kernel halts during first boot > hi, > > > "Stefan" == Stefan Rieken <[EMAIL PROTECTED]> writes: > > Stefan> The HURD

Conference ?

Hello, In June, I wanted to organized a Hurd conference in Paris but it was a little late. But, now, we have newers members and we think again about a Hurd conference. We would like to organize it for June 2001. But maybe it's a little late again. We need yur opinion ! Do you think you could com