Page: Previous  1, 2, 3 ... 10, 11, 12 ... 22, 23, 24  Next

techAdmin
Status: Site Admin
Joined: 26 Sep 2003
Posts: 3735
Location: East Coast, West Coast? I know it's one of them.
Reply Quote
anticapitalista, I can add an option to override the default grub file path, please post a sample from your system, the entire path, including the file name for the grub file.

However, keep in mind that looking ahead, menu.lst doesn't have a long time left here on earth.
Back to top
techAdmin
Status: Site Admin
Joined: 26 Sep 2003
Posts: 3735
Location: East Coast, West Coast? I know it's one of them.
Reply Quote
Actually,, I forgot you can set this yourself I think, simply use the sgfxi sticky options

While i don't say it in that documentation, you can set any hard coded global that is set in the top part of the script to whatever you want, so in the case of sgfxi we find:

:: Code ::
GRUB1='/boot/grub/menu.lst'
GRUB1_FED='/boot/grub/grub.conf'
GRUB2='/boot/grub/grub.cfg'
GRUBED='/etc/default/grub'


Now none of these are hardcoded anywhere else in the script, so the script will test for the content of these global values so you simply take the one you want, let's say GRUB1, and do this:

:: Code ::
echo GRUB1='/etc/somewhere/grub/menu.lst' >> /usr/local/bin/sgfxi-values


now you'll note at about line 358 this:

:: Code ::
# allow user set globals to override script globals
if [ -f $SM_VALUES ];then
   source $SM_VALUES
fi


smxi, svmi, and sgfxi, and I think inxi, have this. smxi, sgfxi, and sgfxi require the config alt file to be named for ex: sgfxi-values and to be in /usr/local/bin

Any hard coded value there, or any option set parameter, can be set there permanently.

For example, let's say we always just want to always use the new skip grub test option -! 33 since the system is fine for sgfxi:

:: Code ::
echo B_SKIP_GRUB_TEST='true' >> /usr/local/bin/sgfxi-values


now sgfxi will always skip the grub test when you type this in: sgfxi

I almost forgot these config hard codings, but as you can see, they are documented, though I don't generally want people to do things out of the ordinary, but you certainly can if you have those needs.

So sgfxi always had this option, sorry I forgot, these things are so damned big now, and have had so many user requests both handled and anticipated, that it gets less and less likely that something isn't covered, but the grub skip test is a good new option, I think smxi has that too.

Keep in mind when you do this, the option or change is ALWAYS an override, to get rid of it you have to delete the entry in the sgfxi-values file, no comments are supported.
Back to top
oldhoghead
Status: Curious
Joined: 05 May 2009
Posts: 5
Location: DFW
Reply Quote
h2,

You are amazing!! Plugged in the value for my particular setup via the echo, script worked flawlessly, finding grub on another partition.
Nice!
Now to update my notes.

cheers,
ohh
Back to top
aus9
Status: Assistant
Joined: 21 Sep 2008
Posts: 358
Location: Australia
Reply Quote
today has a grub2 update and a new error which I have hived off to the grub folder as per

techpatterns.com/forums/about1805.html

Note that I am a simple person on a non-raided non-LVM 32 bit Debian distro and YMMV

EDIT

this appears to be the idiot behind the monitor and I got a new error error as grub2 would not update due to a gremlin with my other distro on sda2....appears resolved for now

sorry to have troubled you

but

the grub LIB link remains as a note to myself in the hope it my help others

Please do not walk away from your smxi updates is the clue I hope?

good luck
Back to top
drb
Status: Contributor
Joined: 09 Jul 2009
Posts: 130
Reply Quote
h2,

There is a mismatch between hplip and hplip-data at the moment but the automatic check doesn't appear to be picking it up prior to dist-upgrade
Back to top
techAdmin
Status: Site Admin
Joined: 26 Sep 2003
Posts: 3735
Location: East Coast, West Coast? I know it's one of them.
Reply Quote
oh, ok, I'll check the versions.

Sadly, there is no standard for numbering mismatches, and even among the same package sets, the parts of the version numbers that create mismatches varies over time. So it makes it always an empirical process to update.
Back to top
update-initramfs
anticapitalista
Status: Contributor
Joined: 13 Jun 2008
Posts: 199
Location: Greece
Reply Quote
Request,

Could the update-initramfs use the -u -t and -k all options?
Back to top
techAdmin
Status: Site Admin
Joined: 26 Sep 2003
Posts: 3735
Location: East Coast, West Coast? I know it's one of them.
Reply Quote
which script are you talking about?

smxi uses either -u or -c depending if it's updating or creating.

sgfxi uses -k -u

What is the -t for?
Back to top
techAdmin
Status: Site Admin
Joined: 26 Sep 2003
Posts: 3735
Location: East Coast, West Coast? I know it's one of them.
Reply Quote
drb re the package version mismatch, make SURE to post the versions, I'm not seeing any version mismatch today.

Do this and post the output when you post a version mismatch bug report:

apt-cache policy hplip hlip-data

that shows me which part of the number string is triggering the version mismatch.
Back to top
anticapitalista
Status: Contributor
Joined: 13 Jun 2008
Posts: 199
Location: Greece
Reply Quote
According to the man page:
-t Allows to take over an custom initramfs with a newer one.

Sometimes, when using smxi and sgfxi, the update-initramfs part doesn't work and gives a message to use with the -t option.

This has also happened when running out of the smxi/sgfxi script as well. It has happened with various kernels, from liquorix to Debian with no obvious reason why it sometimes requests the -t option and other times, default is ok.

The problem is that if users do not see that initramfs has not been updated, the boot will fail, and users will not know why.
Back to top
Display posts from previous:   
Page: Previous  1, 2, 3 ... 10, 11, 12 ... 22, 23, 24  Next
All times are GMT - 8 Hours