we followed this wiki entry to install both amd-64 and i386 from the same fai
install. It works well for us.
http://wiki.fai-project.org/wiki/FAI_Cross-architecture
David Dreezer
Social Strata, Inc.
On Sep 11, 2013, at 11:27 AM, linux-service.be bvba
wrote:
> In order to install ol
Hi,
You might want to add -v -v -v -v to the TFTP_OPTIONS. This will get the max
level of verbosity into your syslog. That will help you to see exactly what
files the client is asking for. Once you know what is being asked for then you
can perhaps better determine why it cannot find the file.
,noatime
raid1 /usrdisk1.9,disk2.9 ext3rw,noatime
raid1 /appdisk1.10,disk2.10 ext3rw,noatime
This is a working config on our 4.0 install for MD raid. Perhaps it gives you a
place to start from?
David Dreezer
>
> We couldn't see an updated manual for FAI 4.0
You could echo that line into /etc/network/interfaces in
/usr/sbin/make-fai-nfsroot or put it into /usr/sbin/fai-setup after
make-fai-nfsroot, couldn't you?
That way any time you rebuild NFSROOT it will add the line for you.
>
> We tend to rebuild our NFSROOT occasionally to keep it patched wi
or not as it has been a long while since we built our FAI install.
If it is not I can provide the for you to edit as you need.
David Dreezer
On Feb 24, 2012, at 8:13 AM, Jeffrey Stolte wrote:
> I would like to add a line like the following to /etc/network/interfaces
> in the FAI NFSROO
would do this for me, but that I
am not seeing?
Am I the only one that has faced a need to make a kernel flag permanent?
Thanks,
David Dreezer
Customer Advocate, Social Strata, Inc.
.
Thanks again!
David Dreezer
Customer Advocate, Social Strata, Inc.
On May 12, 2011, at 4:28 PM, Michael Tautschnig wrote:
> Hi David,
>
> Sorry, I had completely forgotten about this issue. Have you possibly already
> fixed it in the time being? To be honest, I don't have a r
set and I'm not
sure how to go about doing that with Grub2 and FAI. So, while I do have a
server that is now properly installing while using the NTFS kernel and the FAI
flags, I don't yet have a computer that will run on it's own.
But getting the array working is a giant step in the r
m_startall_examined") if (scalar(keys
%FAI::current_raid_config));
297 foreach my $id (keys %FAI::current_raid_config) {
298 my $md = "/dev/md$id";
Thank you for your help on this.
David Dreezer
On May 2, 2011, at 7:43 AM, Michael Tautschnig wrote:
>> Hi Mic
ram => "mdadm --assemble --scan
--config=$FAI::DATADIR/mdadm-from-examine.conf",
I'd guess I'll want to edit both at some point? It's Monday AM, I haven't time
yet this morning to walk through the code and figure which one to edit :)
Haven't yet made it through my first cup of coffee :)
Thanks,
David Dreezer
/dev/md5 is already in use.
mdadm: /dev/md6 is already in use.
mdadm: /dev/md7 is already in use.
mdadm: /dev/md9 is already in use.
mdadm: /dev/md10 is already in use.
mdadm: /dev/md11 is already in use.
Thanks!
David Dreezer
>
> Sorry for getting back to this one very late. If you stil
all the
info that I can
David Dreezer
format-logs.c-fs1qa.tar.gz
Description: GNU Zip compressed data
On Apr 15, 2011, at 6:09 PM, David Dreezer wrote:
>
>>
>> Well, everything should be (more or less) fine here, as long as the kernel
>> actually makes use of this. The e
>
> Well, everything should be (more or less) fine here, as long as the kernel
> actually makes use of this. The error messages are simply an oddity of the
> attempt to export variables here, see
> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=607285. But this shouldn't
> pose any problems to
in SVN and will be part of
> 4.0~beta2+experimental83,
> which was just built.
Excellent. I really appreciate this. Although it's a little thing but it will
save me a lot of time.
David Dreezer
FLAGS=verbose,sshd,createvt FAI_ACTION=install SERVER=mg006
hostname=c-fs1qa.qa01.groupee-inc.net
David Dreezer
On Apr 15, 2011, at 12:40 PM, Michael Tautschnig wrote:
> Hi David,
>
>> The machine came to life eventually. I've attached the system messages as I
>
The machine came to life eventually. I've attached the system messages as I
think it may be helpful
David Dreezer
messages.txt.gz
Description: GNU Zip compressed data
On Apr 7, 2011, at 10:39 AM, David Dreezer wrote:
> Hi Michael,
>
> Interesting.. when I do mdadm --stop -
ping the machine.
I'll make a trip to the data center and see if there is anything on the console
that offers a clue as to the issue that we're seeing. Perhaps I have a faulty
piece of hardware or something?
Thanks,
David Dreezer
On Apr 7, 2011, at 2:32 AM, Michael Tautschnig wr
some insight?
Thank you,
David Dreezer
Customer Advocate, Social Strata, Inc.
formatlog.tar.gz
Description: GNU Zip compressed data
>
> A proper patch would indeed be preferred, but I would have been ok with
> merging
> this one as well if ...
Will do, once we have syntax that we can agree upon for the patch
>
>> # add any 3rd party repository keys that you may need installed into the
>> NFSROOT
>>
>> if [ -d $cfdir/apt/k
>
> Sorry for the earlier nit-picking and then taking so long to finally include
> it.
> But here it is: it's part of 4.0~beta2+experimental70. I suppose you want to
> test whether this final version indeed does the trick.
>
> Best regards,
> Michael
>
Hi Michael,
I'm sorry for the delay. I
Hi Michael,
>
> First of all, obviously it makes almost always sense to share at least the
> ideas, if not the implementation with others. And that's obviously much easier
> if it is integrated into FAI itself. Yet I think two questions are to be
> answered:
>
> - Is this change useful for anybo
t I'm left with is - when is
it best to write something as a hook and when is it best to submit it as a
patch against FAI itself?
As always, thanks for your thoughts.
David Dreezer
On Mar 27, 2011, at 12:20 PM, Michael Tautschnig wrote:
> Hi David,
>
>> I'm starting to dou
OOTCMD apt-key add -"
done;
And for the man page
/etc/fai/apt/sources.list
Copy this file to the NFSROOT.
+ /etc/fai/apt/keys-added/*.asc
+ Repository keys in .asc format found here will be installed into the
NFSROOT
/var/run/fai/make-fai-nfsroot/*-stamp
Stamp files used for recovery.
David Dreezer
>
> Hmm, I thought the point of keys-added-by-m-f-n or whatever you prefer to call
> it would be in using that *inside* the NFSROOT, i.e., doing the cp that you
> previously suggested to such a directory!?
>
> Best,
> Michael
>
I misunderstood your request, entirely. It'll be an easy fix.
> Su
>
> Oh, right, sorry, I had completely missed that $cfdir wasn't actually a config
> space. So my only concern that remains is: wouldn't the sed command possibly
> remove too much of the string?
You're right, I was being redundant on the next line, with the cp. I had sed
strip the path, then I
David Dreezer
>
> That sounds good, but I'd still have a few comments which I believe could even
> improve the code:
>
> - I think find $cfdir -type f -iname *.asc (or maybe even just -name) would be
> safer;
Fair point. I chose i to be more friendly, but isn&
Does this look closer to what you're looking for Michael?
for f in `find $cfdir -iname *.asc | sed 's!.*/!!'`; do
cp -Lp $cfdir/$f $NFSROOT/etc/apt/$f
$ROOTCMD apt-key add /etc/apt/$f
done;
If so I'll create the patch and post it.
David Dreezer
>>>
ested partition
(hd0,msdos8,msdos1).
Thank you so much for your efforts!
David Dreezer
>>
>
> Cool! As such, could you please give 4.0~beta2+experimental69 another try?
> That
> should switch back to a more proper gap between logical partitions. It seems
> that the Linux kernel
>
>
>
>
> parted -s /dev/sdb mklabel msdos
> parted -s /dev/sdb mkpart primary "" 32256B 134249471B
> parted -s /dev/sdb mkpart extended "" 134249472B 60022480895B
> parted -s /dev/sdb mkpart logical "" 144249984B 2281724927B
>
> following without another reboot>
>
> parted -s /dev/sdb mkla
root@fw1dv:~# parted -s /dev/sdb mkpart logical "" 144249984B 2281724927B
Error: You requested a partition from 144MB to 2282MB.
The closest location we can manage is 2282MB to 2282MB.
root@fw1dv:~# parted -s /dev/sdb mkpart logical "" 134250496B 2281724927B
Error: You requested a partition from 1
Same answer, unfortunately.
root@fw1dv:~# parted -s /dev/sdb mklabel msdos; sleep 10;parted -s /dev/sdb
mkpart primary "" 32256B 134249471B; sleep 10;parted -s /dev/sdb mkpart
extended "" 134249472B 60022480895B; sleep 10; parted -s /dev/sdb mkpart
logical "" 134249984B 2281724927B
Warning: The
I did add exit 1; to the top of setup storage, as requested. That made setup
storage exit and then I ran the commands manually as requested. Creating the
logical drive caused the kernel notification error.
Actually, is that the wrong behavior? It isn't what we want, but is it doing
what it isn'
We have a winner, and the winner is parted :)
On Mar 16, 2011, at 7:51 PM, Michael Tautschnig wrote:
>
>
> - add some exit 1; to usr/sbin/setup-storage in the first lines to make it
> abort
> right away to get setup-storage out of the game.
> - manually execute the following commands in the
> Ok, fine, at least the commands are getting executed again. One basic
> question:
> has this system + this configuration ever worked with setup-storage? If yes,
> any
> clue which version that was. Although there is no archive of experimental
> versions its possible to reproduce an earlier sta
I now have
293 # check RAID arrays if there are pre-existing ones
294 &FAI::push_command("true",
295 "", "mdadm_startall_examined") if (scalar(keys
%FAI::current_raid_config));
296 foreach my $id (keys %FAI::current_raid_config) {
297 my $md = "/dev/md$id";
>
> But well, I suppose you are rebooting the system for each run,
Right
> so we can't have
> any old processes hanging around. Then it *must* be the RAID doing stupid
> things. May I ask you to perform the following changes?
>
> - comment out the line in /usr/sbin/setup-storage that does the mda
[..]&FAI::push_command( "mdadm -W --stop $md ; sleep 2 ; cat /proc/mdstat",
"$pre_deps_cl", "self_cleared_$md");
> i.e., add the wait, change the sleep to a smaller value (to speed things up),
> and print /proc/mdstat? I have absolutely no idea why the kernel claims some
> device is still busy, a
::push_command( "mdadm --stop $md ; sleep 5", "$pre_deps_cl",
"self_cleared_$md");
Which still failed. Format.log and mdadm-from-examine.conf tarred and attached.
David Dreezer
Customer Advocate, Social Strata, Inc.
>
> Let's just hope that David will have the
David Dreezer
Customer Advocate, Social Strata, Inc.
>
> [...]
>
> mdadm --examine --scan --verbose -c partitions
root@fw1dv:~# mdadm --examine --scan --verbose -c partition
ARRAY /dev/md/0 level=raid1 metadata=1.2 num-devices=2
UUID=3b0ecb6c:9462a669:87bd52cd:a3bd3bcd nam
My apologies. This truly is the format.log from the last attempt. I don't see
a return from the new print line, but you'll see the initialized value errors.
David Dreezer
>>
>
> Such a warning would make a lot of sense, but I can't seem to see it in the
> file
&g
Oh gosh, I emailed the wrong format.log...
Just a sec.
David Dreezer
Customer Advocate, Social Strata, Inc.
>>
>
> Such a warning would make a lot of sense, but I can't seem to see it in the
> file
> you attached!? And why is that debug output not in the log
lized on lines 299-302, but I
didn't have time to work though those lines and see why it fails.
David Dreezer
>
> May I ask you to add some debug output before that line, which is
>
> print "Size of current_raid_config: " . scalar(keys
> %FAI::current_raid_config)
6
automatically prepare storage devices
David Dreezer
Customer Advocate, Social Strata, Inc.
>
>
> That's weird! Could you please check whether you can find mdadm-startall in
> /usr/share/fai/setup-storage/Commands.pm? Otherwise you're not using the code
> I'
Different, but still didn't make it all the way through. Log attached as it is
only 5k.
Thank you for all of the work and time that you are putting into this.
David Dreezer
Customer Advocate, Social Strata, Inc.
>>
> [...]
>
> I don't know whether Fredrik Eriksson
(started [2/2]).
Generating udev events for MD arrays...done.
root@fw1dv:~# ls -l /dev/md6
brw-rw 1 root disk 9, 6 Mar 15 17:00 /dev/md6
root@fw1dv:~# ls -l /dev/md/6
lrwxrwxrwx 1 root root 6 Mar 15 17:00 /dev/md/6 -> ../md6
root@fw1dv:~#
David Dreezer
Customer Advocate, Social Strata,
/md6: No such file or directory
root@fw1dv:~#
I'll try the new build today.
David Dreezer
Customer Advocate, Social Strata, Inc.
>
>
> As I might only be able to fix that tomorrow, could you meanwhile help me
> assess
> that my diagnosis of the problem is correct? That is, c
Still broken, unfortunately. The good news is that this time the format.log was
complete. I've attached it for you, Michael, gzipped this time.
David Dreezer
Customer Advocate, Social Strata, Inc.
>
> Sorry for breaking it, that should now be fixed again. Could you give
It does, on hindsight. I probably should have used a grep of the directory for
any *.asc file and use any it finds.
Testing ~62 this afternoon. If there is time left I'll clean that patch up for
you.
David Dreezer
Customer Advocate, Social Strata, Inc.
>>
> I'd be happ
Unfortunately we're broken again in 61. see attached, but I'm not sure there is
much to go on in the log.
David Dreezer
Customer Advocate, Social Strata, Inc.
On Mar 9, 2011, at 5:24 PM, Michael Tautschnig wrote:
>
> Things should also work fine if you upgrade to 4.0~beta
your feedback Michael!
David Dreezer
Customer Advocate, Social Strata, Inc.
make-fai-nfsroot-v2.patch
Description: Binary data
ts.
make-fai-nfsroot.patch
Description: Binary data
David Dreezer
Customer Advocate, Social Strata, Inc.
the best way to get that from the client to the server
and then make the file name change.
Thoughts? Am I headed the right way? Making it too difficult?
David Dreezer
Customer Advocate, Social Strata, Inc.
>>
>
> Sorry, it's not just different formats; only 01-... is the
asking for
Mar 8 17:42:30 mg006 in.tftpd[9529]: RRQ from 10.7.4.91 filename
fai/pxelinux.cfg/01-00-30-48-72-64-64
how is it going to work if my file is named fai/pxelinux.cfg/01003048726464?
I don't think that it will. The file on my tftp server needs to match what the
computer is asking for. Yes?
Thanks,
David Dreezer
al option.
Anything else is less usable, I believe.
David Dreezer
Customer Advocate, Social Strata, Inc.
Hi,
Just to confirm, this patch did work and the server installed as hoped. Thanks!
David Dreezer
Customer Advocate, Social Strata, Inc.
On Mar 8, 2011, at 11:59 AM, Michael Tautschnig wrote:
> Hi again,
>
>
> Never mind, I've been able to reproduce this myself and reve
2hex is my issue, or if it is not being given the
$host by task_chboot that I need it to have for this to work properly. Or, if
I'm totally on the wrong track :)
Any advice would be appreciated
Thank you
David Dreezer
Customer Advocate, Social Strata, Inc.
s that the b drive didn't get smaller since last
week. :)
Any ideas?
David Dreezer
Customer Advocate, Social Strata, Inc.
t
nfs-kernel-server. If you need to make any adjustments to
make-fai-nfsroot.conf be sure to run fai-setup again.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
On Jan 21, 2011, at 2:14 AM
You'll receive that error if your /etc/fai/apt/soures.list is incorrect.
Also, if you are using your own mirror, and you have not updated it in a while,
you will receive exactly that error if the resources files are out of date. If
that is the issue you'll nee dot update your mirror
the client is
doing and where the server is trying to serve the files from.
You might also try man tftpd: http://linux.die.net/man/8/tftpd
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
>
I'm using several $ characters in my password without escaping them in the
debconf file. Did you try that?
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
On Jan 17, 2011, at 2:15 AM,
Hi,
In $FAI/config/debconf/YOURCLASS
mysql-server mysql-server/root_password password myrootpassword
mysql-server mysql-server/root_password_again password myrootpassword
mysql-server mysql-server/root_password seen true
mysql-server mysql-server/root_password_again seen true
David Dreezer
Thank you both. So, essentially you are both saying that this part is not
necessary?
debconf-loadtemplate shared /target/etc/java/sun-java6.license;
That I don't need to be doing anything with the license file?
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ
Oh, sure I can. It actually did work with the 3.3.5 with the disk-align. I'll
try today's 3.4.6 release, remove that part, and see what it does.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @so
; /bin/echo 'set
shared/accepted-sun-dlj-v1-1 true'|debconf-communicate;
apt-get -y install sun-java6-jdk sun-java6-bin sun-java6-jre tomcat6
exit $error
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
48
with a transplanted 3.4.5 setup-storage, yet with the 3.3.5 version
transplanted in it worked just fine. Kind of odd.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
On Dec 22, 2010, at 2:09 P
Hi Michael,
I can pack up logs from 3.3.5 and 40~48. Do I send them here or to you
personally?
I can give 3.4.5 a try, sure. I only went to 4.0 as it was the only version
that I was able to get SW raid working with. I'll try 3.4.5 a try today and
see what it does.
David Dreezer
Cus
nfuse the issue when ultimately it
did not fix the problem.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
On Dec 22, 2010, at 1:28 AM, Mathieu Alorent wrote:
> Hi,
>
> Le mardi 21 déce
to him as I don't think I can (or should) attach it in an email to the group
here.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
You are right. As I had said in a previous email I had edited the file on the
fai server then ran fai-setup again. This did nothing because the file is
written by a deb package. I went back and manually edited it in the nfsroot
and that took care of the problem.
David Dreezer
Customer
ff:ff:ff:ff:ff:ff
David Dreezer
Customer Advocate, Social Strata, Inc.
On Dec 13, 2010, at 11:30 PM, Cristian Ionescu-Idbohrn wrote:
> On Mon, 13 Dec 2010, David Dreezer wrote:
>>
>> This did not work, unfortunately. I patched the get-boot-info script yet
>> FAI still thinks et
I patched it on the fai server, than ran fai-setup -v immediately after.
David Dreezer
Customer Advocate, Social Strata, Inc.
On Dec 14, 2010, at 1:36 AM, Thomas Lange wrote:
>>>>>> On Mon, 13 Dec 2010 20:50:58 -0800, David Dreezer
>>>>>> said:
>
"eth0
eth1"
netdevices="eth0
eth1"
BROADCAST='10.0.1.255'
[cut]
I'm not sure what the next step is.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveCloud · Eve Community
Follow us: @socialstrata
On Dec 13, 201
Hi Thomas,
I left out an important piece of information. That is, we're using the latest
experimental. Don't know fi this makes a difference, I did not look at the 3.x
script.
David Dreezer
Customer Advocate, Social Strata, Inc.
Online Community HQ Since 1996
Hoop.la · LiveC
be modified to handle this in
case Squeeze was purposely changed, or doesn't get fixed any time soon?
David Dreezer
Customer Advocate, Social Strata, Inc.
Sure, I'll try to give it a run over the weekend.
David Dreezer
Customer Advocate, Social Strata, Inc.
On Dec 3, 2010, at 2:27 PM, Michael Tautschnig wrote:
> Hi David,
>
>> The 10-setup in GRUB_PC as provided with 4.0~beta2+experimental41 does not
>> seem to wo
Thanks Michael, that will be a nice change.
David Dreezer
Customer Advocate, Social Strata, Inc.
On Dec 3, 2010, at 2:18 PM, Michael Tautschnig wrote:
> Hi,
>
> [...]
>
>> Might I suggest that the current documentation be changed to remove
>> references to partit
age --output=/boot/grub/core.img ext2 pc
biosdisk raid
$ROOTCMD /usr/sbin/grub-install '(md0)'
$ROOTCMD /usr/sbin/grub-setup '(md0)'
$ROOTCMD /usr/sbin/update-grub
$ROOTCMD update-initramfs -k all -u
;;
esac
exit $error
David Dreezer
Social Strata, Inc.
ze;
* in future releases KB, MB, GB, ... will be treated as 1000 instead
* of 1024 (KiB, MiB, GiB, ...) multipliers */
Thank you,
David Dreezer
On Nov 13, 2010, at 8:48 AM, Michael Tautschnig wrote:
>> Hi Michael,
>>
>> Thank you.
>>
>> I actually did have -
Hi Michael,
Thank you.
I actually did have - - in there but was receiving the following error:
ERROR (line 7): Invalid file: Was expecting /\Z/ but found "logical 1.7G"
instead
I did a search of this newsgroup and found a post by you dated Wed, 29 Apr 2009
05:56:09 -0700
(http://www.mail-
Hi,
I'm trying to setup raid1 on a 2 drive system, using both 3.3.5 and
4.0~beta2+experimental40. I've got USE_SETUP_STORAGE=1 defined in my class.
The file in disc_config is as below. I'm receiving an error that mkfs.logical
cannot be found in PATH. I'm not sure where I'm going wrong. Thank y
81 matches
Mail list logo
