devil-linux-develop — developer discussions

Re: [Devil-linux-develop] changes in build setup for 1.1?

[<hzu...@ra...>]

On 01/05/2004 10:27:36 AM t.tait wrote:
>> > Looks like 1.1 has some revamped build methods,
>>
>> Yes, a little.  :-)
>>
>> > I guess the build order
>> > is now done automatically via the dependencies list in the scripts?
>>
>> Yes, it uses the "Required-Start/Provides/Default-Start" stuff.
>>
>> > But how is a new package added to the menus?
>>
>> That hasn't changed.  Are you familiar with the old method?
>>
>> > Also, is there really a need for a .configure file for each package?
>> > And does the makefile need all those options? The package in
>question
>> > (<http://thekelleys.org.uk/dnsmasq/>) has a very simple makefile
>that
>> > seems to compile by itself just fine....
>>
>> That has nothing to do with DL.  It depends on the package itself.
>> Some packages come with a ./configure file, some don't.  Some have
>> different instructions to build and install the package.  You have
>> to write the script to match the package install instructions.
>>
>>  - BS
>
>I've forgotten how I got it into the menus before:) I lost that lfs
>tree during some gyrations with my build system:( I added the line into
>the build/.config file, but I'm sure thats not all it needs. I also
>need to do the insserv thing when setup on target too. Thanks-

It's best just to execute "make prepare" to get the scripts re-ordered.

For the configuration part, take a look into build/scripts/configuration
and build/scripts/configuration/help

Heiko

Re: [Devil-linux-develop] problem with user upgrade script from 0.5 to 1.1.1-2004-01-04

[Friedrich L. <fl...@fl...>]

Bruce Smith wrote on 05.01.2004 19:32 MET:
>>I've just try how an upgrade from version 0.5 to 1.1.1-2004-01-04
>>works. So far so good but after the upgrade script has asked if it
>>should upgrade/migrate the users too I get the error messages
>>"unary operator expected" (see screenshot).
> 
> 
> I've never tried going from 0.5 to 1.X.  I don't have a 0.5 system to
> try it on any longer.

I've just found the config disk image of an old system which I used to
test what happens during the upgrade process - worst case testing ;-)

> Does 0.5 user shadow password files?  Shadow group files (gshadow)?

I can find the shadow but now gshadow. I've downloaded the 0.5a
release and there's also no gshadow included in that etc.tar.gz.

What I still have to try is to not repackage the 0.5 etc.tar.gz
to etc.tar.bz2. Will it catch that? We will see :-)

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

Re: [Devil-linux-develop] problem with user upgrade script from 0.5 to 1.1.1-2004-01-04

[Bruce S. <bw...@ar...>]

> I've just try how an upgrade from version 0.5 to 1.1.1-2004-01-04
> works. So far so good but after the upgrade script has asked if it
> should upgrade/migrate the users too I get the error messages
> "unary operator expected" (see screenshot).

I've never tried going from 0.5 to 1.X.  I don't have a 0.5 system to
try it on any longer.

Does 0.5 user shadow password files?  Shadow group files (gshadow)?

 - BS

Re: [Devil-linux-develop] error when mounting ISO of hd

[<hzu...@ra...>]

On 01/04/2004 10:55:55 PM Tim Tait wrote:
>I get this error when booting of the HD when it attempts to mount the
>ISO image:
>
>grsec: attempted resource overstep by requesting 2864099328 for
>RLIMIT_STACK against limit 8388608 by (busybox:83) UID(0) EUID(0),
>parent (mount_cdrom:64) UID(0) EUID(0)
>ISO 9660 Extensions: Microsoft Joliet Level 3
>ISO 9660 Extensions: RRIP_1991A
>
>It does boot, but leaves /initrd/mnt with the hd mounted on it.

This is an open "bug".
Since you're mounting the ISO within the initrd, initrd can't be unmounted.
You can fix it if you want. ;-)

Heiko

Re: [Devil-linux-develop] cyrus-imapd abort.

[<hzu...@ra...>]

On 01/05/2004 11:53:47 AM Bruce Smith wrote:
>> >New LFS, mrproper, all the latest updates (as of this minute).
>> >Is this caused by the upgrade to the new version of db-4?
>>
>> I got another error yesterday with cyrus-imapd on my box, but thought
>it's
>> something local....
>>
>> I probably have some time to look into it tonight, or in case you're
>> bored... ;-)
>
>I've been looking, but haven't figured it out yet.  Any hints?

No, sorry.

Heiko

Re: [Devil-linux-develop] [ devil-linux-Feature Requests-864429 ] add coda filesystem

[Friedrich L. <fl...@fl...>]

Diego Torres wrote on 05.01.2004 17:53 MET:
> 
> maybe if i buy you a beer... :))

I guess those times are over, but a nice car will do for Heiko, right? ;-)

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

[Devil-linux-develop] problem with user upgrade script from 0.5 to 1.1.1-2004-01-04

[Friedrich L. <fl...@fl...>]

Hi!

I've just try how an upgrade from version 0.5 to 1.1.1-2004-01-04
works. So far so good but after the upgrade script has asked if it
should upgrade/migrate the users too I get the error messages
"unary operator expected" (see screenshot).

-- 
MfG / Regards
Friedrich Lobenstock
____________________________________________________________________
Friedrich Lobenstock                       Linux Services Lobenstock
URL: http://www.lsl.at/                         Email: fl...@fl...
____________________________________________________________________

Re: [Devil-linux-develop] [ devil-linux-Feature Requests-864429 ] add coda filesystem

[Diego T. <dt...@co...>]

On Mon, Jan 05, 2004 at 11:42:39AM -0500, hzu...@ra... wrote:

> >heiko, any plans for this item? i'm planning on using it on a dl
> >server instead of samba|nfs (since there are binaries for
> >windows!!), so i add +1 to the priority :))
> 
> No plans yet, but you can work on it if you want. ;-)

on first place i'll fix smartmontools now that i've a working system (again) :)

besides that, coda looks terrific to setup/configure, since there are a lot of packages/libs requiered.

maybe if i buy you a beer... :))


-- 
-- gnupg keyfingerprint -- 48AF 5BF9 8F54 2966 64CC  2327 7CD0 DD91 B09D 5799
-- Use of a keyboard or mouse may be linked to serious injuries or disorders.  
Diego Torres - dtorres at anthalia dot org - Madrid / España

Re: [Devil-linux-develop] cyrus-imapd abort.

[Bruce S. <bw...@ar...>]

> >New LFS, mrproper, all the latest updates (as of this minute).
> >Is this caused by the upgrade to the new version of db-4?
> 
> I got another error yesterday with cyrus-imapd on my box, but thought it's
> something local....
> 
> I probably have some time to look into it tonight, or in case you're
> bored... ;-)

I've been looking, but haven't figured it out yet.  Any hints?

 - BS

Re: [Devil-linux-develop] [ devil-linux-Feature Requests-864429 ] add coda filesystem

[<hzu...@ra...>]

On 01/05/2004 08:01:26 AM "SourceForge.net" wrote:
>Feature Requests item #864429, was opened at 2003-12-22 14:20
>Message generated for change (Comment added) made by rapid2k1
>You can respond by visiting:
>https://sourceforge.net/tracker/?func=detail&atid=410646&aid=864429&grou
>p_id=34096
>
>Category: Packages
>Group: None
>Status: Open
>Priority: 5
>Submitted By: Heiko Zuerker (smiley73)
>Assigned to: Nobody/Anonymous (nobody)
>Summary: add coda filesystem

>heiko, any plans for this item? i'm planning on using it on a dl
>server instead of samba|nfs (since there are binaries for
>windows!!), so i add +1 to the priority :))

No plans yet, but you can work on it if you want. ;-)

Heiko

Re: [Devil-linux-develop] cyrus-imapd abort.

[<hzu...@ra...>]

On 01/05/2004 09:59:14 AM Bruce Smith wrote:
>New LFS, mrproper, all the latest updates (as of this minute).
>Is this caused by the upgrade to the new version of db-4?

I got another error yesterday with cyrus-imapd on my box, but thought it's
something local....

I probably have some time to look into it tonight, or in case you're
bored... ;-)

cu
  Heiko

Re: [Devil-linux-develop] changes in build setup for 1.1?

[<t....@co...>]

> > Looks like 1.1 has some revamped build methods, 
> 
> Yes, a little.  :-)
> 
> > I guess the build order 
> > is now done automatically via the dependencies list in the scripts? 
> 
> Yes, it uses the "Required-Start/Provides/Default-Start" stuff.
> 
> > But how is a new package added to the menus?
> 
> That hasn't changed.  Are you familiar with the old method?
> 
> > Also, is there really a need for a .configure file for each package? 
> > And does the makefile need all those options? The package in question 
> > (<http://thekelleys.org.uk/dnsmasq/>) has a very simple makefile that 
> > seems to compile by itself just fine....
> 
> That has nothing to do with DL.  It depends on the package itself.
> Some packages come with a ./configure file, some don't.  Some have
> different instructions to build and install the package.  You have 
> to write the script to match the package install instructions.
> 
>  - BS

I've forgotten how I got it into the menus before:) I lost that lfs tree during some gyrations with my build system:( I added the line into the build/.config file, but I'm sure thats not all it needs. I also need to do the insserv thing when setup on target too. Thanks-

Tim

[Devil-linux-develop] cyrus-imapd abort.

[Bruce S. <bw...@ar...>]

New LFS, mrproper, all the latest updates (as of this minute).
Is this caused by the upgrade to the new version of db-4?

 - BS


ar cr libimap.a append.o mailbox.o mboxlist.o mupdate-client.o mboxname.o message.o config.o imap_err.o mupdate_err.o proc.o setproctitle.o convert_code.o duplicate.o saslclient.o saslserver.o acapmbox.o signals.o annotate.o search_engines.o squat.o squat_internal.o mbdump.o imapparse.o telemetry.o user.o notify.o seen_db.o idle_poll.o
ranlib libimap.a
gcc -L/usr/local/lib -Wl,-rpath,/usr/local/lib    -o imapd \
 ../master/service.o pushstats.o backend.o imapd.o index.o tls.o version.o libimap.a ../acap/libacap.a ../lib/libcyrus.a  -lsasl2   -lssl -lcrypto  -lfl  -lresolv   -ldb-4 -lcom_err
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0xa4f): In function `gettid':
/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:409: undefined reference to `txn_begin'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x1318): In function `mystore':
/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:707: undefined reference to `txn_abort'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x134c):/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:688: undefined reference to `txn_begin'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x13f0):/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:721: undefined reference to `txn_commit'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x16ee): In function `mydelete':
/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:816: undefined reference to `txn_abort'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x172c):/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:798: undefined reference to `txn_begin'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x17d0):/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:830: undefined reference to `txn_commit'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x18d7): In function `mycommit':
/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:878: undefined reference to `txn_commit'
../lib/libcyrus.a(cyrusdb_db3.o)(.text+0x19ec): In function `abort_txn':
/data/build/tmp/cyrus-imapd-2.1.16/lib/cyrusdb_db3.c:917: undefined reference to `txn_abort'
collect2: ld returned 1 exit status
make[2]: *** [imapd] Error 1
make[2]: Leaving directory `/data/build/tmp/cyrus-imapd-2.1.16/imap'
make[1]: *** [all] Error 1
make[1]: Leaving directory `/data/build/tmp/cyrus-imapd-2.1.16'
root:/data/build#

[Devil-linux-develop] [fmII] amavisd-new 20030616-p7 released (Default branch)

[<no...@fr...>]

This email is to inform you about the release of version '20030616-p7' of
'amavisd-new' through freshmeat.net. All URLs and other useful information
can be found at

    http://freshmeat.net/projects/amavisd-new/

The changes in this release are as follows:
Missing modules are loaded when running in chroot. Cached spam results
are disregarded for small messages. The catchall default in SQL
lookups was improved. Encrypted archives are retained for virus
checking. The av_scanners list was updated.  

Project description:
amavisd-new is a high-performance and reliable  interface between MTAs
and one or more content  checkers, including virus scanners, and/or
the  Mail::SpamAssasin Perl module. It talks to the MTA via  (E)SMTP
or LMTP, or by using helper programs. No  timing gaps exist in the
design, which could cause a  mail loss. It is normally positioned at
or near a central  mailer, not necessarily where the user's mailboxes
and  final delivery takes place.  

Trove categories:
[Development Status  ] 5 - Production/Stable
[Intended Audience   ] System Administrators
[License             ] OSI Approved :: GNU General Public License (GPL)
[Operating System    ] POSIX, Unix
[Programming Language] Perl
[Topic               ] Communications :: Email :: Filters 

If you would like to cancel subscription to releases of this project,
login to freshmeat.net and choose 'home' from the personal menubar at the
top of the page. You'll be presented with a list of projects and
categories you're subscribed to in the right column, which you may cancel
by highlighting the project or category in question and clicking the
'delete' button.

Sincerely,
freshmeat.net 
 
____________________________| Advertising |____________________________
This email is sponsored by OSDN's Audience Survey. 

Help shape OSDN's sites and tell us what you think.
Take this five minute survey and you could 
win a $250 Gift Certificate.
 

http://www.wrgsurveys.com/2003/osdntech03.php?site=2 
____________________________| Advertising |____________________________

Re: [Devil-linux-develop] changes in build setup for 1.1?

[Bruce S. <bw...@ar...>]

> Looks like 1.1 has some revamped build methods, 

Yes, a little.  :-)

> I guess the build order 
> is now done automatically via the dependencies list in the scripts? 

Yes, it uses the "Required-Start/Provides/Default-Start" stuff.

> But how is a new package added to the menus?

That hasn't changed.  Are you familiar with the old method?

> Also, is there really a need for a .configure file for each package? 
> And does the makefile need all those options? The package in question 
> (<http://thekelleys.org.uk/dnsmasq/>) has a very simple makefile that 
> seems to compile by itself just fine....

That has nothing to do with DL.  It depends on the package itself.
Some packages come with a ./configure file, some don't.  Some have
different instructions to build and install the package.  You have 
to write the script to match the package install instructions.

 - BS

[Devil-linux-develop] [ devil-linux-Feature Requests-864429 ] add coda filesystem

[SourceForge.net <no...@so...>]

Feature Requests item #864429, was opened at 2003-12-22 14:20
Message generated for change (Comment added) made by rapid2k1
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410646&aid=864429&group_id=34096

Category: Packages
Group: None
Status: Open
Priority: 5
Submitted By: Heiko Zuerker (smiley73)
Assigned to: Nobody/Anonymous (nobody)
Summary: add coda filesystem

Initial Comment:
http://www.coda.cs.cmu.edu/

Coda is an advanced networked filesystem. It has been
developed at CMU since 1987 by the systems group of M.
Satyanarayanan. in the SCS department.

Coda is a distributed filesystem with its origin in
AFS2. It has many features that are very desirable for
network filesystems. Currently, Coda has several
features not found elsewhere.

   1. disconnected operation for mobile computing
   2. is freely available under a liberal license
   3. high performance through client side persistent
caching
   4. server replication
   5. security model for authentication, encryption and
access control
   6. continued operation during partial network
failures in server network
   7. network bandwith adaptation
   8. good scalability
   9. well defined semantics of sharing, even in the
presence of network failures 

----------------------------------------------------------------------

>Comment By: Diego Torres (rapid2k1)
Date: 2004-01-05 13:01

Message:
Logged In: YES 
user_id=367380


heiko, any plans for this item? i'm planning on using it on a dl 
server instead of samba|nfs (since there are binaries for 
windows!!), so i add +1 to the priority :))

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=410646&aid=864429&group_id=34096

[Devil-linux-develop] changes in build setup for 1.1?

[Tim T. <t....@co...>]

Looks like 1.1 has some revamped build methods, I guess the build order 
is now done automatically via the dependencies list in the scripts? But 
how is a new package added to the menus? Also, is there really a need 
for a .configure file for each package? And does the makefile need all 
those options? The package in question 
(<http://thekelleys.org.uk/dnsmasq/>) has a very simple makefile that 
seems to compile by itself just fine....

Thanks-

Tim

Re: [Devil-linux-develop] cvs problems?

[Heiko Z. <he...@zu...>]

Tim Tait wrote:
> 
> I keep getiing "eof recieved" errors from the cvs server, is it up? Or 
> is the problem at my end?
> 

This error oftern happens on the sourceforge side.
Keep on trying.... or wait until tomorrow.

Heiko

Re: [Devil-linux-develop] cvs problems

[Tim T. <t....@co...>]

Roland Pabel wrote:

>Hi,
>I tried to checkout 1.0.3 from cvs today several times from several system, 
>but had no success :
>
>cvs -d:pserver:ano...@cv...:/cvsroot/devil-linux 
>login
>Logging in to 
>:pserver:ano...@cv...:2401/cvsroot/devil-linux
>CVS password:
>cvs [login aborted]: end of file from server (consult above messages if any)
>
>It may be related to this SF status report: 
>http://sourceforge.net/docman/display_doc.php?docid=2352&group_id=1
>( 2003-12-12 09:00:25 - Project CVS Service )   The developer CVS server 
>incurred a glitch today, causing EOF errors for many users. This problem has 
>since been resolved. 
>
>Can someone confirm similar errors?
>thx,
>Roland
>PS: I wrote the scripts to support motion ( http://motion.sourceforge.net/ ) 
>in DL. Once tested, what's the procedure to include it officially?
>  
>
I can get it to work about once every 10 times... The good news, I 
finally got working build of 1.1.1! Smbmount is fine now.

Tim

[Devil-linux-develop] [Patch] HostAP 0.1.2 for DL

[Manu \MaNS\ E. <ee...@ma...>]

Hi All,

this is not really a patch as there is now 3 tarballs, 3 options in 
menuconfig, ... old one (called hostap) has to be erased from scripts/, 
scripts/configuration and scripts/configuration/help, as well as the 
tarball in src/.

Attached a tarball with new scripts and helpfiles ... untargz it in 
build directory.

here are the links to grab the sources packages :

http://hostap.epitest.fi/releases/hostap-driver-0.1.2.tar.gz
http://hostap.epitest.fi/releases/hostap-utils-0.1.2.tar.gz
http://hostap.epitest.fi/releases/hostapd-0.1.0.tar.gz

Next on the list is nagios ...

Regards

MaNU

[Devil-linux-develop] error when mounting ISO of hd

[Tim T. <t....@co...>]

I get this error when booting of the HD when it attempts to mount the 
ISO image:

  grsec: attempted resource overstep by requesting 2864099328 for 
RLIMIT_STACK against limit 8388608 by (busybox:83) UID(0) EUID(0), 
parent (mount_cdrom:64) UID(0) EUID(0)
  ISO 9660 Extensions: Microsoft Joliet Level 3
  ISO 9660 Extensions: RRIP_1991A

It does boot, but leaves /initrd/mnt with the hd mounted on it.

Tim

Re: [Devil-linux-develop] [PATCH] linuxrc

[Tim T. <t....@co...>]

Bruce Smith wrote:

>>>>I made what I think is a few enhancements to the linuxrc script... now 
>>>>this is my 1st attempt at any real script mods, so I may not have it 
>>>>all correct.
>>>>
>>>>First is simple, I have it check floppies for a config 1st, thus 
>>>>allowing someone to temporarily test new configs by overiding one 
>>>>stored on a HD. Probably be good to check other removeable media list 
>>>>USB 1st too but thats trickier. I don't think this raises a security 
>>>>risk, cause if someone can put floppies into your machine they can do 
>>>>a lot worse too.
>>>>
>>>>2nd, if you have multiple possible sources of  the configuration file, 
>>>>if it's not on the first one, it halts the boot to prompt if you want 
>>>>to write a default one to it. This is really bad, because it hangs the 
>>>>system on an unattended reboot. So I tweaked the script to check all 
>>>>media 1st, then if none was found do a 2nd pass to look for blank 
>>>>media. Of course lastly it still checks the CDROM. Is it possible to 
>>>>pass the config source from grub as option, like it works for  
>>>>"root=/dev/hda1"?
>>>>        
>>>>
>>>2nd try with "unified diff" attachment. I couldn't get the cvs diff -run 
>>>to work, sorry.
>>>      
>>>
>>I can't look into it at the moment, I have a full re-compile running.
>>Bruce you got time? Otherwise I'll try to do it tomorrow night.
>>    
>>
>
>I might have time tomorrow.
>
>I haven't looked at the patch yet, but I'm not sure about checking for
>floppy first.  That would check floppy before the USB memory stick.
>I think the USB flash drive should have priority over the floppy. 
>Heiko, what's your opinion on that?
>  
>
I think it needs to check both... my thought was if you have a USB 
config then you are unlikely to leave a floppy in, except to test 
something. Also, I was not sure how to make it check USB vs a SCSI 
drive. Either way I think removable media 1st.. of course since I have 
an IDE CF Card, it get's more convoluted. There probably is no canned 
config which solves every setup, which is why I was asking about passing 
it as a parameter from grub for installs to usb/hd .

>Also, the search order for configs need to match the save-config script.
>It's really confusing if your save goes to a different location than the
>boot is using.  (believe me, I know that first hand! :)   So, if you
>change the boot order, you also need to modify the save-config script.
>
> - BS
>  
>
I agree on the save-config script, that was my next order of business. 
Ideally though, save-config should *know* where the source was, not 
guess. Hot-pluggable media coudl ahve been added since boot, etc. Can't 
we pass it some how from linuxrc?

Tim

Re: [Devil-linux-develop] [PATCH] linuxrc

[Bruce S. <bw...@ar...>]

> > I haven't looked at the patch yet, but I'm not sure about checking for
> > floppy first.  That would check floppy before the USB memory stick.
> > 
> > I think the USB flash drive should have priority over the floppy. 
> > Heiko, what's your opinion on that?
> 
> I think so, too. But that means that SCSI devices come before the floppy 
> comes before IDE devices.

And if save-config has the same search order, and if IDE drives come
before floppies, then save-config will be saving the config to IDE hard
drives if they exist.

I'm not sure that's a good idea either, is it???

> > Also, the search order for configs need to match the save-config script.
> > It's really confusing if your save goes to a different location than the
> > boot is using.  (believe me, I know that first hand! :)   So, if you
> > change the boot order, you also need to modify the save-config script.
> 
> Yes, save-config *and* the initrd need to be in sync.

 - BS

Re: [Devil-linux-develop] [PATCH] linuxrc

[Heiko Z. <he...@zu...>]

Bruce Smith wrote:
>>>>I made what I think is a few enhancements to the linuxrc script... now 
>>>>this is my 1st attempt at any real script mods, so I may not have it 
>>>>all correct.
>>>>
>>>>First is simple, I have it check floppies for a config 1st, thus 
>>>>allowing someone to temporarily test new configs by overiding one 
>>>>stored on a HD. Probably be good to check other removeable media list 
>>>>USB 1st too but thats trickier. I don't think this raises a security 
>>>>risk, cause if someone can put floppies into your machine they can do 
>>>>a lot worse too.
>>>>
>>>>2nd, if you have multiple possible sources of  the configuration file, 
>>>>if it's not on the first one, it halts the boot to prompt if you want 
>>>>to write a default one to it. This is really bad, because it hangs the 
>>>>system on an unattended reboot. So I tweaked the script to check all 
>>>>media 1st, then if none was found do a 2nd pass to look for blank 
>>>>media. Of course lastly it still checks the CDROM. Is it possible to 
>>>>pass the config source from grub as option, like it works for  
>>>>"root=/dev/hda1"?
>>>
>>>2nd try with "unified diff" attachment. I couldn't get the cvs diff -run 
>>>to work, sorry.
>>
>>I can't look into it at the moment, I have a full re-compile running.
>>Bruce you got time? Otherwise I'll try to do it tomorrow night.
> 
> 
> I might have time tomorrow.
> 
> I haven't looked at the patch yet, but I'm not sure about checking for
> floppy first.  That would check floppy before the USB memory stick.
> 
> I think the USB flash drive should have priority over the floppy. 
> Heiko, what's your opinion on that?

I think so, too. But that means that SCSI devices come before the floppy 
comes before IDE devices.

> Also, the search order for configs need to match the save-config script.
> It's really confusing if your save goes to a different location than the
> boot is using.  (believe me, I know that first hand! :)   So, if you
> change the boot order, you also need to modify the save-config script.

Yes, save-config *and* the initrd need to be in sync.

Heiko

Re: [Devil-linux-develop] [PATCH] linuxrc

[Bruce S. <bw...@ar...>]

> >> I made what I think is a few enhancements to the linuxrc script... now 
> >> this is my 1st attempt at any real script mods, so I may not have it 
> >> all correct.
> >>
> >> First is simple, I have it check floppies for a config 1st, thus 
> >> allowing someone to temporarily test new configs by overiding one 
> >> stored on a HD. Probably be good to check other removeable media list 
> >> USB 1st too but thats trickier. I don't think this raises a security 
> >> risk, cause if someone can put floppies into your machine they can do 
> >> a lot worse too.
> >>
> >> 2nd, if you have multiple possible sources of  the configuration file, 
> >> if it's not on the first one, it halts the boot to prompt if you want 
> >> to write a default one to it. This is really bad, because it hangs the 
> >> system on an unattended reboot. So I tweaked the script to check all 
> >> media 1st, then if none was found do a 2nd pass to look for blank 
> >> media. Of course lastly it still checks the CDROM. Is it possible to 
> >> pass the config source from grub as option, like it works for  
> >> "root=/dev/hda1"?
> > 
> > 2nd try with "unified diff" attachment. I couldn't get the cvs diff -run 
> > to work, sorry.
> 
> I can't look into it at the moment, I have a full re-compile running.
> Bruce you got time? Otherwise I'll try to do it tomorrow night.

I might have time tomorrow.

I haven't looked at the patch yet, but I'm not sure about checking for
floppy first.  That would check floppy before the USB memory stick.

I think the USB flash drive should have priority over the floppy. 
Heiko, what's your opinion on that?

Also, the search order for configs need to match the save-config script.
It's really confusing if your save goes to a different location than the
boot is using.  (believe me, I know that first hand! :)   So, if you
change the boot order, you also need to modify the save-config script.

 - BS