My plan is to switch to cfengine2 (be)for the etch release.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
I'm asking myself if this bug (of cfengine) also appears with
cfengine2? If not, we can close this bug, when FAI has switched to
cfengine2. Can anyone check cfengine2 ?
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL
IMO it's not that easy to integrate any of the d-i components. Some of
them depend on some special d-i infrastructure (like special libraries
that are only used in d-i or r/w access to some directories), that are
not availabe during a FAI installation. Also, the whole d-i is running
in a ramdisk
by default the command sndmon uses the variable $monserver. This is
the name of the machine to which the messages should be sent. If this
variable is not defined (mostly, because it's not set by default, but
you can define it yourself), it uses the value of $SERVER which is set
by get-boot-info().
On Tue, 13 Jun 2006 19:49:49 +0200, Enrique de la Torre Gordaliza [EMAIL
PROTECTED] said:
I have installed new packages built from svn and the problem continues.
Same
RPC error.
Please give me your boot.log.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with
On Wed, 14 Jun 2006 12:14:40 +0200, Enrique de la Torre Gordaliza [EMAIL
PROTECTED] said:
Installation stops at RPC error and I dont have shell access. How can I
read
the /tmp/boot.log or dmesg from the client? Any boot parameter?
This is an v20z
Yes, I know those machines.
To see which network card is used by the second dhcp call, please edit
the lib/get-boot-info script in your nfsroot. Locate the line:
dhclient -lf /dev/null -cf /etc/dhcp3/dhclient-fai.conf -sf
/etc/dhcp3/dhclient-fai-script $netdevices $bootlog 2 $LOGDIR/dhclient.log
and remove the
Maybe this helps:
http://www.science.uva.nl/research/air/wiki/LogicalInterfaceNames
Try ether=0,0,eth0 (or netdev= which is the same) as kernel parameter
(you can give it with fai-chboot).
You have to force the kernel not to activate the oTher NIC, which is not
defined in dhcpd.conf. The
IMO you can enter the network card BIOS, and disable the PXE
booting. Currently I do not know how to enter it (ecom uses
atl-ctrl-b, intel uses ctrl-s). THere you should may select boot from
local disk first. Maybe this disables the MBA (or PXE) bootig from
this NIC. So it should be possible to
I'm not sure if this bug is really RC. Please specify in more details
what you mean by saying the installation now fails. What is the error
message. Does the make-fai-nfsroot call fail or the installation
itself?
A solution may be to use the -cf and -sf (man dhclient3) flags for
specifying
sk98lin is enabled, but maybe the two others are missing for your hardware.
# CONFIG_SKGE is not set
# CONFIG_SKY2 is not set
CONFIG_SK98LIN=y
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Yes, IMO it's also usefull on amd64. I think all (or nearly all)
network drivers that we use on I386 are also useful on amd64.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
This bug does not affect make-fai-nfsroot but make-fai-bootfloppy. But
I can confirm this bug. It will be fixed soon.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Package: fai-server
Version: 2.10.5
Severity: minor
fai-setup should not set up rsh/ssh on install server if LOGSERVER if
not the install server. This can't deb determined by the hostname or
ip address, since it may be different on a server with multipli
networks cards (and different hostnames
Package: fai-kernels
Version: 1.11.1
Severity: wishlist
If you want to use the iproute command the kernel needs additional
modules. From the iproute packages decription:
At least, the options CONFIG_NETLINK and CONFIG_NETLINK_DEV (or
CONFIG_RTNETLINK) must be compiled into the running
Package: fai
Version: 2.10.5
Severity: normal
Tag: patch
setup_harddisks does not preserve partitions with no mount point.
From: Alexius Ludeman [EMAIL PROTECTED]
when the partition config read:
primary - 9000- preserve4; preserve
mke2fs would still execute.
I'm not sure if the patch really solves this problem. Maybe it
preserved ALL partitions without a mount point. Someone has to test this.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Additional log file:
format
setup_harddisks version 0.38fai
Probing disks: /dev/sda /dev/sdb
Disks found: sda sdb
Using config file: /fai/disk_config/2_DISKCOUNT
Mapping disk name disk1 to sda
Mapping disk name disk2 to sdb
Creating partition table: LC_ALL=C sfdisk -q /dev/sda
Package: fai-kernels
Version: 1.11
Severity: wishlist
Please include the driver for this hardware SCSI device:
CONFIG_SCSI_DPT_I2O
I think also more SATA driver should be added. I propose these:
CONFIG_SCSI_SATA_SX4
CONFIG_SCSI_SATA_SIL
CONFIG_SCSI_SATA_SIL24
CONFIG_SCSI_SATA_SIS
--
regards
Package: fai-kernels
Version: 1.10.3
Severity: wishlist
Please include the driver for the Nvida onbard network card.
Please add this option.
CONFIG_FORCEDETH=y
-- System Information:
Debian Release: 3.1
Architecture: i386 (i686)
Kernel: Linux 2.6.8-3-686
Locale: LANG=C, LC_CTYPE=en_US
/boot/fai/vmlinuz-install needs to be copied by make-fai-nfsroot in
the case you install a new kernel to the nfsroot. The other dirs and
the pxelinux.0 file could be written by fai-setup.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble?
On Wed, 12 Jul 2006 18:40:35 +0200, Holger Levsen [EMAIL PROTECTED]
said:
an hour ago fai-kernels 1.11 was uploaded to the archive. I didn't enable
SATA_SIS for i386 (amd64 has it already) because I didn't see the bug in
the
Vielleicht noch einmal das i386 config mit dem amd64
This bug is not grave and is not the fault of FAI. It's a kernel
problem, that the kernel has a different ordering of network devices,
than the BIOS. Also it only occurs on a certain hardware. I will set
the serverity to normal.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
Package: fai-client
Version: 2.9.1
severity: wishlist
fcopy should have a flag (-n, like dry-run or simulate) for showing
what it would do, without really doing it. This will help many
sysadmin to see if something has to be changed on their system.
--
regards Thomas
--
To UNSUBSCRIBE,
Package: fai-client
Version: 2.9.1
severity: wishlist
I like to remove flag -P from the command fcopy. Is somebody really
using it?
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Package: fai-server
Version: 2.9.1
severity: wishlist
fai-chboot needs a flag for specifying an initramfs image for the
kernel. This will be needed for future install kernels.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact
On Mon, 20 Mar 2006 09:48:57 +0100, Björn Torkelsson [EMAIL PROTECTED]
said:
Left to do is to extend fai-chboot to create the template files (should
they end in .tmpl? Right now we don't use and file extension), but that
shouldn't be that hard. Currently we are creating the
Hi Petter,
I will apply this patch in the next FAI version, which I think will be
uploaded in april. So there's no need for an NMU, I will care about
this bug.
--
regards Thomas
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
On Wed, 25 Oct 2006 17:59:47 +0200, Thomas =?UTF-8?Q?P=C3=B6hnitzsch?=
[EMAIL PROTECTED] said:
+} elsif ($hex =~ /[0-9A-F]+/) {
+my $hexstrip = (split /\./,$hex)[0]; # remove .disable if
necessary
+$n = $hexstrip;
+# hex to ip/subnet address
Again, please provide the output of fcopy -vD for an examples which
causes the error.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Severity serious is only used for policy violations. This bug is grave
or important, but anyway I will fixed is ASAP.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
I like to raise the severity of this bug, since is may also be used for
a DOS attack. A small script that does a lot of calls to rstatd (for
e.g. using rup), you can fill up the log files in /var/log.
IMO this is also very useless data, that is only needed for
debugging. If you do not want to
Hi Lucas,
I will adopt the tcsh package. I still have to learn git, but I think
the tcsh package is a nice first example for learning it. Can you give
me write access to the git repository?
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject
Package: fai-server
Version: 3.4.5
Severity: important
Calling fai-mirror is broken. This is because of the fix for #600715.
Attached is a detailed description.
---BeginMessage---
(explicitly CC'ed Thomas to alert him.)
Hi all,
after update to 3.4.5 I get an error when running fai-mirror
What do you think about this fix?
-print $0: reading config files from directory $classpath\n;
+warn $0: reading config files from directory $classpath\n if $verbose;
It prints the message to stderr as before (before we fixed #600715),
but it only prints this message if verbose (-v) was set.
: #592093
* add tcsh-6.17.02-multibyte.patch Closes: #603545
-- Thomas Lange la...@debian.org Mon, 06 Dec 2010 17:41:17 +0100
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Mmm, my upload was rejected because of a size mismatch. I have to
check why this happened.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
I fixed the problem and now the upload was successfully.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
The attached hook implements this feature. The new variable
FAI_BASEFILEURL is used to download a base.tgz (or .tar.xz,...) via
HTTP or FTP.
If you set for e.g. FAI_BASEFILEURL=http://faiserver/basefiles this
hook will get a listing of all files in the directory basefiles on the
HTTP server
Severity: important
I think it's a normal bug.
Removing the diversion of /etc/init.d/nis from /usr/lib/fai/updatebase
fixes
the problem.
I have to check if the diversion is not needed any more, or if fai
should use policy-rc.d to keep nis from starting during installation.
I think a lot of FAI bugs break unrelated software, because most
functions in FAI are related to other software. But that does not
automatically means the bug is important. Otherwiese mostly all dpkg or
aptitude bugs must also be important, because they cause unrelated
package not to be installed
The diversion of the nis ninit.d script can be removed. Tested with
lenny and squeeze. Patch available in svn commit 6282.
@mika: Please consider this patch for inclusion into the next stable
release, maybe after raising the severity again.
--
regards Thomas
--
To UNSUBSCRIBE, email to
Package: www.debian.org
Severity: normal
Tags: Patch
On the web page http://www.debian.org/events/speakers/ there's a link
to the old FAI project web page
(http://www.informatik.uni-koeln.de/fai/)
Please update this link to it's new location:
http://fai-project.org
The link to my talks is also
Package: fai-server
Severity: wishlist
Please add a new option to fai-chboot this add the flag initial to
FAI_FLAGS. This flag ist used by setup-storage and influences the
creation of partitions marked as preserve_reinstall.
--
regards Thomas
--
To UNSUBSCRIBE, email to
I realized that after the rinse command is executed the nis domainname
is unset (or set to none). The ypbind processes are still running.
# domainname
(none)
Setting the domainname back to its old value let nis running again.
So the problem seems that authconfig only clears the domainname.
--
Here's a small code snippet that will work.
It only matches variable names having chars, numbers or an underscore.
shopt -s extglob
case $word in
*([a-zA-Z0-9_])=*) echo matched $word ;;
esac
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
On Wed, 19 Jan 2011 10:37:48 +, Michael Tautschnig m...@debian.org
said:
I just fixed most of the above this morning in trunk, with any version =
r6270
you should be fine. Well, honestly, the perl-tk isn't there yet, thanks
for
adding it. Should get fixed soon.
svn r6271
To be more spcific. perl-tk is not needed for building the fai
package. It's only needed it you like to check the faimond-gui perl
script for syntax errors. I'm not sure if we really should add a
build-depends on perl-tk.
BTW, adding liblinux-lvm-perl to build-depends may cause lvm2 to be
On Wed, 19 Jan 2011 10:57:50 +, Michael Tautschnig m...@debian.org
said:
If we prefer not to have that build dependency but still keep the syntax
checks
we could also add dummy files (which is what I did on alioth to make the
experimental versions build).
What if we remove
Package: fai
Version: 3.2.14
Severity: wishlist
A package P may not depend on the package locales, but will only
configure itself for locales already available in the system.
This may cause problems, when the package locales is installed after
the package P. Then this package may fall back to
It's not a good idea to replace the logical OR with AND. But what
about supporrting both? This is an idea for a format in
package_config.
logical OR as before:
PACKAGES aptitude CLASSA CLASSB
new logical AND:
PACKAGES aptitude +AND CLASSA CLASSB
--
regards Thomas
--
To UNSUBSCRIBE,
Maybe this is related to the thread below. debootstrap seems to exit with
code 141 in some cases. Can you please check if this applies to your
environment? There's also a proposal for a patch included in the thread.
http://www.mail-archive.com/linux-...@uni-koeln.de/msg01453.html
--
regards
If the red banner says 3.2.11, you are running an old version because
you did not properly rebuild your nfsroot. Maybe you install server
and the nfsroot are using different source.list files.
Please check again your nfsroot. The banner _must_ say 3.2.13.
--
regards Thomas
--
To UNSUBSCRIBE,
Package: fai
Version: 3.2.14
Severity: wishlist
I like to have support for a file, which includes the already
downloaded APT keys (proposal: APT_KEY_FILE)
This list of gpg keys will be added to the APT key management via the
command apt-key. So, all packages signed with those keys are considered
Package: fai-client
Severity: minor
setup-storage creates following line for the swap partitions:
/dev/sda5 noneswaprw 0 2
But it should be 0 0 at the end.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
Package: grub
Version: 0.97-47lenny2
Severity: normal
When I use grub-install for a chroot environment (during an
installation with FAI) I get following error:
# /usr/sbin/grub-install --no-floppy --root-directory=/target /dev/sda
grub-probe: error: Cannot open `/boot/grub/device.map'
This bug was introduced in 0.97-47lenny2 and is located in the file
xfs_freeze.diff
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
On Wed, 28 Jan 2009 11:09:33 +0100, Sebastian Harl s...@tokkee.org said:
Imho faimond should check for
missing modules and print a warning. That warning could also mention
which packages have to be installed.
I like this solution. We have a similar check in install_packages (sub
This bug will be fixed in the next version, since I removed the yes pipe.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
In make-fai-nfsroot the variable NFSROOT is set to
NFSROOT=$NFSROOT/live/filesystem.dir
So it should have been set to the value you expect. I think I've
forgotten to export this variable, so the hooks also can use this
value. Do you think adding the export command in front of this
definition
This patch will not work, since setting HOST= on the kernel command
line does not do anything (maybe since FAI 3.2). IIRC in older version of
FAI this worked, but the code was removed.
So there's no need to patch fai-chboot.
--
regards Thomas
--
To UNSUBSCRIBE, email to
Back to the first post. It's wrong to use an unmodified fai.conf and
try fai softupdate.
FAI_CONFIGDIR is only used on the install server. It's not used by an
install client. Maybe the comment is a little bit confusing. So this
variable should be moved to make-fai-nfsroot.conf to make things
Package: fai-server
Severity: wishlist
Since FAI 2.10 .deb files in /files/packages are not suported any
more (by using $addpackages). But the variable is still used in
fai-mirror. Please remove it and the corresponding code.
--
regards Thomas
--
To UNSUBSCRIBE, email to
Currently I will not include git-core into NFSROOT in the default
configuration, since it needs 10MB additional disk space. The fai-cd
ISO image may become to big.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble?
The next FAI version after 3.2.20 will have a new flag -s/--cspace so
you can specify a value for FAI_CONFIG_SRC.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Since setup_harddisks is not maintained any more and it's replaced by
setup-storage, this patch will not be included. Please file a new bug
report if you like to have this feature for setup-storage.
I will close the bug soon.
--
regards Thomas
--
To UNSUBSCRIBE, email to
The command fai-class(1) now also uses the variable ADDCLASSES for
defining classes. So you can add this to the kernel command line to
define some classes:
ADDCLASSES=MYCLASS1,FOOBAR,ANOTHERCLASS
For more information read the man page of fai-class(1).
--
regards Thomas
--
To UNSUBSCRIBE,
Thanks. It's a good idea to switch from inetd mode to standalone tftp
daemon. I will will fix then also the dependencies.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
fai-quickstarts depends on some version of inetd, because the default
setup of an install server should also provide the time protocol,
which is used by the clients to set the clock.
I will leave this dependency and closes the bug soon.
--
regards Thomas
--
To UNSUBSCRIBE, email to
The package was orphaned over a year ago. Martin worked on it, but
still did not produced a package that could be uploaded. It also was
only prepared for etch, never for lenny nor squeeze.
The last mail on this topic is more than 6 month ago. IMO this
package should be removed from the debian
Tags: fixed-in-experimental
This bug is now fixed in the new package version 6.17.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
It seems that tcsh 6.17 has fixed this issue.
The function NLSStringWidth is now paying attention to the return
value of wcwidth:
if ((l = wcwidth(c)) 0)
Please test, if your issue is solved in version 6.17 of tcsh.
--
regards Thomas
--
To UNSUBSCRIBE, email to
Tags: fixed-in-experimental
Version: 6.17
This bug is fixed in version 6.17
veedel[~/[1]] pwd
/home/lange/[1]
veedel[~/[1]] cd -
veedel[~] echo $version
tcsh 6.17.00 (Astron) 2009-07-10 (i486-intel-linux) options
wide,nls,dl,al,kan,rh,nd,color,filec
--
regards Thomas
--
To UNSUBSCRIBE,
Package: texlive-binaries
Version: 2009-5
Severity: minor
If I install software inside a chroot (squeeze) environment, sometimes
kpsewhich runs into an endless loop and hangs. This is some part of
the output:
Processing triggers for tex-common ...
Running mktexlsr. This may take some time...
Since version 3.1.2 mirgrated to testing, I think this bug can be closed.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
There's a new project called tdpkg which can be used together with dpkg.
The tdpkg shared library is used to speed up dpkg .list files loading
using either tokyocabinet or sqlite3.
http://lethalman.hostei.com/tdpkg.html
--
regards Thomas
--
To UNSUBSCRIBE, email to
Package: newt
Severity: wishlist
There's a new upstream version available at
https://fedorahosted.org/releases/n/e/newt/
newt-0.52.11.tar.gz 24-Sep-2009 15:04 171K GZIP compressed
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
Package: fai-client
Severity: wishlist
People like to use setup-storage without calling it from FAI.
This is how it should work
- install the fai-client package
- export PATH=/usr/lib/fai:${PATH}
- export disklist=$(disk-info | sort)
- setup-storage -X -f yourconfigfile
--
regards Thomas
Package: fai-client
Severity: critical
Tags: security
Tags: pending
When using fai softupdate, install_packages writes a list of all
packages to the file /var/tmp/package, which is located in a world
writeable directory. It also writes to /tmp/packages.list if
FAI_DEBSOURCESDIR is set. These
On Tue, 30 Mar 2010 21:52:27 +0100, Holger Levsen hol...@layer-acht.org
said:
How? (Seriously, how would you do this?)
I, myself will not use subclasses.
definitly be easily possible in FAI, thus either be part of the
featureset or be documented well.
Have a look at #498412, there's a
On Wed, 31 Mar 2010 10:18:25 +0200, Michael Tautschnig m...@debian.org
said:
Would you mind explaining how this could possibly be exploited? There is
A user could create a symlink pointing to a file (e.g. /etc/passwd)
which will then be overwritten when root calls fai softupdate.
Maybe
I'm not quite sure whether this change is intentional but in the current
version of fai-client, disk-info does not output the disk sizes anymore.
This change was intentionally, since the disk size is not needed by
FAI any more. If you need the size in some of your scripts, you have
to
On Tue, 30 Mar 2010 21:36:05 +0200, Gerfried Fuchs rho...@deb.at said:
If they only affect FAI versions from 3.3 - shouldn't it be versioned
like that? In case I missed something feel free to revert, but without a
found version this looks like affecting even lenny.
The last part
On Wed, 31 Mar 2010 22:43:41 +0200, Michael Prokop m...@debian.org said:
The following patch works for me and implements the new option
-B path_to_base.tgz within make-fai-nfsroot to be able to
run something like make-fai-nfsroot -v -B /srv/base.tgz:
On Thu, 17 Dec 2009 16:36:11 +0100, root rudy.geva...@ugent.be said:
I just retile your bug report, so now it has a title :-)
Thanks for reporting the bug.
I used:
bts retitle #561489 Mising support for XEN devices like /dev/xvda
--
regards Thomas
--
To UNSUBSCRIBE, email to
Please provide more information how zypper has to be called when
installing a list of packages.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Just for the record.
This bug is only valid for squeeze and later, not for lenny.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
ssh r...@faiclient faireboot
works perfectly for me with Debian and FAI 3.2.17~lenny1
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
No, AFAIK syslinux is not supported. The syslinux package is only
needed for pxelinux.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
My idea: the fai script should set it to 700 if it's not already defined in
fai.conf.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
One part of the patch in experimental (which replaces ifconfig with ip
calls) removes the variable $mask, so that now the NFSROOT is only
exported to the IP address of the server. I'm pretty sure this will
break things. Or did I overlooked something?
--
regards Thomas
--
To UNSUBSCRIBE,
On Thu, 9 Apr 2009 19:30:53 +0200, Michael Tautschnig m...@debian.org
said:
ip addr includes the mask in the output, so it would be there rightaway,
like
this:
Yep, if if also contains the mask it will work. Seems fine to me, but
I have to check some other things before including
IMO the variable debug should only be set to 0 or 1 or may be
undefined. A test, if this variable is non-numeric is not needed.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Maybe I can limit the execution of more than one dirinstall to the
same directory, or I will just remove the check when doing a
dirinstall and only test if fai is doing softupdates.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
On Sat, 18 Apr 2009 10:06:01 +0200, Michael Tautschnig m...@debian.org
said:
Why not use
set -e
. /etc/fai/fai.conf
. /etc/fai/make-fai-nfsroot.conf
set +e
This is already done in fai-setup and make-fai-nfsroot.
--
regards Thomas
--
To UNSUBSCRIBE, email to
Package: fai-doc
Version: 3.4.4
In the simple examples the script GRUB_PC/10-setup always install grub
into hd0. If your boot disk is not hd0, the host does not boot.
I have following configuration:
four SATA disks, three times 1TB, one 80GB disk. This 80GB should be
my boot disk. The three
Here's the patch:
Index: 10-setup
===
--- 10-setup(revision 875)
+++ 10-setup(working copy)
@@ -5,6 +5,8 @@
$ROOTCMD grub-mkdevicemap -n -m /boot/grub/device.map
$ROOTCMD grub-mkconfig -o /boot/grub/grub.cfg
-$ROOTCMD
Package: fai-doc
Severity: important
I think this bug is at least important. Maybe it can causes data loss,
if grub installs itself into hd0, no matter if this disk is used in a RAID.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of
The script device2grub is not needed for grub2. In FAI we have the
class GRUB_PC which uses the script GRUB_PC/10-setup for setting up
grub2 configuration.
Also have a look at #604938 which fixes a problem in this script. FAI
will use grub-probe (instead of device2grub) for grub2.
IMO we can
- $BOOT_DEVICE may consist of several device names if RAID is used. Hence
we
need to loop over each of them and do what you just proposed.
That's a good reason to use your patch.
--
regards Thomas
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a
201 - 300 of 1142 matches
Mail list logo
