in.tftpd: Missing a critical feature
Package: netstd
Version: 3.07-1
(Specifically, in.tftpd)
Hello,
This is going to be a rather long message, so please bear with me.
I'm going to describe a missing feature from Debian's in.tftpd, why we
need it, and give you a patch to implement it.
Sun's tftpd has an option, -s. Basically, this option chroot's to the
specified directory, than set[ug]id's itself to nobody. After that,
it starts serving files.
Why is this important? Well, we have a number of NCD X terminals.
These terminals try to get files with absolute pathnames like
/usr/lib/X11/ncd/configs. The tftp server should really serve this
file from an appropriate subdirectory below the tftpboot directory --
NOT the system's /usr/lib/X11 directory. That is, the server should
actually hand the client the file
/usr/local/ncdroot/usr/lib/X11/ncd/configs in our case.
The Sun tftpd allows one to start in with -s /usr/local/ncdroot.
This will chroot and then serve files as normal. This method doesn't
work with Debian -- any directory specified with Debian's in.tftpd is
only used for permission checking when the client gives an absolute
method. So all Debian's will do is say that the client has no
permission to access that particular file (it is looking under the
system's /usr/lib/X11 area instead of the tree for the NCD), and
things will fail. The client cannot be changed to use other files.
The net result is that Debian's tftp renders those clients useless
when they try to boot using it.
I really don't know why there is this omission from Debian's tftp
program. I have made a patch to add it in, included below.
Sun's manpage describes the -s option as:
-s Secure. When specified, the directory change must
succeed; and the daemon also changes its root directory
to homedir. This option is set in the default
/etc/inetd.conf file.
The use of tftp does not require an account or password
on the remote system. Due to the lack of authentica-
tion information, tftpd will allow only publicly read-
able files to be accessed. Files may be written only
if they already exist and are publicly writable. Note:
this extends the concept of public to include all
users on all hosts that can be reached through the net-
work; this may not be appropriate on all systems, and
its implications should be considered before enabling
this service.
Below is my patch to fix the problem. I was unable to find any
upstream maintainer e-mail; please forward it upstream.
A few notes. When using -s, in.tftpd must be started as root from
inetd.conf. It will subsequently change its uid,gid to 65534 (nobody,
nogroup). This value is hardcoded. Should it be? The root access is
necessary to be able to call chroot(2).
I have not modified the manpage. This ought to be done.
One cannot use a line in inetd.conf like
chroot /usr/local/ncdroot /usr/sbin/in.tftpd because the system
then looks for the in.tftpd program in
/usr/local/ncdroot/usr/sbin/in.tftpd, and of course doesn't find it.
(Worse -- in some cases it could find a binary for a different
platform.) So the -s support cannot be emulated by other means.
I would apprreciate it if you would integrate this patch into future
releases. This will save us the effort of having to patch tftpd each
time netstd (or tftpd, C libraries, or whatever) is updated. Also,
I'm sure there are others that use the -s in the Sun version.
Here, then, is the patch against tftpd 8.1 in netstd 3.07:
--- tftpd.c.origThu Jun 25 17:26:53 1998
+++ tftpd.c Thu Jun 25 18:29:55 1998
@@ -47,6 +47,11 @@
* This version includes many modifications by Jim Guyton
* [EMAIL PROTECTED].
*/
+
+/*
+ * -s support added by John Goerzen [EMAIL PROTECTED]
+ *
+ */
#include sys/param.h
#include sys/ioctl.h
@@ -117,9 +122,10 @@
register int n;
int ch, on;
struct sockaddr_in sin;
+ int usingchroot = 0, chrootdirfound = 0, res;
openlog(tftpd, LOG_PID, LOG_DAEMON);
- while ((ch = getopt(argc, argv, ln)) != EOF) {
+ while ((ch = getopt(argc, argv, lns)) != EOF) {
switch (ch) {
case 'l':
logging = 1;
@@ -127,6 +133,9 @@
case 'n':
suppress_naks = 1;
break;
+ case 's':
+ usingchroot = 1;
+ break;
default:
syslog(LOG_WARNING, ignoring unknown option -%c, ch);
}
@@ -138,11 +147,36 @@
for (dirp = dirs; optind argc dirp dirs[MAXDIRS];
optind++) {
if (argv[optind][0] == '/') {
- dirp-name = argv[optind];
- dirp-len = strlen(dirp-name);
-
X is not working after the upgrade
Hi! I've just upgraded to the latest X from hamm (3.3.2.2-1, before I had 3.3.2.1-1) and now I can't execute startx anymore. An the end of usual messages it produces: _X11TransSocketUNIXConnect: Can't connect: errno = 111 giving up. xinit: Connection refused (errno 111): unable to connect to X server xinit: No such process (errno 3): Server error. Suspicious lines in the output of strace looks like: [pid 264] execve(/usr/bin/X11/X, [X, :0, -auth, /home/shurik/.Xauthority], [/* 35 vars */]) = -1 EPERM (Operation not permitted) [pid 264] access(X, R_OK) = -1 ENOENT (No such file or directory) [pid 264] write(2, xinit: , 8xinit: ) = 8 [pid 264] write(2, No such file or directory (errno..., 38No such file or directory (errno 2): ) = 38 [pid 264] write(2, no server \X\ in PATH\n, 22no server X in PATH ) = 22 X is certainly in place: -rwsr-xr-x 1 root root 4880 Jun 23 23:46 /usr/bin/X11/X Is it something known? The upgrade was without any visible problems (dselect + apt). I don't want to fill a bug report just after the official Beta release, so I decided to ask here first. Thanks in advance for any help! --- Shurik. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X is not working after the upgrade
It worked for me but. Check you /etc/X11/Xservers file and make sure that it has an Xserver listed such as /usr/X11R6/bin/XF86_SVGA. I had to do this when I first installed hamm because the configuration left it out. Alexander Shumakovitch wrote: Hi! I've just upgraded to the latest X from hamm (3.3.2.2-1, before I had 3.3.2.1-1) and now I can't execute startx anymore. An the end of usual messages it produces: _X11TransSocketUNIXConnect: Can't connect: errno = 111 giving up. xinit: Connection refused (errno 111): unable to connect to X server xinit: No such process (errno 3): Server error. Suspicious lines in the output of strace looks like: [pid 264] execve(/usr/bin/X11/X, [X, :0, -auth, /home/shurik/.Xauthority], [/* 35 vars */]) = -1 EPERM (Operation not permitted) [pid 264] access(X, R_OK) = -1 ENOENT (No such file or directory) [pid 264] write(2, xinit: , 8xinit: ) = 8 [pid 264] write(2, No such file or directory (errno..., 38No such file or directory (errno 2): ) = 38 [pid 264] write(2, no server \X\ in PATH\n, 22no server X in PATH ) = 22 X is certainly in place: -rwsr-xr-x 1 root root 4880 Jun 23 23:46 /usr/bin/X11/X Is it something known? The upgrade was without any visible problems (dselect + apt). I don't want to fill a bug report just after the official Beta release, so I decided to ask here first. Thanks in advance for any help! --- Shurik. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X is not working after the upgrade
On Thu, Jun 25, 1998 at 08:33:29PM -0400, Shaleh wrote: It worked for me but. Check you /etc/X11/Xservers file and make sure that it has an Xserver listed such as /usr/X11R6/bin/XF86_SVGA. I had to do this when I first installed hamm because the configuration left it out. No problems: 11 cat /etc/X11/Xserver /usr/bin/X11/XF86_SVGA Console The first line in this file is the full pathname of the default X server. The second line shows who is allowed to run the X server: RootOnly Console (anyone whose controlling tty is on the console) Anybody /usr/bin/X11/XF86_SVGA is also in place: -rwxr-xr-x 1 root root 2883228 Jun 23 23:45 /usr/bin/X11/XF86_SVGA Looks just like a miracle! Probably, I'll reinstall it tomorrow. Thanks! --- Shurik. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X is not working after the upgrade
On Fri, Jun 26, 1998 at 02:18:34AM +0200, Alexander Shumakovitch wrote: I've just upgraded to the latest X from hamm (3.3.2.2-1, before I had 3.3.2.1-1) and now I can't execute startx anymore. An the end of usual messages it produces: _X11TransSocketUNIXConnect: Can't connect: errno = 111 giving up. xinit: Connection refused (errno 111): unable to connect to X server xinit: No such process (errno 3): Server error. Please be sure you're not using the test build of 3.3.2.2-1. It had a bug in it which GUARANTEES that error if startx or xinit is used. The correct size and md5sum of the xbase 3.3.2.2-1 .deb are: 2167928 cea2e2b7047fb37bcbc51342bb714b29 -- G. Branden Robinson | Purdue University | The noble soul has reverence for itself. [EMAIL PROTECTED] | -- Friedrich Nietzsche http://www.ecn.purdue.edu/~branden/ | pgpHxItcmp9dj.pgp Description: PGP signature
Re: X is not working after the upgrade
On Thu, Jun 25, 1998 at 07:45:39PM -0500, Branden Robinson wrote: Please be sure you're not using the test build of 3.3.2.2-1. It had a bug in it which GUARANTEES that error if startx or xinit is used. The correct size and md5sum of the xbase 3.3.2.2-1 .deb are: 2167928 cea2e2b7047fb37bcbc51342bb714b29 Unfortunately, I can't do it right now. I'm at home now with 56Kbps, but I'll reinstall xbase tomorrow. Nonetheless, I've installed just a version from ftp.de.debian.org. The Packages file looks OK. Could you tell the correct size of xinit, or the other file which could be responsible for this problem, so that I can verify it. Thanks a lot! --- Shurik. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: X is not working after the upgrade
On Fri, Jun 26, 1998 at 03:11:31AM +0200, Alexander Shumakovitch wrote: Unfortunately, I can't do it right now. I'm at home now with 56Kbps, but I'll reinstall xbase tomorrow. Nonetheless, I've installed just a version from ftp.de.debian.org. The Packages file looks OK. Could you tell the correct size of xinit, or the other file which could be responsible for this problem, so that I can verify it. If /usr/X11R6/lib/X11/xinit/xserverrc exists, delete it. -- G. Branden Robinson | Purdue University | kernel panic -- causal failure [EMAIL PROTECTED] | universe will now reboot http://www.ecn.purdue.edu/~branden/ | pgp8POYZ5wZGC.pgp Description: PGP signature
Re: in.tftpd: Missing a critical feature
On Thu, Jun 25, 1998 at 06:38:56PM -0500, John Goerzen wrote: Package: netstd Version: 3.07-1 The bug number for this report hasn't shown up in the web pages yet, so I'm removing bugs.debian.org from the cc: list in this response. You might want to forward it into the bug report. Sun's tftpd has an option, -s. Basically, this option chroot's to the specified directory, than set[ug]id's itself to nobody. After that, it starts serving files. [...] Why is this important? Well, we have a number of NCD X terminals. These terminals try to get files with absolute pathnames like /usr/lib/X11/ncd/configs. The tftp server should really serve this file from an appropriate subdirectory below the tftpboot directory -- NOT the system's /usr/lib/X11 directory. That is, the server should actually hand the client the file /usr/local/ncdroot/usr/lib/X11/ncd/configs in our case. I don't like Sun's chroot mode -- you can't make symlinks out of your /tftpboot into the rest of the filesystem. I did this a lot at work last term, where I would build the software for one of our embedded systems in my home directory, press reset, and it would boot the new image automatically. Instead, could we implement behaviour like this: - discard pathnames with dot-dots, of course. - if no path restriction is specified on the command line, use the given path directly. - if pathname starts with /tftpboot (or whatever's specified on the command line to limit tftpd) pass it through directly. - if pathname starts with / but not /tftpboot, append it to /tftpboot. - otherwise, append it to /tftpboot/ (ie. relative to tftpboot directory). This should cover all usages for tftp that I know of. Have fun, Avery -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bug #23877: Include autoup.sh and apt in hamm/hamm
On 25 Jun 1998, Ben Gertzfield wrote: Subject: please include apt and autoup in hamm/hamm/upgrade-i386/ To: [EMAIL PROTECTED] -- Package: ftp.debian.org,apt,autoup Version: N/A Well, let's just do it! I see no problem with making such a directory for final hamm. imo this directory should also be populated with symlinks to all the packages which autoup.sh needs to do the upgrade. there should also be a copy of the the bo libc5 dpkg in there, to assist with upgrades from rex/buzz. craig -- craig sanders -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
My BitFontEdit as a Debian Package
I'm the author of the BitFontEdit package which I wrote 10 years ago. Only the original version with no name (directory is TermFonts) is still on the net in a few obscure locations. I want to get the improved version back on the net (my free website closed) and thought that one way to do this is to make it into a Debian package. BitFontEdit allows one to create fonts from scratch for dumb terminals. It is really just some filters. One creates a text file (in the format I specify) using any editor and draws the characters they want using the * character for each pixel. Then you run this file thru my filter and get the soft-font which may be downloaded to a terminal. It supports both the Wyse and VT (Dec) type of softfont. I started to modify it so it would also do X-Windows font, but I think that there is already such a program. There is another filter to convert soft-font into the * type drawings. Using this one may edit existing soft-font. At one time I was hoping to extend it a cover some dot-matrix printers. As a Debian package, I'm also hoping that someone will find it and perhaps develop it further. What do you think? Since I wrote the code and some documentation on how to use it, it shouldn't be too hard to package it for Debian (or will it be?). I'm also the author of Text-Terminal-HOWTO. A web-site specializing in terminals was going to make BitFontEdit available, but so far nothing has happened. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: My BitFontEdit as a Debian Package
Since I wrote the code and some documentation on how to use it, it shouldn't be too hard to package it for Debian (or will it be?). Should easy. The main bit of debianising a package, is encapsulated in a makefile (debian/rules), which normally invokes the make clean, make all, and make install steps of your makefile (assuming you have such a thing), and tops and tails the process with the stuff to make the package file. You'll probably need to have a debian system to do it on though. What licence do you distribute this under ? Cheers, Phil. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New web pages are finally up!
In article [EMAIL PROTECTED] you wrote: : The latest version of the Debian web pages are complete. I like them. Good work! A few nits. The link on the developer's corner page that I think should go to the list of developers instead brings up the xearth image. The visit the sponsor image on the home page just brings the image up in an image browser, it does not actually take me anywhere. I didn't have long to spend, will let you know if I run into anything else as I troll later tonight. Bdale -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: netselect - choosing the best FTP server automatically
On Thu, 25 Jun 1998, Avery Pennarun wrote: A while ago on debian-devel I proposed an algorithm that would allow APT to choose the best possible server for each user from a large list automatically. It could also be used for other tasks, eg. choosing a good SQUID neighbour or IRC server. This is pretty good, but it seems to loose much meaning for me with several equal servers, ftp1.us.debian.org 111 ms 16 hops 91% ok (31/34) llug.sep.bnl.gov83 ms 15 hops 100% ok (110/110) ftp.debian.org 80 ms 19 hops 93% ok (40/43) ftp.cdrom.com 87 ms 13 hops 92% ok (39/42) -- ftp1.us.debian.org 115 ms 16 hops 93% ok (31/33) llug.sep.bnl.gov79 ms 15 hops 85% ok (18/21) ftp.debian.org 100 ms 19 hops 90% ok (28/31) ftp.cdrom.com 105 ms 13 hops 90% ok (27/30) I'm sitting on a high speed lan conection (ie I get ~100k/s from llug) Presumably this program pushes the network a bit hard and that is why there is packet loss a straight ping to any of these sites will get 0% loss. My question is how can we build a single weighted score for each site? Let the user pick one of the top two or so. But how do you weight? Someone should definately work on this - I can make use if in APT in a big way. Jason -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Hamm Bug Stamp-Out List for June 25, 1998
In article [EMAIL PROTECTED] you wrote: : 21884 libc6-dev: relative links between top-level dirs : The upstream maintainer (Ulrich D.) insists that the relative links are : correct and that making /usr a symlink to something else is evil. I'm not sure I completely understand what the links in libc6-dev look like, and I don't have time to go look right now. A few quick words about why relative links are important in some circumstances seems in order, though. If you're working in a multi-machine networked environment, and are accessing another system using NFS (particularly with an automounted host map), an absolute link can easily violate the principle of least astonishment... if you are tracking down a remote machine's filesystems, and cross an absolute symbolic link, you're back on your local system's filesystem! That's almost never what you meant, or what you wanted. For this and related reasons, we have an almost absolute prohibition against absolute symlinks in my shop, except in cases where it really doesn't matter. For example, we have lots of symlinks that actually point off to subtrees on a mount from a central NFS server... since these all point to the same actual filesystem regardless of which machine you're on, it's ok for them to be absolute. In summary, while I don't know the details in this particular case, relative links are more often a good thing than a bad thing... Bdale -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: netselect - choosing the best FTP server automatically
On Thu, Jun 25, 1998 at 09:18:24PM -0600, Jason Gunthorpe wrote: On Thu, 25 Jun 1998, Avery Pennarun wrote: [netselect http://www.worldvisions.ca/~apenwarr/netselect-0.1.tar.gz] This is pretty good, but it seems to loose much meaning for me with several equal servers, ftp1.us.debian.org 111 ms 16 hops 91% ok (31/34) llug.sep.bnl.gov83 ms 15 hops 100% ok (110/110) ftp.debian.org 80 ms 19 hops 93% ok (40/43) ftp.cdrom.com 87 ms 13 hops 92% ok (39/42) -- ftp1.us.debian.org 115 ms 16 hops 93% ok (31/33) llug.sep.bnl.gov79 ms 15 hops 85% ok (18/21) ftp.debian.org 100 ms 19 hops 90% ok (28/31) ftp.cdrom.com 105 ms 13 hops 90% ok (27/30) I'm sitting on a high speed lan conection (ie I get ~100k/s from llug) Presumably this program pushes the network a bit hard and that is why there is packet loss a straight ping to any of these sites will get 0% loss. There are two big problems I've noticed with the first version (I wrote it all last night and have no more time to work on it for at least three weeks; sorry): - it stops as soon as the hop count of all sites is determined, AND the minimum time has elapsed. (The default minimum isn't very strict.) Requests that it sends out might not be back before it gives up! - because of random delays (and because if the TTL attempt is too low, it isn't valid for a timing check) the number of packets sent to each host varies wildly. Too-low packet counts can make this really stand out, since getting 4/5 packets is much worse than 29/30, even though only one packet was lost. I don't consider the ping time variance a big deal -- because it sends out lots of packets at once, they will get queued up a bit, but (theoretically, anyway) since each host is equally affected this effect should mostly average out. I'm not surprised by +/- 50 ms, though. If the two bugs above were fixed, I think this would be usable. After all, comparing equal hosts isn't all that useful anyway. The ones we want to filter out are the ones with really large lag times ( 500 ms) and large packet losses ( 10%). Bandwidth comparison a la bing would also be nice, but I definitely don't have time to implement that anytime soon. My question is how can we build a single weighted score for each site? Let the user pick one of the top two or so. But how do you weight? Assuming that the packet loss and timing statistics were made more reliable... I would say that packet loss and delay are of almost equal importance, and hop count considerably less. With bugfixes we should be able to attribute about +/- 5% of packet loss and 50ms of delay to random chance, so we need a nice binary comparison routine. Here's a random one: - gravitate, say, 3% packet loss to each site, toward the average. ie. if A has 85% connectivity and B has 95% connectivity, adjust to 88% and 92%. If the average is less than 3% away, just set both to the average. - similarly with 25ms of lag time. Given 200ms and 300ms, we get 225ms and 275ms. - Take percentage difference for each and subtract them (note that smaller lag is better, and larger connectivity is better). Hmm... - A -- B - (88-90)/90 = -2.22% (92-90)/90 = +2.22% (225-250)/250 = -10%(275-250)/250 = +10% Subtract: 7.78% Subtract: -7.78% Note that due to the way percentage difference works, we only actually need one of these. - Higher score wins. Disclaimer: IANAS. (I Am Not A Statistician :)) Anyone is invited to work on the this since I have very little time to do so. Please forward me any improvements or ideas. Have fun, Avery -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: New web pages are finally up!
Bdale wrote: The link on the developer's corner page that I think should go to the list of developers instead brings up the xearth image. The visit the sponsor image on the home page just brings the image up in an image browser, it does not actually take me anywhere. Good catch. This shows really have to be careful when using content negotiation. I've fixed both of these and they will be at your favorite mirror within 12 hours or so. Jay Treacy -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Away from my computer for a while
I will be out of town for a quite long period (until August 14). Feel free to do NMUs of any of my packages, if it is needed. -- Rafael Laboissiere Institut de la Communication Parlee | Email: [EMAIL PROTECTED] UPRESS A CNRS 5009 / INPG | Voice: +33 4.76.57.48.49 46, av. Felix Viallet | Fax: +33 4.76.57.47.10 F-38031 Grenoble CEDEX 1 France | URL: http://www.icp.inpg.fr/~rafael -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Hamm Bug Stamp-Out List for June 25, 1998
Hi, Bdale == Bdale Garbee [EMAIL PROTECTED] writes: Bdale If you're working in a multi-machine networked environment, Bdale and are accessing another system using NFS (particularly with Bdale an automounted host map), an absolute link can easily violate Bdale the principle of least astonishment... if you are tracking Bdale down a remote machine's filesystems, and cross an absolute Bdale symbolic link, you're back on your local system's filesystem! Bdale That's almost never what you meant, or what you wanted. On the other hand, if you are using automount, or AFS, or DFS, or even, in some environments, NFS, you may keep all mounts under, say, /mnt (eg /usr - /mnt/i386/usr), in which case, any relative links between top level directories loose. In this case, a relative symlink also violates the princple of least astonishment, and indeed, does not even work. I think, on balance, the policy works for the most common case. manoj 3.3.5. Symbolic links - In general, symbolic links within a toplevel directory should be relative, and symbolic links pointing from one toplevel directory into another should be absolute. (A toplevel directory is a sub-directory of the root directory `/'.) In addition, symbolic links should be specified as short as possible, i.e., link targets like `foo/../bar' are depreciated. -- One machine can do the work of fifty ordinary men. No machine can do the work of one extraordinary man. Elbert Hubbard ...yet. Karl Lehenbauer Manoj Srivastava [EMAIL PROTECTED] http://www.datasync.com/%7Esrivasta/ Key C7261095 fingerprint = CB D9 F4 12 68 07 E4 05 CC 2D 27 12 1D F5 E8 6E -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Hamm Bug Stamp-Out List for June 25, 1998
The upstream maintainer (Ulrich D.) insists that the relative links are correct and that making /usr a symlink to something else is evil. but in some situations they are necessary : when useing nfsroot with special setup. for improved speed i want a codafs as the basic layer, so /usr - /coda/usr, /opt-/coda/opt, ... but anyone doing such stuff should be able to fix symlinks himself, so this isn't a big problem. andreas -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Intend to package Snd and CLM
Hi ! I'm planning to package Snd (a powerful Sound Editor) and the Common Lisp Music system for slink. Any suggestions, comments ?? Guenter (Please CC public replies to me .. I'm not on the list) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
tcpd with xinetd
Hello, a few years ago i patches tcpd to cooperate with xinetd; the patch is on ftp://ftp.tuke.sk/pub/unix/security/tcpd_xinetd.patch The problem is xinetd doesn't allow to define argv[0] thus tcpd would call itself; this patch increases argv if argv[0] eauqls to tcpd, in.tcpd or in.frontd (defined in Makefile). This allows to join advantages of xinetd and tcpd; better checks etc. I'd like to make patch to xinetd which would join it more close - xinetd could call hosts_access() from libwrap; I've heard some FreeBSD developers are doing that. Could anyone comment this ? OR probably make new xinetd package with this patch included ? -- Matus fantomas Uhlar, sysadmin at NETLAB+ Kosice, Slovakia BIC coord for *.sk; admin of netlab.irc.sk; co-admin of irc.felk.cvut.cz -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Intent to package xruskb and netenv
Brian Mays writes: Package: netenv Version: 0.81-1 Description: Configure your system for different network environments. Netenv creates a file containing variable assignments which reflect the current environment. It is especially useful for laptop computers, since it can be used by the PCMCIA setup scheme (like the one included in the Debian pcmcia-cs package) and the plip setup script included as an example in this package. You can also use netenv configure your windowmanager or your printing environment. Exactly what I need right now. When will it be available? Or else where do I get the source? Also I'm willing to beta test it. Michael -- Dr. Michael Meskes, Project-Manager| topsystem Systemhaus GmbH [EMAIL PROTECTED]| Europark A2, Adenauerstr. 20 [EMAIL PROTECTED] | 52146 Wuerselen Go SF49ers! Go Rhein Fire! | Tel: (+49) 2405/4670-44 Use Debian GNU/Linux! | Fax: (+49) 2405/4670-10 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Oops
Sorry for the last one. Should have gone to request. But I'm in a hurry. My boss decided to strip my remote login rights without telling me. Boy, am I glad I leave this company! Michael -- Dr. Michael Meskes, Project-Manager| topsystem Systemhaus GmbH [EMAIL PROTECTED]| Europark A2, Adenauerstr. 20 [EMAIL PROTECTED] | 52146 Wuerselen Go SF49ers! Go Rhein Fire! | Tel: (+49) 2405/4670-44 Use Debian GNU/Linux! | Fax: (+49) 2405/4670-10 -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Hamm Bug Stamp-Out List for June 25, 1998
On 26 Jun 1998, Manoj Srivastava wrote: Hi, Bdale == Bdale Garbee [EMAIL PROTECTED] writes: Bdale If you're working in a multi-machine networked environment, Bdale and are accessing another system using NFS (particularly with Bdale an automounted host map), an absolute link can easily violate Bdale the principle of least astonishment... if you are tracking Bdale down a remote machine's filesystems, and cross an absolute Bdale symbolic link, you're back on your local system's filesystem! Bdale That's almost never what you meant, or what you wanted. On the other hand, if you are using automount, or AFS, or DFS, or even, in some environments, NFS, you may keep all mounts under, say, /mnt (eg /usr - /mnt/i386/usr), in which case, any relative links between top level directories loose. In this case, a relative symlink also violates the princple of least astonishment, and indeed, does not even work. I think, on balance, the policy works for the most common case. manoj Thanks for quoting Policy on this. Upon reading the actual text, I don't think this policy applies to libc6-dev. 3.3.5. Symbolic links - In general, symbolic links within a toplevel directory should be relative, and symbolic links pointing from one toplevel directory into another should be absolute. (A toplevel directory is a sub-directory of the root directory `/'.) Note: from one toplevel directory into another The links are from /usr/lib, which is a subdirectory of a top level directory. Yes, it points to a top level directory, but that only satifies one half of the above criterion. In addition, symbolic links should be specified as short as possible, i.e., link targets like `foo/../bar' are depreciated. I guess I'm just stupid ;-) but I don't understand what the example is trying to tell me. Waiting is, Dwarf -- _-_-_-_-_- Author of The Debian Linux User's Guide _-_-_-_-_-_- aka Dale Scheetz Phone: 1 (850) 656-9769 Flexible Software 11000 McCrackin Road e-mail: [EMAIL PROTECTED] Tallahassee, FL 32308 _-_-_-_-_-_- If you don't see what you want, just ask _-_-_-_-_-_-_- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Possible bug in Apache 1.3.0...
[Also sent to debian-devel, since I'm interested to hear about other people having noticed this problem] I run a medium sized webserver (~80.000 hits/day) with apache/1.3.0, however it occasionnaly stops answering http queries and forks 120 http daemons, while I've set the max. forking limit to 20. When this happens, the only way to get apache to serve http requests again is too reload it. After much hair pulling and logfile browsing, it appears that the problem always shown up while someone was browsing pages with either Net_Vampire/2.4 or Net_Vampire/2.41. This morning, apache stopped answering queries again, and even after retarting it, requests would stall after 3 minutes; I saw in the access_log that s/o was currently using Net_Vampire on my site. Apache started to work fine again only after I used a firewall rule to deny any tcp access from the IP address of this Net Vampire user. Is it a known bug? Is there a better work-around than mine? Feedback/Ideas welcomed... Cordialement, -- - Vincent RENARDIAS [EMAIL PROTECTED],pipo.com,debian.org} - - Debian/GNU Linux: Pipo:WAW: - - http://www.fr.debian.orghttp://www.pipo.com http://www.waw.com - --- - La fonctionnalite Son Visuel vous delivre des avertissements visuels. - - [Message durant l'installation de Windows95]:wq -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: My BitFontEdit as a Debian Package
David Lawyer [EMAIL PROTECTED] wrote: I'm the author of the BitFontEdit package which I wrote 10 years ago. Only the original version with no name (directory is TermFonts) is still on the net in a few obscure locations. I want to get the improved version back on the net (my free website closed) and thought that one way to do this is to make it into a Debian package. Sounds like a good idea. We're not yet to the point where it's completely trivial to throw together a debian package. There's a lot of possiblities which need to be allowed for. And, of course, we want everything to work together reasonably well. Skim when encounter stuff that's not applicable to you (shared libraries, emacs configuration, ...). In theory, you should be able to run through all of this in a few hours (especially for a simple package like what you're talking about). Anyways, here's some meta-instructions: [Joining debian] see the debian developer's reference: http://www.debian.org/doc/packaging-manuals/developers-reference/ [Designing the package] see the debian policy manual http://www.debian.org/doc/debian-policy/ [Implementing the package] this documentation needs to be pieced together better, but there's more than one way to approach this: debian packaging manual http://www.debian.org/doc/packaging-manuals/packaging.html New-Maintainer's mini-howto, deb-make http://va.debian.org/~jaldhar/index.html http://www.cis.udel.edu/~lowe/deb-make [Testing the package] You should have lintian check over your package(s), http://www.debian.org/lintian Once things seem clean enough, you should also install the package on a debian system and make sure it looks and acts right. [Uploading the package] see above: the bit about joining debian. BitFontEdit allows one to create fonts from scratch for dumb terminals. It is really just some filters. One creates a text file (in the format I specify) using any editor and draws the characters they want using the * character for each pixel. Then you run this file thru my filter and get the soft-font which may be downloaded to a terminal. It supports both the Wyse and VT (Dec) type of softfont. I started to modify it so it would also do X-Windows font, but I think that there is already such a program. I think it would be nice to support X-windows fonts (bdf, preferably, I think). It would also be nice to write out console fonts. But that's up to you. -- Raul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
