Re: Chaos Congress 2003

Peter 'p2' De Schrijver <[EMAIL PROTECTED]> writes: > Anyone going to the Chaos Congress in Berlin this year ? I will be there, at least occasionaly. In case you'll be at a dutch supermarket, could you buy few jars of Calve Pindakaas for me? There's not much left here. :) moritz

Re: high temperature with gnumach and hurd

"Alfred M. Szmidt" <[EMAIL PROTECTED]> writes: > Sounds like the machine_idle patch that never go applied when GNU Mach > 1.3 was released. Uhm, what about: 2002-05-23 Marcus Brinkmann <[EMAIL PROTECTED]> * i386/i386at/model_dep.c: Include and . (machine_idle): New fu

Re: Status of the ppc port

Robert Millan <[EMAIL PROTECTED]> writes: > when storeio and pfinet work (how do you use it without storeio?) Filesystems can be accessed without storeio - libstore is enough. Peter: Great work. moritz -- ((gpg-key-id . "6F984199") (email . "[EMAIL PROTECTED]") (webpage

Re: console client no workee

Roland McGrath <[EMAIL PROTECTED]> writes: > hurd:~# console -d ncursesw /dev/vcs/1 That should be /dev/vcs, not /dev/vcs/1. moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 __

Re: init: cannot reboot

Lionel Elie Mamane <[EMAIL PROTECTED]> writes: > Launched the console server. [BANG] Are you able to reproduce that? How did you start it? moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 ___

empty symlinks, e2fsck

Hi, I created an empty symlink on an ext2 filesystem while I was in GNU/Hurd. Later on that partitions was checked by e2fsck (in GNU/Linux) and when it encountered that symlink, it reported a filesystem inconsistency, which I had to repair manually. That doesn't seem so sane to me, either ext2fs

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > That's actually not a problem, because each walk through the union > fs requires a retry. The library is supposed to keep track of how > many retries it gets, and handle ELOOP itself. Still, if you imagine many that users create a unionfs based

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > What we really want is for the user to do a retry of the name as it > exists in the "real" location, and then if that results in ENOENT, > we want the user to return back to the filesystem for another name > to try. Well, here you are only consid

Re: ftpfs broken ?

You can get some helpful information out of ftpfs like this: Start an active ftpfs instance: $ settrans -afg node /hurd/ftpfs --HANG= Now you have seconds left to attach gdb to the ftpfs process on a different terminal: get the PID via ps and then: $ gdb /hurd/ftpfs Let the process continue

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > Oh, that. Blech blech blech. Blech is also corking. > And, of course, this matters in just this case! Because it's a > union, and so the node is found in *two* directories and it's not at > all clear which one is right. I'm not sure wether I

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > What exactly would the problem be there? Maybe I've missed a beat > in the conversation. Maybe I am overlooking something, I am not that familiar with libdiskfs. My question is: given the situation that dir_lookup is called to re-open a node, w

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > I think the right fix is to have lookups for "" do all the normal > processing when you open a file. Well, yes, but the problem of relative symbolic links is not yet solved, is it? moritz -- [EMAIL PROTECTED] - http://duesseldor

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > I think that's why I originally stated "." which Roland corrected to > "". Well, "." would not work for non-directories, of course. > It might well be that we have a hole in the interface here. Blech. So... fs interface change - anyone? =)

Re: Unionfs, looking up links and translators

[EMAIL PROTECTED] (Thomas Bushnell, BSG) writes: > Instead, you fetch the actual node, and then tell the user to reauth > *that* node. Are you sure the needed functionality is implemented? I tried that, it does not work (with a retry name of ""); the user keeps the underlying node, he doesn't ge

Unionfs, looking up links and translators

Hi, I was thinking about what unionfs should do with symbolic links and translators on the underlying filesystems; i think if unionfs's _S_dir_lookup would return retry names in that case, that would be reasonable. It would create some problems if unionfs would simply use file_name_lookup (wi

Re: Unionfs

Wolfgang Jaehrling <[EMAIL PROTECTED]> writes: > /usr/bin/ld: netfs.o(.debug_info+0x6399): unresolvable relocation > against symbol `_netfs_translator_callback1' I am not sure what this is; I don't see it here. > I noticed that lnode_ref_remove() and lnode_uninstall() recursively > call each oth

Unionfs

Hello, after not doing any work on my former shadowfs implementation anymore for many months, I started working on a new implementation (`unionfs') few weeks ago. It is quite uncomplete at the moment and has not few problems. Compared with my former implementation, this one uses a completely n

Re: Mach Interface Generator

"Frank Saar" <[EMAIL PROTECTED]> writes: > Are there any good documents for understanding the mach interface > generator out there ? Here you will also find information about MIG: http://www-2.cs.cmu.edu/afs/cs/project/mach/public/www/doc/osf.html. moritz -- [EMAIL PROTECTED] -

proposing libnetfs interface change

Hi there, there are (or will be) at least two Hurd servers, which hand out "forein" ports to the user - ports, not managed by themselves. * socketio gives out ports managed by the socket server, * unionfs gives out ports to the underlying filesystems. Instead of letting these servers imp

file_chmod in unionfs, copy-on-write

Hi, just out of curiosity: how should the yet-to-be-written unionfs implement copy-on-write for file_chmod()? unionfs usually gives out ports to the underlying files; when in copy-on-write mode, it should give out a port to which the user can safely write without modifying the `real' data. I

Re: implementing bash's magic

Marcus Brinkmann <[EMAIL PROTECTED]> writes: > Ok, this is simple enough to be in one single file like fakeroot in > directory trans/. Ok, done: http://duesseldorc.ccc.de/~moritz/files/socketio.c.gz. There are some minor improvements since netio version 0.2. io_restrict is not yet used. There a

libstore/kids.c question

Hi, libstore/kids.c contains the following code: /* Set STORE's current children list to (a copy of) CHILDREN and NUM_CHILDREN. */ error_t store_set_children (struct store *store, struct store *const *children, size_t num_children) { unsigned size = num_children * sizeof

Re: libiohelp

Michael Koch <[EMAIL PROTECTED]> writes: > I wonder why this bug is known/reported for more then three months > and nobody fixed it. It was fixed for quite some time, but now the bug is back it seems. It would be good, if somebody could apply that trivial fix soon... moritz -- [

Re: [PATCH] ftpfs

Just curious; was there something wrong with that patch? moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 ___ Bug-hurd mailing list [EMAIL PROTECTED]

configuration files for Hurd servers?

Hi, was there ever a plan to make Hurd servers support configuration files? Right now the only way to configure them is via command line options. I think it could be convenient in the future to have them support configuration files; there could be per-server files and files, which apply to all s

[PATCH] libftpconn

The problem, which I introduced with my patches, was that I used dirname() without keeping in mind that the original path is modified after using that function. Here is a patch against libftpconn/unix.c. 2002-10-17 Moritz Schulte <[EMAIL PROTECTED]> * unix.c (ftp_conn_unix_start_get_s

Re: implementing bash's magic

[Lookups like /dev/{tcp,udp}/host/port] I have written a translator, which implements exactly this: netio. Can be fetched from http://duesseldorf.ccc.de/~moritz/files/netio-0.2.tar.gz. moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 15

Re: [PATCH] ftpfs/libftpconn

[EMAIL PROTECTED] (Niels Möller) writes: >> I haven't found a way in the FTP protocol, which helps us here. > > In theory, the command "mlst" should help. Aah, this seems to be a very good idea in the FTP protocol, especially because it's easy to parse. If that is well supported by FTP servers,

[PATCH] ftpfs/libftpconn

t my patch for libftpconn does. The patch for ftpfs does two things: 0) disable 'list's of the root node (that doesn't make much sense anyway); 1) initialize the root node's stat information. I hope everything is alright. ftpfs: 2002-04-13 Moritz Schulte <[EMAIL PROTECTED]

Re: Hurd logging. (was zalloc panic)

Marcus Brinkmann <[EMAIL PROTECTED]> writes: > Of course, we really have to look at the specific case, there is > probably no blanket statement for such things. All errors that can > be recovered from should of course not cause instant server death. Yes, I agree absolutely. mor

Re: Hurd logging. (was zalloc panic)

Marcus Brinkmann <[EMAIL PROTECTED]> writes: > I think that would only even more obfuscate the situation you are > trying to debug. The point I try to make is that it's imho better to let the user decide what should happen in such a situation and give him at least the chance to get control over

Re: Hurd logging. (was zalloc panic)

Marcus Brinkmann <[EMAIL PROTECTED]> writes: > I would say that if you recognize a situation that would make all > further writes (to a file, to a socket) trigger another emergency > log (or, let's say there is a significant likelihood for it), it > would make sense for the server to log a messag

Re: Hurd logging. (was zalloc panic)

Jon Arney <[EMAIL PROTECTED]> writes: > The possibility exists for users to flood the log port. Just an idea... What I think would be pretty cool: let the logging function use /servers/log (or whatever) if the server is running as root and ~/servers/log (or whatever) otherwise. This would be q

Re: Panic: zalloc!

Jon Arney <[EMAIL PROTECTED]> writes: >> When we use oskit-mach we can get better data with gdb over serial >> line. > > If you will be switching uKernels, does this mean that the zalloc > problem is not worth investigating since it will probably go away > with gnumach, or am I missing something.

netio-0.1

Hi there, I've written a small translator "netio", which provides a filesystem interface to sockets. The sockets are created by lookups of the form {tcp,udp}/HOST/PORT. IO on the resulting IO port triggers IO on the underlying socket. With some bash redirection tricks you can have a simple net

L4 meets the Hurd at the Congress in Berlin

Hi there, the CCC congress was fun for me, I hope for the others, too. Not only that there were some Hurd people, but also two L4 hackers from Karlsruhe. Neal had his really nice talk on the Hurd and after that the L4 people talked spontaneouslly a bit about L4. Jeroen and me tried to summariz

Re: httpfs, tarfs --help

Jeroen Dekkers <[EMAIL PROTECTED]> writes: > Running gzippped and bzip2'd binaries is already supported in exec > IIRC. Well, that support - at least for gzipped binaries - is broken. moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 1

Re: libnetfs --examples

arun v <[EMAIL PROTECTED]> writes: > Can any one suggest some good sample pgms. implemented using > libnetfs. $ find . -name "*.c" | xargs grep "netfs_init ();" ./ftpfs/ftpfs.c: netfs_init (); ./hostmux/hostmux.c: netfs_init (); ./nfs/main.c: netfs_init (); ./usermux/usermux.c: netfs_init ()

Re: tar file system

Jeroen Dekkers <[EMAIL PROTECTED]> writes: > You can find the current format used by GNU tar in the comments of > the code IIRC. It's info page describes the different formats too. moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 15BE

Re: tar file system

arun v <[EMAIL PROTECTED]> writes: > Basically what libraries, i need to use -- i think libdiskfs will be > needed! No, you should use libnetfs, because the nodes in the directory hierarchy provided by tarfs are "virtual" - in a way. > How can i extract the tar files to directory? I think you'

Re: [PATCH] symlink lookups in ext2fs

Hi, hmm, is there anything wrong with that patch? Please tell me if that is the case, so that I can fix it. moritz -- [EMAIL PROTECTED] - http://duesseldorf.ccc.de/~moritz/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 _

Re: [PATCH] symlink lookups in ext2fs

Roland McGrath <[EMAIL PROTECTED]> writes: > A patch without a bug report is almost always ignored. Sorry, I thought it was obvious, since it's a little change. Here the bug, now in verbose mode: moritz@hurd:~$ touch foo moritz@hurd:~$ ln -s foo foo2 moritz@hurd:~$ ln -s $PWD/foo foo3 moritz@h

[PATCH] symlink lookups in ext2fs

Hi, 2001-12-03 Moritz Schulte <[EMAIL PROTECTED]> * libdiskfs/dir-lookup.c (diskfs_S_dir_lookup): Don't forget the trailing slash when looking up absolute symlinks. --- dir-lookup.c.orig Mon Dec 3 17:35:09 2001 +++ dir-lookup.cMon Dec 3 17:36:04 200

Re: st_nlink of directory nodes in shadowfs

g that you how suggested would be more or less a rewrite.. moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ "In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." - Linus Torvalds. GPG fingerprin

Re: st_nlink of directory nodes in shadowfs

le directory lookup... moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ "In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." - Linus Torvalds. GPG fingerprint = 3A14 3923

Re: st_nlink of directory nodes in shadowfs

dd and we would calculate a st_nlink, which can be to high, because Shadowfs has always only one directory entry. That was my first question: wether that is acceptable. moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ "In short: just say NO TO DRU

Re: st_nlink of directory nodes in shadowfs

t the performance aspect. If there are shadowed directories, which contain _many_ entries and the user simply looks up that directory in shadowfs, shadowfs would have to first read _all_ the directory entries of _all_ of the underlying directories just to provide the 100% right st_nlink number.

Re: st_nlink of directory nodes in shadowfs

high. So, that is acceptable? moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ "In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." - Linus Torvalds. GPG fingerprint = 3A14 3923 15

st_nlink of directory nodes in shadowfs

be able to rule out duplicate names and therefore provide the correct st_nlink number of a looked up virtual directory node. moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ "In short: just say NO TO DRUGS, and maybe you won't end up like

Bug#116317: Assertion failure in ext2fs/libports

Package: Hurd Version: N/A Severity: normal I just got this crash: ext2fs.static: ../../libports/port-ref.c:31: ports_port_ref: Assertion 'pi->refcnt || pi->weakrefcnt' failed. It seems it's not reproducible. moritz -- Moritz Schulte <[EMAIL PROTECTED]> htt

root node in shadowfs

locking problem or are there better approaches? Thanks, moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ "In short: just say NO TO DRUGS, and maybe you won't end up like the Hurd people." - Linus To

Re: Shadowfs - some notes

mentation, creating all the directories on > read-only lookups seems to be what BSD does. Seems a little iffy to > me, but if it's ok for them I guess it's not so bad. I think it's a good way. It makes everything much easier. Instead of caring about creating whole directory h

Shadowfs - some notes

system, where newly created files in shadowfs should actually be stored. moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ Debian/GNU supporter - http://www.debian.org/ http://www.gnu.org/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199

Bug#115311: Assertion failure in libdiskfs while moving directory

Moritz Schulte <[EMAIL PROTECTED]> writes: > ext2fs.static: ../../libdiskfs/dir-renamed.c:140: diskfs_rename_dir: > Assertion `tmpnp == fdp' failed. Fix: --- dir-renamed.c.orig Fri Oct 12 05:35:48 2001 +++ dir-renamed.c Fri Oct 12 04:40:32 2001 @@ -137,12 +137,12 @@

Bug#115311: Assertion failure in libdiskfs while moving directory

re # CRASH. moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ Debian/GNU supporter - http://www.debian.org/ http://www.gnu.org/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 ___ Bug-hurd mai

ports in symlink server (was: Re: ssh as user: setrlimit failed)

_send_t; uids: idarray_t SCP; gids: idarray_t SCP); hth, moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ Debian/GNU supporter - http://www.debian.org/ http://www.gnu.org/ GPG fingerpr

Re: Adding new device support

cs.utah.edu/flux/oskit/ moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ Debian/GNU supporter - http://www.debian.org/ http://www.gnu.org/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 _

[PATCH] init crashing in a subhurd

CH_SEND_INVALID_DEST) goto procbad; Bye, moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ Debian/GNU supporter - http://www.debian.org/ http://www.gnu.org/ GPG fingerprint = 3A14 3923 15BE FD57 FC06 B501 0841 2D7B 6F98 4199 __

Re: error compilimg hurd cvs source

27; to '-O2' helped (it seems to be a bug in gcc). You can do that in the file 'Makeconf' in the Hurd source tree. hth, moritz -- Moritz Schulte <[EMAIL PROTECTED]> http://www.chaosdorf.de/moritz/ Debian/GNU supporter - http://www.debian.org/ http://w

Details about a ShadowFS Implementation

" is accessed. IMHO, there need to be a way to get the _name_ ("bar"), not the node struct, of the accessed directory. Then i can simply map it to the shadowed filesystem by appending it to the names of these filesystems: => /path[12]/bar Or is this a totally wrong approach? How sh