Does anyone know if any of the SLES releases 9 or 10 support passing the
linuxrc option netdevice=<hardware mac address> ?
I noticed Red Hat has had this for sometime and it seems totally missing
in SuSE which seems really odd.
I'm still scratching my head if we really are still passing only by
kernel interface name.
I read in the change log for linuxrc-1.8.6-3 a change that "sounds" like
this feature.
* Mon Mar 14 2005 - snwint(a)suse.de
- s390: fix segfault in net_list_s390_devs()
- extra root image was sometimes not used
- allow netdevices to be specified by hwaddress
But it looks like linuxrc-1.8.6-3 is only available in SuSE Professional
9.3.
"any" commentary would be sincerely appreciated
the feature also seems to be fairly minor since any delimiter seperating
the octets in the hwaddr string could be used to identify the difference
between an interface name and interface hw address
if i could figure out a generic method of overriding the linuxrc with a
custom script and daisy chain it back into linuxrc after setting up the
interface name, I would.. and have a slight idea on how to do that.. but
i'd rather find out i'm just missing something and the feature is
already there
Thanks!
I dont have kinternet installed.
This is what doesnt seem to work:
<software>
<patterns config:type="list">
<pattern>default</pattern>
<pattern>Basis-Devel</pattern>
<pattern>x11</pattern>
</patterns>
<packages config:type="list">
<packages>perl-libwww-perl</packages>
<packages>net-snmp</packages>
<packages>kernel-source</packages>
</packages>
<remove-packages config:type="list">
<packages>pptp</packages>
<packages>smpppd</packages>
<packages>ppp</packages>
<packages>wvdial</packages>
<packages>mutt</packages>
</remove-packages>
</software>
In y2log it mentions that those packages should be removed, but it goes ahead and installs them anyway. I can do it manually on the command line:
# rpm -e pptp smpppd ppp wvdial mutt
and that works fine so I know I have no dependency problems.
matt
> -----Original Message-----
> From: Uwe Gansert [mailto:ug@suse.de]
> Sent: Friday, 30 June 2006 12:50 AM
> To: suse-autoinstall(a)suse.com
> Subject: Re: [suse-autoinstall] SLES 10 RC2 remove-packages section?
>
>
> On Thursday 29 June 2006 16:45, Manfred Hollstein wrote:
>
> > If that fails, you have kinternet installed; add that to
> the list and it
> > should work.
>
> good point.
> <remove-packages> can not break dependencies and so it's very
> likely that
> there is still a package that needs those packages.
> I did a very small installation with no X - so it worked for me.
>
> --
> ciao, Uwe Gansert
>
> Uwe Gansert, Server Technologies Team
> SUSE LINUX Products GmbH, Maxfeldstrasse 5, D-90409 Nuernberg, Germany
> Fax: +49-(0)911-74053-476, Web: http://www.suse.de/~ug
> now playing Covenant - Wasteland
>
> --
> To unsubscribe, e-mail: suse-autoinstall-unsubscribe(a)suse.com
> For additional commands, e-mail: suse-autoinstall-help(a)suse.com
>
>
This email and any attachments may contain privileged and confidential information and are intended for the named addressee only. If you have received this e-mail in error, please notify the sender and delete this e-mail immediately. Any confidentiality, privilege or copyright is not waived or lost because this e-mail has been sent to you in error. It is your responsibility to check this e-mail and any attachments for viruses. No warranty is made that this material is free from computer virus or any other defect or error. Any loss/damage incurred by using this material is not the sender's responsibility. The sender's entire liability will be limited to resupplying the material.
What does "Can't detect Profile Version" usually indicate?
>From y2log..
===
2007-02-28 01:27:04 <1> 10.42.37.174(3375) [YCP] AutoinstScripts.ycp:545 Script Execution command: /bin/sh /tmp/YaST2-03375-S80cIe/pre-scripts/pre.sh 2&> /tmp/YaST2-03375-S80cIe/pre-scripts/logs/pre.sh.log
2007-02-28 01:27:04 <1> 10.42.37.174(3375) [YCP] XML.ycp:113 Reading /tmp/profile/modified.xml
2007-02-28 01:27:04 <1> 10.42.37.174(3375) [agent-xml] XmlAgent.cc(Read):758 Setting XML generic error handler
2007-02-28 01:27:04 <1> 10.42.37.174(3375) [agent-xml] XmlAgent.cc(Read):793 Freeing generic XML error handler
2007-02-28 01:27:04 <1> 10.42.37.174(3375) [YCP] Profile.ycp:167 importing profile
2007-02-28 01:27:04 <1> 10.42.37.174(3375) [YCP] Profile.ycp:149 Can't detect Profile Version
===
The bizarre thing is that I'm trying to use a pre.sh script to replace
some lines in the profile, writing it out as /tmp/profile/modified.xml
and this fails. But, if I take exactly that modified.xml file (scp'd
off via the console) and feed that back into the same host again on a new
attempt at installing, I get the above error, but the autoinst.xml
(which should be invalid too since it's identical[*]) works fine. This
makes no sense... why would it reject modified.xml but work just fine as
the fallback?
Is there something in the parsing of modified.xml that is a little more
strict I might be missing?
[*]there are differences in a couple of the scripts, the parsing I'm
doing removes the backslashes that I used i some sed commands, so I
double backslash them. This doesn't effect the profile itself though.
A diff shows that the autoinst.xml (which works) vs the modified.xml
(which is rejected) are the same, other then the backslashes inside the
scripts CDATA blocks.
--
Mike Marion-Unix SysAdmin/Staff Engineer-http://www.qualcomm.com
Grandpa: "Quick! We have to kill the boy!"
Marge: "How'd you know he's a vampire?"
Grandpa: "He's a vampire?!? Ahhh...." [runs away] ==> Simpsons
--
To unsubscribe, e-mail: opensuse-autoinstall+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-autoinstall+help(a)opensuse.org
Hi All,
I've managed to use the ldap client (<ldap>) and server
(<ldap-server>) sections of my autoinst.xml to setup the
server with both as ldap server and ldap client.
I would assume that the ldap client stuff will work eventually,
but there's a problem with the server part: the ldap database
does not get initialized, I still have to go in and do that
step manually.
Does anyone know how to automate this step?
Any thoughts appreciated
/Lars
--
To unsubscribe, e-mail: opensuse-autoinstall+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-autoinstall+help(a)opensuse.org
All,
I have searched this list and www.suse.de/~ug/.cannot and cannot find any
documentation on adding a service pack to a SLES10 installation repository -
is this documented anywhere?
Thanks,
--
Jeremy Silver
jeremy(a)broadware.com
--
To unsubscribe, e-mail: opensuse-autoinstall+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-autoinstall+help(a)opensuse.org
I'm trying to use the sysconfig setting as described in
<http://www.suse.com/~ug/autoyast_doc/createprofile.sysconfig.html>
like so:
<sysconfig config:type="list" >
<sysconfig_entry>
<sysconfig_key>DISPLAYMANAGER_AUTOLOGIN</sysconfig_key>
<sysconfig_path>/etc/sysconfig/displaymanager</sysconfig_path>
<sysconfig_value>root</sysconfig_value>
</sysconfig_entry>
</sysconfig>
That doesn't bite, there's no change being performed.
What am I doing wrong here?
/Lars Stavholm
--
To unsubscribe, e-mail: opensuse-autoinstall+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-autoinstall+help(a)opensuse.org
Hi,
Both on SLES 9 and SLES 10 when you configure the NTP client via the
autoyast profile, when the system is installed the config file /etc/
ntp.conf is lacking the line:
driftfile /var/lib/ntp/drift/ntp.drift
Which means that every time the machine is rebooted NTP has to
establish the drift from scratch which means the time oscillates
randomly for about 15 minutes after booting up. This in turn causes
problems with time sensitive applications and in particular our web
authentication servers have a problem as suddenly things start
appearing to be in the future which is treated as an error...
The autoyast snippet we use to do the ntp client config is (I have
obfuscated the ip addresses of the NTP servers):
<quote>
<ntp-client>
<peers config:type="list">
<peer>
<initial config:type="boolean">false</initial>
<key>server</key>
<value>blah0.blah.blah</value>
</peer>
<peer>
<initial config:type="boolean">false</initial>
<key>server</key>
<value>blah1.blah.blah</value>
</peer>
<peer>
<initial config:type="boolean">false</initial>
<key>server</key>
<value>blah2.blah.blah</value>
</peer>
</peers>
<start_at_boot config:type="boolean">true</start_at_boot>
</ntp-client>
</quote>
This results in this ntp config file:
<quote>
server blah0.blah.blah
server blah1.blah.blah
server blah2.blah.blah
# alternate log file
# logconfig =syncstatus + sysevents
# logconfig =all
# statsdir /tmp/ # directory for statistics files
# filegen peerstats file peerstats type day enable
# filegen loopstats file loopstats type day enable
# filegen clockstats file clockstats type day enable
#
# Authentication stuff
#
# keys /etc/ntp.keys # path for keys file
# trustedkey 1 2 3 4 5 6 14 15 # define trusted keys
# requestkey 15 # key (7) for accessing server variables
# controlkey 15 # key (6) for accessing server variables
</quote>
Is this a bug or are we missing some magic line(s) in the autoyast
profile to cause a drift file to be added so our web authentication
servers stop being confused when they are rebooted? The autoyast
documentation does not seem to suggest that there is a way to
configure a drift file in the autoyast profile so I think this is a
bug with autoyast...
Thanks a lot in advance!
Best regards,
Anton
--
Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @)
Unix Support, Computing Service, University of Cambridge, CB2 3QH, UK
Linux NTFS maintainer, http://www.linux-ntfs.org/
--
To unsubscribe, e-mail: opensuse-autoinstall+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-autoinstall+help(a)opensuse.org
While working on a SLES10 image, I'm running into a problem where it's
ignoring my updates path, even though I've signed content and put the
key into the initrd file, and included the signature parts in the
profile:
My general section looks like:
<general>
<mode>
<confirm config:type="boolean">false</confirm>
<forceboot config:type="boolean">false</forceboot>
</mode>
<mouse>
<id>none</id>
</mouse>
<signature_handling>
<accept_file_without_checksum
config:type="boolean">true</accept_file_without_checksum>
<accept_unsigned_file
config:type="boolean">true</accept_unsigned_file>
<accept_unknown_gpg_key
config:type="boolean">true</accept_unknown_gpg_key>
<accept_verification_failed
config:type="boolean">true</accept_verification_failed>
</signature_handling>
</general>
but I still get "Can't find packages openafs, openafs-client,
openafs-devel" (Our builds of openafs) that are in updates, and y2log
shows:
2007-03-26 18:16:35 <0> 10.42.37.135(3306) [zypp] ExternalProgram.cc(checkStatus):333 pid 3816 successfully completed
2007-03-26 18:16:35 <1> 10.42.37.135(3306) [zypp::KeyRing] KeyRing.cc(verifyFileSignatureWorkflow):328 File [/var/tmp/TmpDir.CnJ7LU/DATA/content] ( /content (nfs://10.46.16.35/local/mnt/x86_64/sles10/updates) ) signed with unknown key []
2007-03-26 18:16:35 <1> 10.42.37.135(3306) [zypp::KeyRing] KeyRing.cc(verifyFileSignatureWorkflow):336 User does not want to accept unknown key
2007-03-26 18:16:35 <5> 10.42.37.135(3306) [base] Exception.cc(log):94 SuseTagsImpl.cc(downloadMetadata):261 THROW: SuseTagsImpl.cc(downloadMetadata):261: Error. Source signature does not validate and user does not want to continue.
2007-03-26 18:16:35 <0>
10.42.37.135(3306) [zypp] PathInfo.cc(_Log_Result):295 recursive_rmdir /var/tmp/TmpDir.CnJ7LU
2007-03-26 18:16:35 <0> 10.42.37.135(3306) [zypp] TmpPath.cc(~Impl):78 TmpPath cleaned up /var/tmp/TmpDir.CnJ7LU{d 0700 0/0}
2007-03-26 18:16:35 <5> 10.42.37.135(3306) [base] Exception.cc(log):94 SuseTagsImpl.cc(saveMetadataTo):339 THROW: SuseTagsImpl.cc(saveMetadataTo):339: Downloading metadata failed (is a susetags source?) or user did not accept remote source. Aborting refresh.
It explicitly says "User does not want to accept unknown key" which
isn't true. An almost identical setup for SLED10 is working (it's
updates path was create with create_update_source.sh from SLED10, the
above from the SLES10 version of the script).
The key also shouldn't show up as unknown (I would think) since it had
imported the key left in the initrd.
Also.. there seem to be tons of things missing from the dtd files so
that using xmllint --valid is basically useless because things like
signature_handling aren't even defined (nor are it's interior bits).
I have applied the sles10 root fix v.4 script too.
--
Mike Marion-Unix SysAdmin/Staff IT Engineer-http://www.qualcomm.com
Peggy: "12 years old and drinking a beer!?!"
Bobby: "I didn't even like it!"
Hank: "Well now you're just trying to get me mad!" ==> King of the Hill
--
To unsubscribe, e-mail: opensuse-autoinstall+unsubscribe(a)opensuse.org
For additional commands, e-mail: opensuse-autoinstall+help(a)opensuse.org
Hi, all,
When we develop the software, one tough issue happens during the
auto-installation of SLES 10 with autoyast scripts (see below).
Pls give us some suggestions or solutions to this issue.
Look forward to your reply. Thanks!!
[Issue summary]:Autoinstall SLES10 with autoyast scripts.
If there are four primary partitions on the disk, three of them
are used to install OS, then the fourth partition can not be remained.
Why? And how to keep the fourth primary partition?
[Issue details]
Partition List:
Device Boot Start End Blocks Id System
/dev/hdc1 * 1 582 293296+ 83 Linux
/dev/hdc2 583 1746 586656 82 Linux swap
/dev/hdc3 1747 17248 7813008 83 Linux
/dev/hdc4 17249 39770 11351088 83 Linux
SLES10 is planned to install on the previous three partitions:
/dev/hdc1---------/boot, formatted as ext2 in advanced, put autoinstall
scripts on this partition(path: /prep/autoinst.xml)
/dev/hdc2---------swap, formatted in the process of auto-install
/dev/hdc3---------/, formatted as reiser in the process of auto-install
/dev/hdc4-------suppose that this partition is a data section need to be
remained.
Step 1: make auto-install scripts - "autoinst.xml" as follows:
Reference to
http://forgeftp.novell.com/yast/doc/SLES10/autoinstall/CreateProfile.Partit…
...
<partitioning config:type="list">
<drive>
<device>/dev/hdc</device>
<partitions config:type="list">
<partition>
<create config:type="boolean">false</create>
<format config:type="boolean">false</format>
<mount>/boot</mount>
<partition_nr config:type="integer">1</partition_nr>
</partition>
<partition>
<create config:type="boolean">false</create>
<format config:type="boolean">true</format>
<filesystem config:type="symbol">swap</filesystem>
<partition_nr config:type="integer">2</partition_nr>
<mount>swap</mount>
</partition>
<partition>
<create config:type="boolean">false</create>
<format config:type="boolean">true</format>
<filesystem config:type="symbol">reiser</filesystem>
<mount>/</mount>
<partition_nr config:type="integer">3</partition_nr>
</partition>
</partitions>
<use>1,2,3</use>
</drive>
</partitioning>
...
Step 2: Boot by SLES10 DVD, set startup parameter:
autoyast=device://hdc1/prep/autoinst.xml
Step 3: The error would be prompted when the process comes to "Create
Partition Plan". Then the installation would abort.
At this time, check /var/log/YaST2/y2log, we have found that error
exists in /usr/share/YaST2/include/autoinstall/autopart.ycp.
Try other way:
1>Change scripts like <use>linux</use> or <use>all</use>
No error would be promped. But the fourth primary partition - hdc4 is
deleted.
2>Make use of OpenSUSE 10.2 DVD to boot system and don't change scripts
(keep the same script above). The whole process of auto-install is normal.
3>When the error happens,
/usr/share/YaST2/include/autoinstall/autopart.ycp is replaced by
autopart.ycp of OpenSUSE10.2.
Then rebuild /usr/share/YaST2/modules/AutoinstStorage.ycp to generate
AutoinstStorage.ybc.
After updating this new AutoinstStorage.ybc in SLES10, the whole process
of auto-install is normal.
Whether is this one bug of SLES 10?
Furthermore, if the fourth partition is extended section (see below), the
auto-install of SLES 10 is also ok.
/dev/hdc1 * 1 582 293296+ 83 Linux
/dev/hdc2 583 1746 586656 82 Linux swap
/dev/hdc3 1747 17248 7813008 83 Linux
/dev/hdc4 17249 39770 11351088 5 Extended
/dev/hdc5 17249 39770 11351056+ 83 Linux
By the way, refer to
http://www.suse.de/~ug/autoyast_changes_SLES10SP1.html, SLES10SP1 has
resolved this issue on the condition of four primary partitions. Is that
right? Pls confirm it. If it's true, when will SLES10SP1 be released?
Best Regards!
*****************************************
Cheng Ningting(Carol Cheng)
Server Business Department
Lenovo Group Limited
6#,ShangDi ChuangYe Road,HaiDian,BEIJING
TEL:(010)58864185
E_Mail:chengnt(a)lenovo.com
****************************************
We are the new, we are the future, we are LENOVO!