Re: fcopy exits with an error if file is not updated because its up2date
> Hi, > > fcopy exits with an error if a file is not updated because its up2date, I > believe that is a bug. Because everything _is_ ok. > > So I used to always do "fcopy -i foo", which today I have changed to "fcopy > foo || true" - which at least will tell me in the logs that something is > wrong. But I would prefer not to get an exitcode!=0 from fcopy in the first > place. > > Opinions? > At least I'd also consider it a somewhat strange behaviour; but I can't say for sure whether this is a bug or a feature; Henning or Thomas definitively had a reason for implementing it that way. However, I'd argue for some further command line parameter (I know, fcopy already has quite a lot...) or some environment variable to make fcopy not exit with an error in that special case. Best, Michael pgpfpPI1XdzV4.pgp Description: PGP signature
fcopy exits with an error if file is not updated because its up2date
Hi, fcopy exits with an error if a file is not updated because its up2date, I believe that is a bug. Because everything _is_ ok. So I used to always do "fcopy -i foo", which today I have changed to "fcopy foo || true" - which at least will tell me in the logs that something is wrong. But I would prefer not to get an exitcode!=0 from fcopy in the first place. Opinions? BTW: the fcopy manpage says "OPTIONS Options -c, -s, -t (or their analog variables) and a SOURCE are required." This is not true anymore. regards, Holger P.S.: resent because I have not received it on the list. pgp6oR0yM4X4D.pgp Description: PGP signature
Re: fcopy exits with an error if file is not updated because its up2date
* Holger Levsen <[EMAIL PROTECTED]> [070411 17:07]: > Hi, > > fcopy exits with an error if a file is not updated because its up2date, I > believe that is a bug. Because everything _is_ ok. > > So I used to always do "fcopy -i foo", which today I have changed to "fcopy > foo || true" - which at least will tell me in the logs that something is > wrong. But I would prefer not to get an exitcode!=0 from fcopy in the first > place. > > Opinions? > Hi, yes, I agree! I'd prefer that behaviour, too. I also think that the present behaviour is a bug... regards, Oliver -- * Oliver Osburg- Software and System Architect * * email: [EMAIL PROTECTED] phone: +49 179 50 76 784 * * note: All above is my opinion unless said otherwise * * One Unix to rule them all, One Resolver to find them, * * One IP to bring them all and in the zone bind them.*
fcopy exits with an error if file is not updated because its up2date
Hi, fcopy exits with an error if a file is not updated because its up2date, I believe that is a bug. Because everything _is_ ok. So I used to always do "fcopy -i foo", which today I have changed to "fcopy foo || true" - which at least will tell me in the logs that something is wrong. But I would prefer not to get an exitcode!=0 from fcopy in the first place. Opinions? BTW: the fcopy manpage says "OPTIONS Options -c, -s, -t (or their analog variables) and a SOURCE are required." This is not true anymore. regards, Holger pgpzVocnYjqeo.pgp Description: PGP signature
Extending fcopy
Hi all, Janning has proposed quite a few enhancements to fcopy lately. I think the resulting discussion should be bundled in a single thread and it should actually move to linux-fai-devel (@Janning: don't know, whether you are subscribed to this one already). Just to summarize, the suggestions were: - some -k/--keep option (which I'd rather call --keep-permissions :-) ) to retain the permissions of the file which should be overwritten (BTW: what should happen if there is no such file?) - some way to return the number of files that were actually copied. My objections were - Adding code to a mission-critical tool like fcopy is dangerous. - More options may confuse users. - The intended behaviour may be simulated using pre-/postinst scripts. In my opinion the first and the second are still valid, however, the third has been proven to be _really_ clumsy. So I guess these options should be added, if a proper patch can be provided. Hoping for further discussion, Michael pgpldixqDwsaE.pgp Description: PGP signature
Re: fcopy feature request: report if a file was copied
[...] > > yes but imagine > fcopy -ir /etc/postfix/ > and i need a postinst script with >/etc/init.d/postfix reload > > i have to place postinst inside > /etc/postfix/main.cf/BASE/postinst > /etc/postfix/master.cf/BASE/postinst > right? > Nearly correct - there is no such directory as BASE, but rather you'll have the files /etc/postfix/main.cf/BASE and /etc/postfix/main.cf/postinst > and if i did change master.conf and main.conf postinst would be called two > times. and what if i have 30 files or more in /etc/postfix like virtual > tables ... > Would you really change all of them at once? And furthermore, in case of postfix there is usually no need for reloading it, it will at most be a call to postmap (which should definitely go into the postinst script). [...] Even though I do see (very few) applications for such a code, the same objections as for your -k,--keep-permissions apply here: It might add quit a bit of code, which may cause quite a lot of trouble. Don't get me wrong, I'd hope for a lot more of such ideas, but they must be discussed thoroughly, even more so when touching mission-critical tools such as fcopy. Best, Michael pgphlfbnvz0lR.pgp Description: PGP signature
Re: another fcopy feature request -k --keep-modes
> > sometimes i don't care about the owner:group:modes:timestamp > example /etc/ssh/sshd_config. I just want to copy my content > to /etc/ssh/sshd_config, but owner, group and modes should be kept. > > Of course i can add -m but then i need to know the correct permissions and > ownership of each file i want to change. > > I just want to copy a new file and KEEP the original file modes and > ownerships. with fcopy you have -m -M or modes from source file. > > but It would be nice to have > > -k --keep-modes[=owner,group,modes,time] > to keep all or named file stats from the original file > > i can write a patch if it sounds reasonable. > Sorry for leaving this one open for so long, but I'd like to add a few comments here: - Of course, this is indeed something useful and a fully acknowledge your reasoning. - BUT: I'm fearing a real overload of fcopy with more and more options for each and every need. This will likely irritate users. - Adding more and more code to fcopy makes it prone to even more bugs and harder to fix any of them. - I guess one can implement your idea in a simple preinst script, which will work as soon as your permission-patch gets included: preinst: --- #!/bin/sh chmod --reference=$FAI_ROOT/etc/ssh/sshd_config $2 --- Surely, this is a bit clumsy and not as flexible as =owner,group.modes,time, but - it is simple - won't confuse anybody HTH, Michael pgpgPTL9xcjDY.pgp Description: PGP signature
Re: fcopy feature request: report if a file was copied
Am Mittwoch, 6. Dezember 2006 17:01 schrieben Sie: > > Am Mittwoch, 6. Dezember 2006 16:13 schrieb Michael Tautschnig: > > > > hi, > > > > > > > > as i use softupdate a lot , it would be nice to have a fcopy feature > > > > which reports if any files or file permissions were changed. Look at > > > > my script in BASE/50-ntp > > > > > > [...] > > > > > > I guess postinst is what you really want (and it is already there, > > > after all). See the man page. > > > > you are right, but > > > > - postinst is called for every file which is copied > > - > > Which is what you intended, as I understood it - this does _not_ mean that > postinst is called for every file that fcopy is run on, but only in those > cases where the file is _really_ copied, i.e., overwritten. yes but imagine fcopy -ir /etc/postfix/ and i need a postinst script with /etc/init.d/postfix reload i have to place postinst inside /etc/postfix/main.cf/BASE/postinst /etc/postfix/master.cf/BASE/postinst right? and if i did change master.conf and main.conf postinst would be called two times. and what if i have 30 files or more in /etc/postfix like virtual tables ... with my approach it would be `fcopy -irR /etc/postfix/` -gt "0" && /etc/init.d/postfix reload (maybe this is not correct bash syntax, just example code ) kind regards Janning
Re: fcopy feature request: report if a file was copied
> Am Mittwoch, 6. Dezember 2006 16:13 schrieb Michael Tautschnig: > > > hi, > > > > > > as i use softupdate a lot , it would be nice to have a fcopy feature > > > which reports if any files or file permissions were changed. Look at my > > > script in BASE/50-ntp > > > > [...] > > > > I guess postinst is what you really want (and it is already there, after > > all). See the man page. > > you are right, but > > - postinst is called for every file which is copied > - > Which is what you intended, as I understood it - this does _not_ mean that postinst is called for every file that fcopy is run on, but only in those cases where the file is _really_ copied, i.e., overwritten. Best, Michael pgpMQXkhLyFLJ.pgp Description: PGP signature
another fcopy feature request -k --keep-modes
sometimes i don't care about the owner:group:modes:timestamp example /etc/ssh/sshd_config. I just want to copy my content to /etc/ssh/sshd_config, but owner, group and modes should be kept. Of course i can add -m but then i need to know the correct permissions and ownership of each file i want to change. I just want to copy a new file and KEEP the original file modes and ownerships. with fcopy you have -m -M or modes from source file. but It would be nice to have -k --keep-modes[=owner,group,modes,time] to keep all or named file stats from the original file i can write a patch if it sounds reasonable. kind regards Janning
Re: fcopy feature request: report if a file was copied
Am Mittwoch, 6. Dezember 2006 16:46 schrieb Janning Vygen: > Am Mittwoch, 6. Dezember 2006 16:13 schrieb Michael Tautschnig: > > > hi, > > > > > > as i use softupdate a lot , it would be nice to have a fcopy feature > > > which reports if any files or file permissions were changed. Look at my > > > script in BASE/50-ntp > > > > [...] > > > > I guess postinst is what you really want (and it is already there, after > > all). See the man page. > > you are right, but > > - postinst is called for every file which is copied sorry, i clicked return-enter to early. my second reason why i don't like postinst: - its more transparent to have it inside the script which fcopies the files. kind regards janning
Re: fcopy feature request: report if a file was copied
Am Mittwoch, 6. Dezember 2006 16:13 schrieb Michael Tautschnig: > > hi, > > > > as i use softupdate a lot , it would be nice to have a fcopy feature > > which reports if any files or file permissions were changed. Look at my > > script in BASE/50-ntp > > [...] > > I guess postinst is what you really want (and it is already there, after > all). See the man page. you are right, but - postinst is called for every file which is copied - > However, the permission thing is obviously not captured (yet). > > Best, > Michael -- PLANWERK 6 websolutions Venloer Straße 8, 40477 Düsseldorf Tel: (0211) 302666-0 Fax: (0211) 302666-10 http://www.planwerk6.de/
Re: fcopy feature request: report if a file was copied
> hi, > > as i use softupdate a lot , it would be nice to have a fcopy feature which > reports if any files or file permissions were changed. Look at my script in > BASE/50-ntp > [...] I guess postinst is what you really want (and it is already there, after all). See the man page. However, the permission thing is obviously not captured (yet). Best, Michael pgp5yRmGlAlET.pgp Description: PGP signature
Re: fcopy feature request: report if a file was copied
Am Mittwoch, 6. Dezember 2006 15:35 schrieb Thomas Lange: > >>>>> On Wed, 6 Dec 2006 15:23:58 +0100, Janning Vygen <[EMAIL PROTECTED]> > >>>>> said: > >>>>> > > if [ `fcopy -iM /etc/ntp.conf` -gt "0" ] > > > > it would be nice to have an option like > > -n --number report number of files changed > > and let fcopy report the number of files changed > > Good idea. thanks. > But then fcopy must not write any error or verbose messages > when using -n. Or should -n report the number of files changes using > the exit/return code? we have two options: - using error code: i thought about using return code first, but isn't it a little bit non-unix-stlye to use exit code for reporting? - using stdout: we could print verbose messages to stderr with "warn" and use stdout only for reporting number of files. i think using stdout is best, just reporting a number. The script could even build a nice report with "-R full" to send to administrator showing diffs and chmod/chown/utime if you get output with -R something was changed, so you can always reload the daemon. so my proposal would be: 1. send all verbose output to stderr 2. create a new option --report [number|filenames|full|mail] "number" (default if -R is given): reports number of file changes to stdout "filenames" list all files which has changed one per line "full" list all files and their differences of content or modes "mail" sends full report to stdout and [EMAIL PROTECTED] But i am unsure how to name the option as "-r" is already used, as -l --list and -c --count are used, too. kind regards Janning
Re: Fcopy patch regarding debian bug report #360184
On Wed, Dec 06, 2006 at 12:01:03PM +0100, Janning Vygen wrote: > + return 1 if ( $suid != $duid or $sgid != $dgid or $smode != $dmode or > $stime != $dtime ); thinking about (pre/post)inst scripts, comparing the time will lead to problems... (as destfile may have a different mtime because it has been modified by one of these scripts). -- c u henning
Re: fcopy feature request: report if a file was copied
>>>>> On Wed, 6 Dec 2006 15:23:58 +0100, Janning Vygen <[EMAIL PROTECTED]> said: > if [ `fcopy -iM /etc/ntp.conf` -gt "0" ] > it would be nice to have an option like > -n --number report number of files changed > and let fcopy report the number of files changed Good idea. But then fcopy must not write any error or verbose messages when using -n. Or should -n report the number of files changes using the exit/return code? -- regards Thomas
Re: Fcopy patch regarding debian bug report #360184
Am Mittwoch, 6. Dezember 2006 15:24 schrieb Henning Glawe:
> On Wed, Dec 06, 2006 at 12:01:03PM +0100, Janning Vygen wrote:
> > + return 1 if ( $suid != $duid or $sgid != $dgid or $smode != $dmode or
> > $stime != $dtime );
>
> thinking about (pre/post)inst scripts, comparing the time will lead to
> problems... (as destfile may have a different mtime because it has been
> modified by one of these scripts).
that's not a problem in my patch as set_mode is called anyway and only once
here regardless of the result of compare_modes:
original code in LATEST
set_mode($ps,$destfile,$class);
runscript("postinst",$ps,$destfile,$class);
original code in DEBIAN STABLE is
postinst($ps,$destfile,$class);
set_mode($ps,$destfile,$class);
so set_mode is called anyway after or before postinst. I dont know why the
utime is called at all. Is it importand to have the time set to the time of
the configuration file?
in my opinion all modtime code could be removed, but i kept it as i didn't
understand it and just wanted to fix the bug.
kind regards,
janning
--
PLANWERK 6 websolutions
Venloer Straße 8, 40477 Düsseldorf
Tel: (0211) 302666-0
Fax: (0211) 302666-10
http://www.planwerk6.de/
fcopy feature request: report if a file was copied
hi, as i use softupdate a lot , it would be nice to have a fcopy feature which reports if any files or file permissions were changed. Look at my script in BASE/50-ntp #!/bin/sh fcopy -iM /etc/ntp.conf /etc/init.d/ntp-server stop ntpd -q -g hwclock --systohc /etc/init.d/ntp-server start I dont want to start and stop my ntp server on each softupdate, only on my first install or when /etc/ntp.conf changes it would be nice to have something like this: if [ `fcopy -iM /etc/ntp.conf` -gt "0" ] then /etc/init.d/ntp-server stop ntpd -q -g hwclock --systohc /etc/init.d/ntp-server start fi it would be nice to have an option like -n --number report number of files changed and let fcopy report the number of files changed i would like to write a patch for fcopy implementing this if it does make sense and would become part of fai. kind regards Janning -- PLANWERK 6 websolutions Venloer Straße 8, 40477 Düsseldorf Tel: (0211) 302666-0 Fax: (0211) 302666-10 http://www.planwerk6.de/
Fcopy patch regarding debian bug report #360184
Hi fai,
attached there are two patch files for fcopy, one for debian stable version
and one for the latest from fai (which send Thomas Lange to me yesterday)
These Patch files should fix the bug
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=360184
"In case only the modes are changed, but the actual file stays the same,
these changes are not propagated; a possible workaround is IMHO
to call set_mode even if a file is preserved."
It just fixes the part of code where $opt_update is not set as i do not
understood what -p really does.
the main difference is in this line
if ( compare($tmpfile,$destfile) or compare_mode($ps,$destfile,$class)) {
instead of
if ( compare($tmpfile,$destfile) ){
There are a few new subroutines:
sub get_sourcefile_mode
sub get_destfile_mode
sub compare_mode
and some minor modifications to
sub set_mode
to use these new subroutines and avoid repeated code.
It was tested a few times in my environment and it works for me, but maybe
won't work for you.
i hope it works and will help anybody. Feel free to use/modify/discard but
don't claim me for any bugs.
Can anybody review this patch and maybe upload it to http://bugs.debian.org/
as i am not used to handle bug reporting databases.
kind regards
Janning
PS: i hope its allowed to send attachments to the list. netiquette is still a
TODO on
http://faiwiki.informatik.uni-koeln.de/index.php/Linux-fai_mailing_list_infos
--- fcopy 2006-12-06 11:49:47.306784009 +0100
+++ fcopy.deb31.fixed 2006-12-06 11:47:56.833502527 +0100
@@ -123,8 +123,7 @@
}
} else {
# compare literally
-
-if ( compare($sourcefile,$destfile)) {
+if ( compare($sourcefile,$destfile) or compare_mode($ps,$destfile,$class)) {
$logcomment="\t# new (literal)";
} else {
$logcomment="\t# preserved (literal)" if $logfile;
@@ -187,26 +186,55 @@
sub set_mode {
# set target file's owner, group, mode and time
- # use owner,group,mode from -m or from the file file-modes or
- # use the values from the source file
my ($sourcefile,$destfile,$class) = @_;
+ my ($uid,$gid,$mode,$stime) = get_sourcefile_mode($sourcefile,$class);
+ warn "chown/chmod u:$uid g:$gid m:$mode $destfile\n" if $debug;
+ chown ($uid,$gid, $destfile) || ewarn("chown $uid, $gid $destfile failed. $!");
+ chmod ($mode, $destfile) || ewarn("chmod $mode $destfile failed. $!");
+ utime ($stime,$stime, $destfile) || ewarn("utime for $destfile failed. $!");
+}
+# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
+sub get_sourcefile_mode {
+
+ # owner,group,mode from -m or from the file file-modes or
+ # the values from the source file
+ my ($sourcefile,$class) = @_;
my ($uid,$gid,$owner,$group,$mode);
# get mtime,uid,gid,mode from source file
my ($stime,@defmodes) = (stat("$sourcefile/$class"))[9,4,5,2];
-
if ($modeset) { # use -m values
($owner,$group,$mode) = @opt_modes;
} elsif (-f "$sourcefile/file-modes"){
($owner,$group,$mode) = read_file_mode("$sourcefile/file-modes",$class);
} else { # use values from source file
($owner,$group,$mode) = @defmodes;
+# see perldoc -f stat
+$mode = $mode & 0;
}
-
($uid,$gid) = name2num($owner,$group);
- warn "chown/chmod u:$uid g:$gid m:$mode $destfile\n" if $debug;
- chown ($uid,$gid, $destfile) || ewarn("chown $owner $group $destfile failed. $!");
- chmod ($mode, $destfile) || ewarn("chmod $mode $destfile failed. $!");
- utime ($stime,$stime, $destfile) || ewarn("utime for $destfile failed. $!");
+ return ($uid,$gid,$mode,$stime);
+}
+# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
+sub get_destfile_mode {
+
+ my $destfile = shift;
+ my ($owner,$group,$mode,$time) = (stat($destfile))[4,5,2,9];
+ my ($uid,$gid) = name2num($owner,$group);
+ # see perldoc -f stat
+ $mode = $mode & 0;
+ return ($uid,$gid,$mode,$time);
+}
+# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
+sub compare_mode {
+
+ # compare destination file's mode with source
+ # return 1 if differs / 0 if equals
+ my ($sourcefile,$destfile,$class) = @_;
+ my ($suid,$sgid,$smode,$stime) = get_sourcefile_mode($sourcefile,$class);
+ my ($duid,$dgid,$dmode,$dtime) = get_destfile_mode($destfile);
+
+ return 1 if ( $suid != $duid or $sgid != $dgid or $smode != $dmode or $stime != $dtime );
+ return 0;
}
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
sub check_mopt {
--- fcopy 2006-12-06 11:19:47.0 +0100
+++ fcopy.fixed 2006-12-06 11:19:47.0 +0100
@@ -145,7 +145,7 @@
runscript("preinst",$ps,$tmpfile,$class);
};
-if ( compare($tmpfile,$destfile)) {
+if ( compare($tmpfile,$destfile) or compare_mode($ps,$destfile,$class)) {
$logcomment=
Re: fcopy
Am Dienstag, 5. Dezember 2006 21:04 schrieb Michael Tautschnig: > > >>>>> On Tue, 5 Dec 2006 15:46:35 +0100, Janning Vygen > > >>>>> <[EMAIL PROTECTED]> said: > > >>>>> > > > If i fcopy a file which is alread there with same content, but the > > > file in SOURCE has other file access permissions, why it is not > > > copied? > > > > I should be copied, but the permissions should be changed. > > Which fcopy are you using? > > Maybe try fcopy -vD to get debuggin and verbose output. > > > > > Why does fcopy compares files only literally? > > > > Maybe a small bug. But first try -D for debugging output. > > > > If you join the IRC channel #fai on oftc.net and paste your output to > > paste.debian.net we may help you very quickly. > > Did you solve this issue already? Otherwise this is probably just what > #360184 is about: > > http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=360184 Yes, it seams to be the same bug. And it's still there in the newest version. IMHO this is a bug to worry about if you use softupdate, which can result in a security hole: i chmod a script executable in /etc/network/if-pre-up.d/ which contains iptables statements. Tthe executable flag was NOT copied and so my firewall is not in the state it should be. chmod a script can be an easy way to enable/disable a part of a firewall. This bug might be only a problem with softupdate of course. I am debugging fcopy at the moment and will send my patch today. kind regards, janning -- PLANWERK 6 websolutions Venloer Straße 8, 40477 Düsseldorf Tel: (0211) 302666-0 Fax: (0211) 302666-10 http://www.planwerk6.de/
Re: fcopy
> >>>>> On Tue, 5 Dec 2006 15:46:35 +0100, Janning Vygen <[EMAIL PROTECTED]> > >>>>> said: > > > If i fcopy a file which is alread there with same content, but the file > in > > SOURCE has other file access permissions, why it is not copied? > I should be copied, but the permissions should be changed. > Which fcopy are you using? > Maybe try fcopy -vD to get debuggin and verbose output. > > > Why does fcopy compares files only literally? > Maybe a small bug. But first try -D for debugging output. > > If you join the IRC channel #fai on oftc.net and paste your output to > paste.debian.net we may help you very quickly. > Did you solve this issue already? Otherwise this is probably just what #360184 is about: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=360184 Best, Michael pgpahTlDDPFfu.pgp Description: PGP signature
Re: fcopy
>>>>> On Tue, 5 Dec 2006 15:46:35 +0100, Janning Vygen <[EMAIL PROTECTED]> said: > If i fcopy a file which is alread there with same content, but the file in > SOURCE has other file access permissions, why it is not copied? I should be copied, but the permissions should be changed. Which fcopy are you using? Maybe try fcopy -vD to get debuggin and verbose output. > Why does fcopy compares files only literally? Maybe a small bug. But first try -D for debugging output. If you join the IRC channel #fai on oftc.net and paste your output to paste.debian.net we may help you very quickly. -- regards Thomas
fcopy
Hi, i have read the fcopy man page and didn't find a hint on the following problem: If i fcopy a file which is alread there with same content, but the file in SOURCE has other file access permissions, why it is not copied? I even tried to set the permissions with -m but no way. If the file is given and has the same content, nothing is copied but "preserved" as fcopy states. I would expect fcopy to copy the file if anything like content or permissions has changed. Ok, it is somewhat explaind in the -p option but i don't have a changelog file. I have no nfsroot and no cvs, i just rsync my /fai to the remote server and run 'fai -N softupdate'. And if i change file access permissions they are not copied which is bad in my opinion. Why does fcopy compares files only literally? kind regards Janning
Re: generating links via fcopy
> Hi!
>
> We'll be using fai to upgrade our Woody workstations to Sarge and
> maintain them. But sometimes we'd like to create links on the target
> machine in the same way, we copy files via fcopy (because we have to
> generate a lot of them).
Isn't it just as easy to create the links in one of your scripts, i.e.
> Example: a relative link is located for example in
> $fairoot/files/etc/mylink.extension/MYCLASS and points to
> ../usr/local/myrealfile.extension. It should be copied as is and
> /etc/mylink.extension should remain as relative link. Current behavior
> is following the link and copying the target file.
>
fcopy /usr/local/myrealfile.extension && ln -s /usr/local/myrealfile.extension
${target}/etc/mylink.extension
> We'd like to extend fcopy that it uses readlink/symlink to keep links
> intact if the parameter for example -H is present and the current file
> is a link.
>
> Has someone did something like this before? Is the dev-team interested
> in the patch?
>
In my opinion, this is somewhat dangerous: It changes the semantics of the
files/ directory, although, of course, only if the -H switch is given.
I'd rather suggest to build a small function:
fcopy_links() {
local fcopy_args sources eff_sources
for arg ; do
if echo $arg | grep -q "^-" then
fcopy_args="$fcopy_args $arg"
else
sources="$sources $arg"
fi
done
for i in $sources ; do
if [ -L ${FAI}/$i ] ; then
$eff_sources="$eff_sources $i/`readlink ${FAI}/$i`"
ln -s $i/`readlink ${FAI}/$i` ${target}/$i
else
$eff_sources="$eff_sources $i"
fi
done
fcopy $fcopy_args $eff_sources
}
HTH,
Michael
generating links via fcopy
Hi! We'll be using fai to upgrade our Woody workstations to Sarge and maintain them. But sometimes we'd like to create links on the target machine in the same way, we copy files via fcopy (because we have to generate a lot of them). Example: a relative link is located for example in $fairoot/files/etc/mylink.extension/MYCLASS and points to ../usr/local/myrealfile.extension. It should be copied as is and /etc/mylink.extension should remain as relative link. Current behavior is following the link and copying the target file. We'd like to extend fcopy that it uses readlink/symlink to keep links intact if the parameter for example -H is present and the current file is a link. Has someone did something like this before? Is the dev-team interested in the patch? Bye, Patrick Cornelissen
Re: fcopy for single files ?
Michael Tautschnig wrote: [...] The point is: fcopy (and all other parts of FAI too) uses classes to determine, which version of a file should be copied, because, e.g., /etc/hostname can't be the same on all hosts installed with FAI. Thus, all versions of this single file are kept in one directory /fai/files/etc/hostname and fcopy determines, which of them has to copied to the host being set up. Consider the following example: Your hosts are "alpha" and "beta" and you want to copy /etc/hostname using fcopy. In that case, you need /fai/files/etc/hostname/alpha and /fai/files/etc/hostname/beta Then "fcopy /etc/hostname" will do the job for you. HTH, Michael Thanks. It's works as I want now. I was in the bad context of something like /fai/files/alpha/etc/hostname /fai/files/beta/etc/hostname The fcopy message could be more clear for newbees like me, distinguishing 2 cases: if file exists and is not a directory -> "$ps should be a directory containing templates" if not existent -> "inexistent directory: $ps" Thanks again. -- Fabrice
Re: fcopy for single files ?
> I don't understand if "fcopy" can be used to copy single file. > > Maybe Yes: > > - The is examples in "/usr/share/doc/fai/examples" containing single copy : > simple/scripts/FAIBASE/10-misc contains "fcopy -i /etc/modules" > > - It seems useful indeed. > > Maybe No: > > - The manual is saying "Each SOURCE argument is a directory" > > - The logs of "successfull installed computers" as > http://www.informatik.uni-koeln.de/fai/logs/demohost/last-install/shell.log > show things as > > "fcopy: /fai/files/etc/hostname is not a directory. Not copied." > > [...] The point is: fcopy (and all other parts of FAI too) uses classes to determine, which version of a file should be copied, because, e.g., /etc/hostname can't be the same on all hosts installed with FAI. Thus, all versions of this single file are kept in one directory /fai/files/etc/hostname and fcopy determines, which of them has to copied to the host being set up. Consider the following example: Your hosts are "alpha" and "beta" and you want to copy /etc/hostname using fcopy. In that case, you need /fai/files/etc/hostname/alpha and /fai/files/etc/hostname/beta Then "fcopy /etc/hostname" will do the job for you. HTH, Michael
fcopy for single files ?
I don't understand if "fcopy" can be used to copy single file. Maybe Yes: - The is examples in "/usr/share/doc/fai/examples" containing single copy : simple/scripts/FAIBASE/10-misc contains "fcopy -i /etc/modules" - It seems useful indeed. Maybe No: - The manual is saying "Each SOURCE argument is a directory" - The logs of "successfull installed computers" as http://www.informatik.uni-koeln.de/fai/logs/demohost/last-install/shell.log show things as "fcopy: /fai/files/etc/hostname is not a directory. Not copied." Maybe I'm wrong... or there is some bug somewere ? I'm talking about fcopy "Version 1.23, 24-march-2005" coming with fai_2.8.4_all.deb -- [EMAIL PROTECTED] LPN-CNRS, route de Nozay, 91460 Marcoussis (FRANCE)
Re: recursive fcopy for the complete files directory?
[...]
>
> Which is somehow ugly (I often don't like the idea anyway of having
> these files named as classes, but when i thought about having another
> intermediate directory with the class name and put the file in it, I
> didn't like that also, so it seems the way it is is the one that sucks
> less).
> I'd like to able to just manage a whole
> directory of files belonging together to one useful configuration
> in a single directory like:
>
> /files/etc/fai/UBUNTU-FAI-SERVER/
> fai.conf
> make-fai-nfsroot.conf
> menu-lst
> sources.list
>
> /files/etc/fai/FAI-SERVER/
> fai.conf
> make-fai-nfsroot.conf
> menu-lst
> sources.list
>
> Would it be a bad idea if I'd add an option to fcopy that could handle
> this? Hmm, I guess it could work and be nice, but it might interfere
> with the "-r" when I use it as Henning does...
>
[...]
As far as I can see, at least keeping such a structure within the
"/files"-subtree wouldn't break "fcopy -r" -- it simply wouldn't do anything
with those files, as long as they aren't named like any of your classes.
Thus, you could probably do it with a quite simple bash-function for each of
your directories that demand special treatment:
copy_dir() {
#params: directory
local dir=$1
local class_dir
if [ ! -d "$FAI/files/$dir" ] ; then
echo "Source directory not found!"
return 1
fi
for c in $classes ; do
[ -d "$FAI/files/$dir/$c" ] && class_dir="$FAI/files/$dir/$c"
done
if [ -z $class_dir ] ; then
echo "No usable directory found!"
return 2
fi
local current=`pwd`
local pattern
cd $class_dir
if [ `ls -d .* | wc -l` -gt 2 ] ; then
tar cfp - --exclude="CVS" --exclude=".svn" --exclude="postinst" {.[^.]*,*}
| tar xf - -C ${target}/$dir
else
tar cfp - --exclude="CVS" --exclude=".svn" --exclude="postinst" * | tar xf
- -C ${target}/$dir
fi
cd $current
if [ -x $class_dir/postinst ] ; then
$class_dir/postinst
return $?
else
return 0
fi
}
Well, there is no way to support "file-modes", at least no using the current
format of file-modes.
HTH,
Michael
Re: recursive fcopy for the complete files directory?
On Thu, Jun 02, 2005 at 12:06:38PM +0200, Henning Sprang wrote: > On Thu, 2005-06-02 at 10:58 +0200, Henning Glawe wrote: > > But the major question would be: is this really necessairy? > > Necessary in terms of it doesn't work otherwise, no. > > I agree that I shouldn't think about it any further. I guess I will > continue to, for some services, store the configuration directory in a > separate subversion tree and manage it there, checking it out at install > time. what about keeping those class-dependent subtrees in a tar, and using ftar, instead? Steffen
Re: recursive fcopy for the complete files directory?
On Thu, 2005-06-02 at 10:58 +0200, Henning Glawe wrote: > But the major question would be: is this really necessairy? Necessary in terms of it doesn't work otherwise, no. I agree that I shouldn't think about it any further. I guess I will continue to, for some services, store the configuration directory in a separate subversion tree and manage it there, checking it out at install time. Henning
Re: recursive fcopy for the complete files directory?
On Thu, Jun 02, 2005 at 10:33:28AM +0200, Henning Sprang wrote: > /files/etc/fai/ > fai.conf/ > UBUNTU-FAI-SERVER > FAI-SERVER > make-fai-nfsroot.conf/ > UBUNTU-FAI-SERVER > FAI-SERVER > menu-lst/ > UBUNTU-FAI-SERVER > FAI-SERVER > sources.list/ > UBUNTU-FAI-SERVER > FAI-SERVER > > Which is somehow ugly (I often don't like the idea anyway of having > these files named as classes, but when i thought about having another > intermediate directory with the class name and put the file in it, I > didn't like that also, so it seems the way it is is the one that sucks > less). > I'd like to able to just manage a whole > directory of files belonging together to one useful configuration > in a single directory like: > > /files/etc/fai/UBUNTU-FAI-SERVER/ > fai.conf > make-fai-nfsroot.conf > menu-lst > sources.list > > /files/etc/fai/FAI-SERVER/ > fai.conf > make-fai-nfsroot.conf > menu-lst > sources.list > > Would it be a bad idea if I'd add an option to fcopy that could handle > this? Hmm, I guess it could work and be nice, but it might interfere > with the "-r" when I use it as Henning does... not only this, but it would also break consistency of the files/ hierarchy (which is, btw, already broken if you use the files/packages cruft ;) ) if you want to do something like this, open a new hierarchy and don't use files/. But the major question would be: is this really necessairy? > P.S: I know about ftar, but I wanna do version control - if I put > *.tar's in the files dir, there's no easy way to get diff's from the > text files inside. I'd recommend to use fcopy just as-is... -- c u henning
Re: recursive fcopy for the complete files directory?
On Thu, 2005-06-02 at 08:51 +0200, Henning Glawe wrote: > [...] > thats what fcopy's recursion code is for: I am calling > > fcopy -i -r . > > in scripts/DEFAULT/S01recursive_fcopy.sh, that means at the really first > scripts step (the -i sets the exitcode always to 0, because if you are doing > a recursive fcopy, the probability of some not-matching files is high). > > I am using this construct for at least 3 years now, so it should be pretty > safe. > Cool, will try that later. Thanks! Another thing comes to my mind that I wanted to do some time ago, but I don't remember the correct example back then - maybe the FAI config is a good one: For some services, there is not just one file, but a directory with multiple files in it which belong together as configuration. Now, if I have for example two different fai configs, say one for a sarge install server, one for an ubuntu install server, but I want them to reside on the installed machine in /etc/fai, assuming I never want an install server that does them both together (which I want finally), I have to make a setup like this in files: /files/etc/fai/ fai.conf/ UBUNTU-FAI-SERVER FAI-SERVER make-fai-nfsroot.conf/ UBUNTU-FAI-SERVER FAI-SERVER menu-lst/ UBUNTU-FAI-SERVER FAI-SERVER sources.list/ UBUNTU-FAI-SERVER FAI-SERVER Which is somehow ugly (I often don't like the idea anyway of having these files named as classes, but when i thought about having another intermediate directory with the class name and put the file in it, I didn't like that also, so it seems the way it is is the one that sucks less). I'd like to able to just manage a whole directory of files belonging together to one useful configuration in a single directory like: /files/etc/fai/UBUNTU-FAI-SERVER/ fai.conf make-fai-nfsroot.conf menu-lst sources.list /files/etc/fai/FAI-SERVER/ fai.conf make-fai-nfsroot.conf menu-lst sources.list Would it be a bad idea if I'd add an option to fcopy that could handle this? Hmm, I guess it could work and be nice, but it might interfere with the "-r" when I use it as Henning does... P.S: I know about ftar, but I wanna do version control - if I put *.tar's in the files dir, there's no easy way to get diff's from the text files inside. Henning
Re: recursive fcopy for the complete files directory?
Hi Henning,
On Wed, Jun 01, 2005 at 10:19:56PM +0200, Henning Sprang wrote:
> Hi,
> Maybe I miss something, but when I rename some files
> in /usr/local/share/fai/files, I always have to change the corresponding
> script that is responsible for copying this file to it's place.
>
> Similar double-actions are required when adding or removing a file.
>
> I would rather like to do a single fcopy command that makes sure every
> file in "files" gets copied to the install client according to matching
> class definitions, so I don't have to take care of anything else then.
I'm using a script to accomplish this which roughly looks this way:
(snipped and pasted)
# install all files available (using fcopy)
filelist=`cd /fai/files; find . -type f -print \
| sed 's-^\./\(.*\)/.*-\1-' | grep -v 'CVS$' | grep -v '^packages' \
| sort | uniq`
# ugly patch: remove /etc/resolv.conf
(cd /tmp/target/etc; /bin/mv -f resolv.conf resolv.conf.BAD)
for file in $filelist ; do
# -B: no backup files! would be included in conf.d directories :(
# -i: ignore "no matching class"
# [-M: default modes: would destroy executables]
# [-v: verbose output, only for debugging]
fcopy -Bi $file
done
The resolv.conf thing is still in there from old days (dunno whether
it's still needed): fcopy would try to follow the symlink and replace
a file in the ro-mounted root.
If you're using SVN, you'd have to extend the find | grep a bit...
> Is something like that available? Or do I forget/oversee something which
> makes this a bad idea?
I've been doing this for quite a long time and nobody complained...
Note that some files would be fcopied before!
Cheers,
Steffen
--
Steffen Grunewald * MPI fuer Gravitationsphysik (Albert-Einstein-Institut)
SciencePark Golm, Am Mühlenberg 1, D-14476 Potsdam * http://www.aei.mpg.de
* e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:7298}
No Word/PPT mails - http://www.gnu.org/philosophy/no-word-attachments.html
Re: recursive fcopy for the complete files directory?
On Wed, Jun 01, 2005 at 10:19:56PM +0200, Henning Sprang wrote: > Maybe I miss something, but when I rename some files > in /usr/local/share/fai/files, I always have to change the corresponding > script that is responsible for copying this file to it's place. > > Similar double-actions are required when adding or removing a file. > > I would rather like to do a single fcopy command that makes sure every > file in "files" gets copied to the install client according to matching > class definitions, so I don't have to take care of anything else then. > > Is something like that available? Or do I forget/oversee something which > makes this a bad idea? thats what fcopy's recursion code is for: I am calling fcopy -i -r . in scripts/DEFAULT/S01recursive_fcopy.sh, that means at the really first scripts step (the -i sets the exitcode always to 0, because if you are doing a recursive fcopy, the probability of some not-matching files is high). I am using this construct for at least 3 years now, so it should be pretty safe. -- c u henning
Re: recursive fcopy for the complete files directory?
> Hi, > Maybe I miss something, but when I rename some files > in /usr/local/share/fai/files, I always have to change the corresponding > script that is responsible for copying this file to it's place. > > Similar double-actions are required when adding or removing a file. > > I would rather like to do a single fcopy command that makes sure every > file in "files" gets copied to the install client according to matching > class definitions, so I don't have to take care of anything else then. > > Is something like that available? Or do I forget/oversee something which > makes this a bad idea? > There shouldn't be any problem, as long as you take care of - the correct file modes (probably using "file-modes") - the classes, you are using -- I personally tend to use "DEFAULT" or similar classes, that would catch to many hosts, in cases where I know, that my scripts only copy it onto the relevant hosts. AFAIK Henning Glawe just does "fcopy -r" in one of his scripts, so it does work! Best regards, Michael
recursive fcopy for the complete files directory?
Hi, Maybe I miss something, but when I rename some files in /usr/local/share/fai/files, I always have to change the corresponding script that is responsible for copying this file to it's place. Similar double-actions are required when adding or removing a file. I would rather like to do a single fcopy command that makes sure every file in "files" gets copied to the install client according to matching class definitions, so I don't have to take care of anything else then. Is something like that available? Or do I forget/oversee something which makes this a bad idea? Henning
Crossposting to FAI Lists and Debian BTS - was: "Re: fai: fcopy does not resolve uid/gid"
Hy, As Thomas already pointed out to me in private mail some time ago: can we agree that mails to the Debian BTS are not cross-posted to any of the FAI lists, except it is really so very important that we don't want anybody to miss the information? Reason: everybody who has subscribed to the fai package in the Debian BTS gets these mails twice, and I would like to propose instead, that everybody interested in the FAI package and all it's bugs just subscribes the FAI package in the debian BTS. Because it's really hard to find where to subscribe (at least even I had to search some minutes - and i just did it last week) - it's there: http://packages.qa.debian.org/f/fai.html please let me know if this is an inappropriate idea... Henning
Re: fai: fcopy does not resolve uid/gid
On Tue, May 24, 2005 at 09:24:22AM +0200, Steffen Grunewald wrote:
> On Mon, May 23, 2005 at 08:36:40PM +0200, Michael Tautschnig wrote:
> > diff -rN -u old-MT_FAI/scripts/fcopy new-MT_FAI/scripts/fcopy
> [...]
> > - my $uid = ($user =~ /^\d+$/) ? $user : getpwnam $user;
> > - my $gid = ($group =~ /^\d+$/) ? $group : getgrnam $group;
> > + my $uid = ($user =~ /^\d+$/) ? $user : `$ENV{ROOTCMD} perl -e '\$uid =
> > getpwnam $user; print \$uid'`;
> > + my $gid = ($group =~ /^\d+$/) ? $group : `$ENV{ROOTCMD} perl -e '\$gid =
> > getgrnam $group; print \$gid'`;
[...]
> I'll try to give it a try... had intended to do a full installation today
> anyway :-)
Except that the "fcopy" script resides in /usr/sbin (both locally and in
nfsroot), the patch appears to work. And I think that's the proper way...
Thomas, can you put this patch on your list for upcoming versions?
(it might even pass Vorlon's review... thinking about it a bit longer,
this *may* be a critical bug (since it forces the installer to remove
some access limitations to make files accessible at all), feel free to
change the bug level...)
Cheers,
Steffen
--
Steffen Grunewald * MPI fuer Gravitationsphysik (Albert-Einstein-Institut)
SciencePark Golm, Am Mühlenberg 1, D-14476 Potsdam * http://www.aei.mpg.de
* e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:7298}
No Word/PPT mails - http://www.gnu.org/philosophy/no-word-attachments.html
Re: fai: fcopy does not resolve uid/gid
> On Mon, May 23, 2005 at 08:36:40PM +0200, Michael Tautschnig wrote:
> > Hi Steffen,
> >
> > if you're still interested in that problem - could you give the following
> > patch
> > a try? IMHO the problem is, that logcheck does not exist within your
> > nfsroot, it
> > only exists on the future client. Thus I'd chroot before doing
> > getpwnam/getgrnam.
> >
> > diff -rN -u old-MT_FAI/scripts/fcopy new-MT_FAI/scripts/fcopy
> [...]
> > - my $uid = ($user =~ /^\d+$/) ? $user : getpwnam $user;
> > - my $gid = ($group =~ /^\d+$/) ? $group : getgrnam $group;
> > + my $uid = ($user =~ /^\d+$/) ? $user : `$ENV{ROOTCMD} perl -e '\$uid =
> > getpwnam $user; print \$uid'`;
> > + my $gid = ($group =~ /^\d+$/) ? $group : `$ENV{ROOTCMD} perl -e '\$gid =
> > getgrnam $group; print \$gid'`;
>
> Will $ROOTCMD be set properly for upgrades?
It's just empty, which should not be a problem (fcopy does not use perl -w, so
not even warnings will be issued).
>
> I'll try to give it a try... had intended to do a full installation today
> anyway :-)
>
Thanks!
Michael
Re: fai: fcopy does not resolve uid/gid
On Mon, May 23, 2005 at 08:36:40PM +0200, Michael Tautschnig wrote:
> Hi Steffen,
>
> if you're still interested in that problem - could you give the following
> patch
> a try? IMHO the problem is, that logcheck does not exist within your nfsroot,
> it
> only exists on the future client. Thus I'd chroot before doing
> getpwnam/getgrnam.
>
> diff -rN -u old-MT_FAI/scripts/fcopy new-MT_FAI/scripts/fcopy
[...]
> - my $uid = ($user =~ /^\d+$/) ? $user : getpwnam $user;
> - my $gid = ($group =~ /^\d+$/) ? $group : getgrnam $group;
> + my $uid = ($user =~ /^\d+$/) ? $user : `$ENV{ROOTCMD} perl -e '\$uid =
> getpwnam $user; print \$uid'`;
> + my $gid = ($group =~ /^\d+$/) ? $group : `$ENV{ROOTCMD} perl -e '\$gid =
> getgrnam $group; print \$gid'`;
Will $ROOTCMD be set properly for upgrades?
I'll try to give it a try... had intended to do a full installation today
anyway :-)
Cheers,
Steffen
Re: fai: fcopy does not resolve uid/gid
Hi Steffen,
if you're still interested in that problem - could you give the following patch
a try? IMHO the problem is, that logcheck does not exist within your nfsroot, it
only exists on the future client. Thus I'd chroot before doing
getpwnam/getgrnam.
diff -rN -u old-MT_FAI/scripts/fcopy new-MT_FAI/scripts/fcopy
--- old-MT_FAI/scripts/fcopy2005-05-17 15:01:06.0 +0200
+++ new-MT_FAI/scripts/fcopy2005-05-23 20:33:40.0 +0200
@@ -178,8 +178,8 @@
# convert names to numeric uid, gid
my ($user,$group) = @_;
- my $uid = ($user =~ /^\d+$/) ? $user : getpwnam $user;
- my $gid = ($group =~ /^\d+$/) ? $group : getgrnam $group;
+ my $uid = ($user =~ /^\d+$/) ? $user : `$ENV{ROOTCMD} perl -e '\$uid =
getpwnam $user; print \$uid'`;
+ my $gid = ($group =~ /^\d+$/) ? $group : `$ENV{ROOTCMD} perl -e '\$gid =
getgrnam $group; print \$gid'`;
warn "name2id $user = $uid ; $group = $gid\n" if $debug;
return ($uid,$gid);
}
Thanks in advance,
Michael
questions: fcopy, how to encrypt password, network config
I was able to get fai to install with the 2.6.8 kernel (on client), but for some reason it's not online (I had assumed it would have used dhcp). So in trying to configure fai more, I have some general questions that I wasn't able to find quickly on google :) - I see fcopy used with only the source files specified. e.g. fcopy /etc/host.conf /etc/nsswitch.conf /etc/hosts.equiv. Is the base dir the nfsroot directory? ie. if I have nfsroot=/diskless_ro then it's copying from /diskless_ro/etc/host.conf? - this question might be related to the previous one, but where is the /etc/network/interfaces that's used? I want to make sure that dhcp is used. - in the FAIBASE.var class, I see the rootpw='56hNVqht51tzc' as default. How can I easily generate this password? - I need to specify kernel boot parameters for the installed client kernel, e.g. I have in my other computers this grub stuff: title Debian GNU/Linux, kernel 2.6.8-1-386 root(hd0,0) kernel /boot/vmlinuz-2.6.8-1-386 root=/dev/hda1 ro console=ttyS1,19200 console=tty0 initrd /boot/initrd.img-2.6.8-1-386 savedefault boot The kernel installed from aptitude automatically generates it's own config, so how do I add this? ...I think that's enough questions for now... cheers, dan --
Re: fcopy gives permission denied
On Tuesday 15 February 2005 17:12, Holger Levsen wrote: > if the file in /fai/files also has permissions of root,root,0600 then it's > not readable when exported (try it on the client). You have to loose the > permissions which might introduce a security breach. Yes this works. Thanks
Re: fcopy gives permission denied
Hi, On Tuesday 15 February 2005 16:36, you wrote: > gevaert:/home/fai/antldap/last# cat /ed/fai_configdir/scripts/LDAPS > #!/bin/sh > fcopy -ir -m root,root,600 /etc/ldap/slapd.conf > > > the shell.log reports: > > shell.log:fcopy: copy /fai/files/etc/ldap/slapd.conf/LDAPS to > /tmp/target/etc/ldap/slapd.conf failed. Permission denied if the file in /fai/files also has permissions of root,root,0600 then it's not readable when exported (try it on the client). You have to loose the permissions which might introduce a security breach. regards, Holger
fcopy gives permission denied
Hello, I have the following script: gevaert:/home/fai/antldap/last# cat /ed/fai_configdir/scripts/LDAPS #!/bin/sh fcopy -ir -m root,root,600 /etc/ldap/slapd.conf the shell.log reports: shell.log:fcopy: copy /fai/files/etc/ldap/slapd.conf/LDAPS to /tmp/target/etc/ldap/slapd.conf failed. Permission denied Why can't I copy that file? I don't see any problem in the original file have permissions 600. And then being replace with a newer file. Thanks in advance, Rudy
Re: Passwords in FAI (Was: Re: Copy directories with fcopy)
> On Thu, 13 Jan 2005 14:23:37 +0200, [EMAIL PROTECTED] (Markus Törnqvist) > said: > Must I enable crypts manually somewhere? In which format is the > default root password after fai? 16-bit MD5? Password are normal crypt password without using /etc/shadow. This will change in fai 2.6.6 (comming soon). Have a look at scripts/LAST in fai 2.6.6 -- regards Thomas
Re: Passwords in FAI (Was: Re: Copy directories with fcopy)
On Thu, Jan 13, 2005 at 02:23:37PM +0200, Markus TXrnqvist wrote: > On Wed, Jan 12, 2005 at 07:05:26PM +0100, Henning Glawe wrote: > >1) copying passwd is not a good idea: it contains many dynamic entries > > generated by packages (and thus changes quite often) > > Well, in this case it's pretty much a good idea :) > I'd like to company sysadmin team to be able to log in, if they > change their passwords, it's not an installation problem. What about having a locked root password and installing your sysadmin public ssh keys instead? > Customer accounts are added manually afterwards in either case. On the YP server, we run a script to extract user accounts (strip everything with UID below 1000) and upload this to FAIconfig CVS periodically... It's still work in progress, but it's quite close: http://pandora.aei.mpg.de/cgi-bin/viewcvs.cgi/faiconfig/ Of course, hints are welcome... Cheers, Steffen
Re: Passwords in FAI (Was: Re: Copy directories with fcopy)
On Thu, Jan 13, 2005 at 02:23:37PM +0200, Markus Tïrnqvist wrote: > On Wed, Jan 12, 2005 at 07:05:26PM +0100, Henning Glawe wrote: > >1) copying passwd is not a good idea: it contains many dynamic entries > > generated by packages (and thus changes quite often) > > Well, in this case it's pretty much a good idea :) > I'd like to company sysadmin team to be able to log in, if they > change their passwords, it's not an installation problem. > > Customer accounts are added manually afterwards in either case. but if you install software, the passwd tends to change already: packages such as ssh create users and/or groups, and the order in which the installation happens determines the uids; if you later install a non-matching /etc/passwd, the permissions in the filesystem are garbled. thats why I use a script to add the necessary account(s) using 'adduser' and postprocess /etc/passwd afterwards to insert the correct password hashes. > >2) unless you run fcopy recursively somewhere you need to call it for each > > file. > > I made a script like USERS/S666 or something to write the files but > there's a new problem. > > It doesn't authenticate, no matter what. > I compared the PAM configuration and all's the same there as on the > install server, as well as the file contents and permissions. > > Must I enable crypts manually somewhere? In which format is the > default root password after fai? 16-bit MD5? this depends on the configuration you are using. try to check $FAI/scripts/* for things messing around with the passwords... One quick idea: maybe it is a problem with shadow passwords: if they are enabled, password hashes are in /etc/shadow and not in /etc/passwd... -- c u henning
Passwords in FAI (Was: Re: Copy directories with fcopy)
On Wed, Jan 12, 2005 at 07:05:26PM +0100, Henning Glawe wrote: >1) copying passwd is not a good idea: it contains many dynamic entries > generated by packages (and thus changes quite often) Well, in this case it's pretty much a good idea :) I'd like to company sysadmin team to be able to log in, if they change their passwords, it's not an installation problem. Customer accounts are added manually afterwards in either case. >2) unless you run fcopy recursively somewhere you need to call it for each > file. I made a script like USERS/S666 or something to write the files but there's a new problem. It doesn't authenticate, no matter what. I compared the PAM configuration and all's the same there as on the install server, as well as the file contents and permissions. Must I enable crypts manually somewhere? In which format is the default root password after fai? 16-bit MD5? So close yet so far ;) Thanks! -- mjt
Re: Maintaining passwd [was Re: Copy directories with fcopy]
On Wed, Jan 12, 2005 at 06:58:27PM +, Michael Gellman wrote: > On Wed, 2005-01-12 at 19:05 +0100, Henning Glawe wrote: > > 1) copying passwd is not a good idea: it contains many dynamic entries > >generated by packages (and thus changes quite often) > > Can I ask how you handle adding user accounts to systems installed with > FAI? I currently store my passwd and group file in FAI, and have to make > sure to manually update it to add new users, which is beginning to > become an administrative problem. well, if the client(s) has got network access, a directory service like nis or ldap is the best way (from the viewpoint of administrative costs). if this is not possible, I'd add a script adding the users via adduser, and substituting the passwords (if needed). -- c u henning
Maintaining passwd [was Re: Copy directories with fcopy]
On Wed, 2005-01-12 at 19:05 +0100, Henning Glawe wrote: > 1) copying passwd is not a good idea: it contains many dynamic entries >generated by packages (and thus changes quite often) Can I ask how you handle adding user accounts to systems installed with FAI? I currently store my passwd and group file in FAI, and have to make sure to manually update it to add new users, which is beginning to become an administrative problem. --Mike
Re: Copy directories with fcopy
On Wed, Jan 12, 2005 at 07:58:20PM +0200, Markus Tïrnqvist wrote: > nurse:/mnt/z/fai/files/etc# ls -l passwd > total 4 > -rw-r--r-- 1 root root 1463 Jan 10 18:53 USERS > nurse:/mnt/z/fai/files/etc# ls -ld passwd > drwxr-xr-x 2 root root 72 Jan 12 18:00 passwd > > Basically I can do it in a separate script, I mean, write the > passwd and group files, but is this a feature that passwd etc > are not copied? > > This is mildly annoying :) 1) copying passwd is not a good idea: it contains many dynamic entries generated by packages (and thus changes quite often) 2) unless you run fcopy recursively somewhere you need to call it for each file. -- c u henning
Re: Copy directories with fcopy
On Fri, Jan 07, 2005 at 01:22:39PM +0100, Thomas Lange wrote: >fcopy is made for directory structure where you only need to copy a >couple of files. It's not intended for copying the whole directory >tree. Therefore you can use ftar(8). This refuses to work for me: nurse:/mnt/z/fai/files/etc# ls -l passwd total 4 -rw-r--r-- 1 root root 1463 Jan 10 18:53 USERS nurse:/mnt/z/fai/files/etc# ls -ld passwd drwxr-xr-x 2 root root 72 Jan 12 18:00 passwd Basically I can do it in a separate script, I mean, write the passwd and group files, but is this a feature that passwd etc are not copied? This is mildly annoying :) Thanks! -- mjt
Re: Copy directories with fcopy
Thomas Lange wrote: Fine! Seems that you are satisfied with your autom. installation of your first batch. Can you please fill out the fai questionnaire? http://www.informatik.uni-koeln.de/fai/questionnaire I will do that. fcopy is made for directory structure where you only need to copy a couple of files. It's not intended for copying the whole directory tree. Therefore you can use ftar(8). Thanks, I get it now! Cheers, Ben.
Re: Copy directories with fcopy
Steffen Grunewald wrote: I'm wondering if it's possible to use fcopy to just copy an entire directory structure for a defined class, rather than creating directories with the filenames and renaming the actual file as the classname? I have tried various ways but cannot do it Could ftar be what you're looking for? (If not, I didn't understand the problem, which is no surprise at all, it's Friday...) Yes! Thanks Steffen, didn't even think about that, it works fine. Cheers, Ben.
Re: Copy directories with fcopy
>>>>> On Fri, 07 Jan 2005 11:56:22 +, Ben Willcox <[EMAIL PROTECTED]> said: > Hi there, > Trying to use FAI again for a second batch of computers with a new > configuration. Fine! Seems that you are satisfied with your autom. installation of your first batch. Can you please fill out the fai questionnaire? http://www.informatik.uni-koeln.de/fai/questionnaire > I'm wondering if it's possible to use fcopy to just copy an entire > directory structure for a defined class, rather than creating > directories with the filenames and renaming the actual file as the > classname? I have tried various ways but cannot do it fcopy is made for directory structure where you only need to copy a couple of files. It's not intended for copying the whole directory tree. Therefore you can use ftar(8). -- regards Thomas
Re: Copy directories with fcopy
On Fri, Jan 07, 2005 at 11:56:22AM +, Ben Willcox wrote: > Hi there, > > Trying to use FAI again for a second batch of computers with a new > configuration. > > I'm wondering if it's possible to use fcopy to just copy an entire > directory structure for a defined class, rather than creating > directories with the filenames and renaming the actual file as the > classname? I have tried various ways but cannot do it Could ftar be what you're looking for? (If not, I didn't understand the problem, which is no surprise at all, it's Friday...) Steffen
Copy directories with fcopy
Hi there, Trying to use FAI again for a second batch of computers with a new configuration. I'm wondering if it's possible to use fcopy to just copy an entire directory structure for a defined class, rather than creating directories with the filenames and renaming the actual file as the classname? I have tried various ways but cannot do it I hope what I'm asking makes sense, but from what I understand the way the /fai/files directory works is you will build a common file structure, which contains class specific files within that structure. To me that seems a little confusing when you have many different types of machine build, and more importantly it is a lot of work creating directories as the filename and renaming files etc, when there are many files in there - it would seem easier if I could create a separate subdirectory within /fai/files per class, and then have a filestructure underneath that which would get copied to the relevent client machines. Perhaps I am using FAI incorrectly, but I would normally build one machine manually and get it working to my desired spec, and then try to configure FAI to install the same config to my fresh batch of machines. What I would like to do is directly copy the config files from /etc or wherever on my manually installed machine to the FAI server using scp, and then FAI copies these back to the auto installed machines using the same directory structure. Is this the correct way to do it? Thanks, Ben.
Re: fcopy problem
On Sun, Nov 28, 2004 at 06:16:24PM +0100, Holger Levsen wrote: > As a workaround you can do a "$ROOTCMD chgrp" afterwards... but: please fill > a bug in the debian BTS (bug tracking system, http://bugs.debian.org) to help > Thomas _and_others_. Good point, done. S
Re: fcopy problem
Hi, > The problem I can see is that within /usr/sbin/fcopy, the name2num > subroutine does not properly return the gid for group logcheck because > in the install system there isn't such a group - although it should > better look in the target system. The result is that the file > permissions are set, the owner (root) is set, but the group pertains to > be root too, so logcheck won't be able to access the file, causing the > exclusion to fail. > Any ideas? As a workaround you can do a "$ROOTCMD chgrp" afterwards... but: please fill a bug in the debian BTS (bug tracking system, http://bugs.debian.org) to help Thomas _and_others_. regards, Holger
fcopy problem
Hi,
I'd like to add some exclusion rules for logcheck, and put them into
/etc/logcheck/ignore.d.server/ subdirectories. I also added file-modes
files (with root logcheck 0640 in them).
The problem I can see is that within /usr/sbin/fcopy, the name2num
subroutine does not properly return the gid for group logcheck because
in the install system there isn't such a group - although it should
better look in the target system. The result is that the file
permissions are set, the owner (root) is set, but the group pertains to
be root too, so logcheck won't be able to access the file, causing the
exclusion to fail.
Of course, a simple workaround would be to set the permissions to 644,
and ignore the group settings - but I could imagine that under certain
circumstances this feature might be more important (SSL web server etc.)
The "getpwnam"/"getgrnam" calls should work in the target environment
instead of the install env...
Any ideas?
Cheers,
Steffen
--
Steffen Grunewald * * * Merlin cluster admin (http://pandora.aei.mpg.de)
Albert-Einstein-Institut (MPI Gravitationsphysik, http://www.aei.mpg.de)
Science Park Golm, Am Mühlenberg 1, 14476 Potsdam, Germany
e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:7298}
Re: fcopy and ignoring .svn directories when copying recursively
On Fri, Nov 05, 2004 at 11:12:31AM +0100, Steffen Grunewald wrote:
> If (as I suppose) subversion internal information is stored in a .svn/
> directory in every subdir of the tree, then the fifth line should be
> replaced by
>
> -d && ! ($_ =~ /^.*CVS$/) && ! ($_ =~ /^.*\.svn/) && ($nlink=(lstat($_))[3])
> &&
>
> (question to Thomas: shouldn't there be a \/ prepended to CVS - and .svn ?)
To answer the question (this messy bit of code was one of my first
contributions to FAI...):
when using File::Find, $_ only contains the basename, not the complete path,
so matching on '/' would break the code.
the attached patch (already submitted to thomas) contains a clean solution:
it cleans up the code, making it actually readable and makes the list of
ignored directories configurable.
--
c u
henning
--- orig/man/fcopy.8
+++ mod/man/fcopy.8
@@ -65,6 +65,11 @@
.B \-i
Ignore warnings about no matching class and always exit with 0.
.TP
+.BI "\-I " dir[,dir]
+Override list of ignored directories when recursing. If not given, it is
+taken from $FCOPY_IGNOREDIRS and as a last chance an internal list of known
+revision control system's metadata is used (CVS, .cvs, {arch}, .arch\-ids).
+.TP
.B \-l
Do not copy if destination is a symbolic link
.TP
@@ -101,7 +106,7 @@
.B \-r
Copy recursively (traverse down the tree). Copy all files below
SOURCE. These are all subdirectory leaves in the SOURCE tree. Ignore
-CVS directories.
+"ignored" directories (see "\-I" for details).
.TP
.BI "\-s " source_dir
Look for SOURCE files relative to \fIsource_dir\fR. Use $FAI/files if
--- orig/scripts/fcopy
+++ mod/scripts/fcopy
@@ -57,10 +57,12 @@
my $nobackup;
my $opt_update;
my $backupdir;
+my @ignoredirs = qw'CVS .svn .arch-ids {arch}';
# getopts:
our ($opt_s, $opt_t, $opt_r, $opt_m, $opt_M, $opt_v, $opt_d, $opt_D, $opt_i);
our ($opt_B, $opt_c, $opt_C, $opt_h, $opt_F, $opt_l, $opt_L, $opt_P, $opt_b);
+our $opt_I;
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
sub copy_one {
@@ -322,13 +324,14 @@
-F file Read list of sources from file.
-h Show summary of options.
-i Exit with 0 when no class applies.
+ -I dir[,dir] Override default list of ignored subdirectories
-l Do not copy if destination is a symbolic link.
-L file Log destination and used class to file
-m user,group,mode Set user, group and mode for copied files.
-M Same as -m root,root,0644
-P log,changes Copy if class or source for class has changed since
previous run
- -r Copy recursivly but skip CVS directories.
+ -r Copy recursivly but skip ignored directories.
-s source_dirLook for source files relative to source_dir.
-t target_dirCopy files relativ to target_dir.
-b backup_dirWhere to save backups of overwritten files
@@ -342,16 +345,21 @@
sub rfilter {
# Filter for recursive copying
- my $nlink;
- -d && ! ($_ =~ /^CVS$/) && ($nlink=(lstat($_))[3]) &&
-( (-d 'CVS') && ($nlink==3) || ($nlink==2) ) &&
- push @rlist,$File::Find::name;
+
+ # are we in a directory ? should we ignore it ?
+ my $location=$_;
+ (-d and (! grep $location eq $_,@ignoredirs )) or return 0;
+ # a directory without subdirs has two hard links
+ # don't count @ignoredirs as subdirs
+ my $subdirs=(lstat($_))[3] - 2 - grep(-d,@ignoredirs);
+ # push leaf
+ push @rlist,$File::Find::name unless $subdirs;
}
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
# main program
$|=1;
-getopts('Ms:t:rm:vidDc:C:hF:lL:P:Bb:');
+getopts('Ms:t:rm:vidDc:C:hF:lL:P:Bb:I:');
$opt_h && usage;
$opt_M and $opt_m="root,root,0644"; # set default modes
$opt_m && check_mopt();
@@ -380,6 +388,9 @@
$opt_C and @classes = read_classes($opt_C);
warn join ' ','Classes:',@classes,"\n" if $debug;
$opt_F and @ARGV = read_files($opt_F);
+$ENV{'FCOPY_IGNOREDIRS'} and @ignoredirs = split
/\s+/,$ENV{'FCOPY_IGNOREDIRS'};
+$opt_I and @ignoredirs = split /,/,$opt_I;
+
die "fcopy: source undefined\n" unless $source;
die "fcopy: target undefined\n" unless $target;
Summary: provide a more intelligent mechanism to ignore dirs in fcopy
Keywords: fcopy feature
- rewrite rfilter to check against a list @ignoredirs
- initialize @ignoredirs for config managed in CVS, subversion and GNU arch
- include command line option "-I blah[,blubb]" and environment variable
$FCOPY_IGNOREDIRS to override @ignoredirs
- comment the code and make it understandable (YEAH!)
provide a more intelligent mechanism to ignore dirs in fcopy
Re: fcopy and ignoring .svn directories when copying recursively
Hi,
On Fri, Nov 05, 2004 at 09:41:11AM -, [EMAIL PROTECTED] wrote:
> I've tried with the backslashed escape, but it has no effect. Is there
> anything else I can try?
> >> sub rfilter {
> >>
> >> # Filter for recursive copying
> >> my $nlink;
> >> -d && ! ($_ =~ /^.*CVS$/) && ($nlink=(lstat($_))[3]) &&
> >> ( (-d 'CVS') && ($nlink==3) || ($nlink==2) ) &&
> >> push @rlist,$File::Find::name;
> >> }
If (as I suppose) subversion internal information is stored in a .svn/
directory in every subdir of the tree, then the fifth line should be
replaced by
-d && ! ($_ =~ /^.*CVS$/) && ! ($_ =~ /^.*\.svn/) && ($nlink=(lstat($_))[3]) &&
(question to Thomas: shouldn't there be a \/ prepended to CVS - and .svn ?)
Cheers,
Steffen
--
Steffen Grunewald * * * Merlin cluster admin (http://pandora.aei.mpg.de)
Albert-Einstein-Institut (MPI Gravitationsphysik, http://www.aei.mpg.de)
Science Park Golm, Am Mühlenberg 1, 14476 Potsdam, Germany
e-mail: steffen.grunewald(*)aei.mpg.de * +49-331-567-{fon:7233,fax:7298}
Re: fcopy and ignoring .svn directories when copying recursively
Aplolgies, but I have checked the script that the test install client was
running and the fcopy script was not the amended one.
I am now testing re-install with the amended script and will keep the list
informed as to my results.
thanks
Stephen
>
> Hi Geert,
>
> Thank you for your response.
>
> I've tried with the backslashed escape, but it has no effect. Is there
> anything else I can try?
>
> (If this does not work, I will have to individually specify each file that
> I need to fcopy in a shell script with a for statement. This is messy and
> I would like to avoid this.)
>
> Here is some sample output from fcopy when it hits the .svn dirs:
>
>
> fcopy: no matching file for any class for root/.profile/.svn/tmp/wcprops
> defined.
> fcopy: no matching file for any class for root/.profile/.svn/tmp/props
> defined.
> fcopy: no matching file for any class for root/.profile/.svn/tmp/prop-base
> defined.
> fcopy: no matching file for any class for root/.profile/.svn/tmp/text-base
> defined.
>
> thanks
>
> Stephen
>
>> On Thu, Nov 04, 2004 at 01:46:46PM -, [EMAIL PROTECTED] wrote:
>>>
>>> Hi
>>>
>>> I'm using subversion to keep tabs of the /usr/local/share/fai
>>> directory,
>>> but have run into a problem when using fcopy recursively.
>>>
>>> I know that fcopy is set to ignore CVS files/dirs (it says so in the
>>> man
>>> page) so I tried to hack the perl from /usr/sbin/fcopy to accomodate
>>> .svn
>>> dirs. However, I can't seem to get the code to ignore .svn files! (My
>>> perl
>>> is not great.)
>>>
>>> Here is the code:
>>>
>>> sub rfilter {
>>>
>>> # Filter for recursive copying
>>> my $nlink;
>>> -d && ! ($_ =~ /^.*CVS$/) && ($nlink=(lstat($_))[3]) &&
>>> ( (-d 'CVS') && ($nlink==3) || ($nlink==2) ) &&
>>> push @rlist,$File::Find::name;
>>> }
>>>
>>> I have replaced CVS with '.'svn but when I run fcopy, my .svn files are
>>> not ignored and the fcopy fails.
>>>
>>> What am I doing wrong?
>>
>> try \.svn ( a dot is a wildcard, the backslash escapes it )
>>
>>> thanks in advance. (Danke!)
>>
>> Report your results back.
>>
>>>
>>> Stephen
>>
>> Geert Stappers
>>
>
>
Re: fcopy and ignoring .svn directories when copying recursively
Hi Geert,
Thank you for your response.
I've tried with the backslashed escape, but it has no effect. Is there
anything else I can try?
(If this does not work, I will have to individually specify each file that
I need to fcopy in a shell script with a for statement. This is messy and
I would like to avoid this.)
Here is some sample output from fcopy when it hits the .svn dirs:
fcopy: no matching file for any class for root/.profile/.svn/tmp/wcprops
defined.
fcopy: no matching file for any class for root/.profile/.svn/tmp/props
defined.
fcopy: no matching file for any class for root/.profile/.svn/tmp/prop-base
defined.
fcopy: no matching file for any class for root/.profile/.svn/tmp/text-base
defined.
thanks
Stephen
> On Thu, Nov 04, 2004 at 01:46:46PM -, [EMAIL PROTECTED] wrote:
>>
>> Hi
>>
>> I'm using subversion to keep tabs of the /usr/local/share/fai
>> directory,
>> but have run into a problem when using fcopy recursively.
>>
>> I know that fcopy is set to ignore CVS files/dirs (it says so in the man
>> page) so I tried to hack the perl from /usr/sbin/fcopy to accomodate
>> .svn
>> dirs. However, I can't seem to get the code to ignore .svn files! (My
>> perl
>> is not great.)
>>
>> Here is the code:
>>
>> sub rfilter {
>>
>> # Filter for recursive copying
>> my $nlink;
>> -d && ! ($_ =~ /^.*CVS$/) && ($nlink=(lstat($_))[3]) &&
>> ( (-d 'CVS') && ($nlink==3) || ($nlink==2) ) &&
>> push @rlist,$File::Find::name;
>> }
>>
>> I have replaced CVS with '.'svn but when I run fcopy, my .svn files are
>> not ignored and the fcopy fails.
>>
>> What am I doing wrong?
>
> try \.svn ( a dot is a wildcard, the backslash escapes it )
>
>> thanks in advance. (Danke!)
>
> Report your results back.
>
>>
>> Stephen
>
> Geert Stappers
>
Re: fcopy and ignoring .svn directories when copying recursively
On Thu, Nov 04, 2004 at 01:46:46PM -, [EMAIL PROTECTED] wrote:
>
> Hi
>
> I'm using subversion to keep tabs of the /usr/local/share/fai directory,
> but have run into a problem when using fcopy recursively.
>
> I know that fcopy is set to ignore CVS files/dirs (it says so in the man
> page) so I tried to hack the perl from /usr/sbin/fcopy to accomodate .svn
> dirs. However, I can't seem to get the code to ignore .svn files! (My perl
> is not great.)
>
> Here is the code:
>
> sub rfilter {
>
> # Filter for recursive copying
> my $nlink;
> -d && ! ($_ =~ /^.*CVS$/) && ($nlink=(lstat($_))[3]) &&
> ( (-d 'CVS') && ($nlink==3) || ($nlink==2) ) &&
> push @rlist,$File::Find::name;
> }
>
> I have replaced CVS with '.'svn but when I run fcopy, my .svn files are
> not ignored and the fcopy fails.
>
> What am I doing wrong?
try \.svn ( a dot is a wildcard, the backslash escapes it )
> thanks in advance. (Danke!)
Report your results back.
>
> Stephen
Geert Stappers
signature.asc
Description: Digital signature
fcopy and ignoring .svn directories when copying recursively
Hi
I'm using subversion to keep tabs of the /usr/local/share/fai directory,
but have run into a problem when using fcopy recursively.
I know that fcopy is set to ignore CVS files/dirs (it says so in the man
page) so I tried to hack the perl from /usr/sbin/fcopy to accomodate .svn
dirs. However, I can't seem to get the code to ignore .svn files! (My perl
is not great.)
Here is the code:
sub rfilter {
# Filter for recursive copying
my $nlink;
-d && ! ($_ =~ /^.*CVS$/) && ($nlink=(lstat($_))[3]) &&
( (-d 'CVS') && ($nlink==3) || ($nlink==2) ) &&
push @rlist,$File::Find::name;
}
I have replaced CVS with '.'svn but when I run fcopy, my .svn files are
not ignored and the fcopy fails.
What am I doing wrong?
thanks in advance. (Danke!)
Stephen
Re: fcopy: "file-modes" for directories?
Hello Holger install -m 0700 -d $target/some/dir ? fcopy /some/dir/somefile Sorry, I did not tell you the idea behind. I try to be more in line with Henning Glawe's patched fai: I call fcopy just once: fcopy -B -r / This is not really important now, but it might be easier in the future to convert the fai setup to maintain configurations of running systems. Cheers Harry
Re: fcopy: "file-modes" for directories?
> for directories instead of plain files? Is there a reason why this > would not make sense, or am I just the first one who could use this Maybe it makes sense, but maybe you can help yourself with this: install -m 0700 -d $target/some/dir ? fcopy /some/dir/somefile
fcopy: "file-modes" for directories?
As you know, fcopy has the capability to chown/chmod copied files using an additional file-modes file. Has anybody out there ever missed this for directories instead of plain files? Is there a reason why this would not make sense, or am I just the first one who could use this feature? Cheers Harry
Re: fcopy and classes?
>>>>> On Sat, 12 Jan 2002 20:15:40 +0100, Christoph Keller <[EMAIL PROTECTED]> >said: > Hello, is there any way to tell fcopy to use the file which > matches the class at the beginning? The Problem is that for > some host I need a different config file all other hosts have > one config. I set extra classes in the host file but in Why not using the hostname class ? The priority in FAI_CLASSES and $classes is from low to high. DEFAULT is the first class with the lowest priority. At the end the class with the hostname is of high priority. The class LAST has the highest priority but should not be used for fcopy. So put all your hosts into the class EXAMPLE2 and create a config file with the hostname and this will be used by fcopy. Also try fcopy -vD to get debug output. You can also influence the order of defined classes by creating new scripts in /fai/class/. -- Thomas
fcopy and classes?
Hello, is there any way to tell fcopy to use the file which matches the class at the beginning? The Problem is that for some host I need a different config file all other hosts have one config. I set extra classes in the host file but in FAI_CLASSES the class I definded in the host-file appears at the beginning. So for fcopy the priority for this file is less than that for all hosts. This is not what I want. Or do I use fcopy the wrong way? Christoph --
new command fcopy; first version
fcopy, a new command for copying file with classes is available. Using fcopy, all copy sections in the cfengine scripts will be obsolete. There's also a manual page available. Please look at it, and send me any comment on it. fcopy is available at: http://www.informatik.uni-koeln.de/fai/download It will be available in the CVS repository next week and in the next FAI version (not available the next week ;-). -- Thomas -- Thomas Lange Institut fuer Informatikmailto:[EMAIL PROTECTED] Universitaet zu Koeln Pohligstr. 1Telefon: +49 221 470 5303 50969 KoelnFax: +49 221 470 5317 1024D/AB9B66FD AEA6 A8C1 BD8E 67C4 8EF6 8BCA DC13 E54E AB9B 66FD --
