Page 8 of 19

Posted: Thu 03 Nov 2011, 06:08
by thkn777
fyi: Tested on Lucid Puppy 5.2.8, PUPMODE=7 --> NOT working :cry:

Failure with PUPMODE=7

Posted: Thu 03 Nov 2011, 11:01
by shinobar
thanks puppynoob101 and thkn777 for the report.
I could reproduce the failure with PUPMODE=7.

Currently it does not support PUPMODE=7 (saving session in entire partition on a flash device).
Use the bootmanager to load sfs.
But i recommend to make a fresh install and save your session in a file to make PUPMODE=13, not to the entire partition.

EDIT: sfs_load v.1.3 supports PUPMODE=7, but i still recommend saving your session in a file(so that the PUPMODE=13), not to the partition.

Russian translation

Posted: Sat 05 Nov 2011, 13:00
by rodin.s
I have updated Russian translation for sfs_load.

UPDATE: sfs_load-1.3

Posted: Sun 13 Nov 2011, 13:10
by shinobar
13 Nov 2011 v1.3: fix could not unload excess sfs(>6) with some version of busybox, PUPMODE=7(same as PUPMODE=2), error message at unload failure

Posted: Thu 17 Nov 2011, 11:01
by darkcity
added to wiki
http://puppylinux.org/wikka/SFSLoad

also have you seen, SFSdirloader, seems like SFSLoad has a similar function?

http://www.murga-linux.com/puppy/viewtopic.php?t=44601/

Posted: Thu 17 Nov 2011, 11:10
by nooby
Yes but as I remember????? SFSdirloader does not work on all puppies.

I had to do very time consuming manual editing for it to work on my lupu513 that actually are a Snow Puppy 5. And I have forgotten how
to do it again if I would need to add anything.

So I would instead recomment either Shinobar's load on the fly
or Seaside's SFS-Exec which is more manual but that one can fail too
and it has no error checking. One can load same file twice and it would
not complain about it. One have to keep track of what one have loaded
and not using pencil or pen or memory or write it on some notes digitally

Posted: Thu 17 Nov 2011, 11:21
by darkcity
seems like sfsdirloader isn't reliable, have you had good experience with SFSLoad?

may take SFSdirloader off wiki filesystem page here-
http://puppylinux.org/wikka/SoftwareFilesystem

Posted: Thu 17 Nov 2011, 11:46
by nooby
seems like sfsdirloader isn't reliable
To be fair to Seaside he does tell under what circumstances
it does work reliably. His set up at home. So if the one I use
differ then he had no way of testing this at his home :)

So it is a bit of a lottery or know how but it is reliable at his place :)

Nope have only tested these and not the one you mention most likely

Posted: Sun 04 Dec 2011, 14:27
by darkcity
There is a problem with the way some SFSs are packaged

SFS-jre-1.7.0.01-20111027.sfs

and

google-earth_6.2.2074.sfs

see thread:-

http://www.murga-linux.com/puppy/viewto ... 949#586949

The permission setting breaks CUPS, is there a way SFS Load could detect and even work around this problem?

SFS permission

Posted: Mon 05 Dec 2011, 01:50
by shinobar
darkcity wrote:see thread:-

http://www.murga-linux.com/puppy/viewto ... 949#586949

The permission setting breaks CUPS, is there a way SFS Load could detect and even work around this problem?
Tnx darkcity for the report. But maybe nothing the SFS_load can do for this.

Re: SFS permission

Posted: Mon 05 Dec 2011, 02:12
by jpeps
shinobar wrote:
darkcity wrote:see thread:-

http://www.murga-linux.com/puppy/viewto ... 949#586949

The permission setting breaks CUPS, is there a way SFS Load could detect and even work around this problem?
Tnx darkcity for the report. But maybe nothing the SFS_load can do for this.
It also helps to backup/restore your cups permissions. cups_filemanager.pet should work.

Spanish locale for sfs_load

Posted: Sun 25 Dec 2011, 07:12
by ferro10n
I haven't seen a Spanish translation for this app. Just in case, I made one

EDIT :oops:

Checked and had to correct a couple of mistakes, if you use the former file users will believe they're unloading an SFS when they're actually downloading one from the Internet. Please use this file instead.

Re: Spanish locale for sfs_load

Posted: Sun 25 Dec 2011, 09:10
by shinobar
ferro10n wrote:I haven't seen a Spanish translation for this app. Just in case, I made one
Tnx ferro10n. Will be contained in next release.

Calling for other language. Now only have spanish, russian and japanese.

Portuguese locale, corrected Spanish locale

Posted: Mon 26 Dec 2011, 01:08
by ferro10n
Sorry, I had to correct a couple of lines in the Spanish locale :oops: See the corrected version two posts above.

This is the Portuguese translation:

Posted: Sat 21 Jan 2012, 11:43
by PenguinPupLin
Sfs_load-1.3 on-the-fly failed to load the devx and kernel source files for Classic Pup 214X-TOP9, after informing unionfs support is experimental.

sfs_load on 214X

Posted: Sat 21 Jan 2012, 15:47
by shinobar
PenguinPupLin wrote:Sfs_load-1.3 on-the-fly failed to load the devx and kernel source files for Classic Pup 214X-TOP9, after informing unionfs support is experimental.
Tnx for the report. Maybe it does not work on Puppy-4.0 and older.
(There are no report on Puppy 3.x.)

Posted: Fri 03 Feb 2012, 00:01
by jemimah
Here is a small patch for the afterwork to deal with the icon cache and to compile glib schemas. I haven't tested it yet.

Code: Select all

--- sfs_load.orig	2012-02-02 18:48:55.000000000 +0800
+++ sfs_load	2012-02-02 18:57:37.000000000 +0800
@@ -261,6 +261,8 @@
   HAS_SCRIPTS="$HAS_SCRIPTS "$(ls $MNTPNT/etc/init.d 2>/dev/null)
   HAS_STARTUP="$HAS_STARTUP "$(ls $MNTPNT/root/Startup 2>/dev/null)
   HAS_PROFILE="$HAS_PROFILE "$(ls $MNTPNT/etc/profile.d 2>/dev/null) # 28 Oct 2011
+  HAS_ICONS="HAS_ICONS "$(find -L $MNTPNT/usr/share/icons/hicolor -not -type d -printf '/%P\n' 2>/dev/null)
+  HAS_GLIB_SCHEMA="HAS_GLIB_SCHEMA "$(ls $MNTPNT/usr/share/glib-2.0/schemas/ 2>/dev/null)
   if [ -s "$MNTPNT/root/Choices/ROX-Filer/PuppyPin" ]; then
     mkdir -p "$MYTMPDIR"
     cp -f "$MNTPNT/root/Choices/ROX-Filer/PuppyPin" "$MYTMPDIR/PuppyPin-$ROOTNAME"
@@ -275,6 +277,8 @@
   HAS_SCRIPTS=$(echo $HAS_SCRIPTS)
   HAS_STARTUP=$(echo $HAS_STARTUP)
   HAS_PROFILE=$(echo $HAS_PROFILE)
+  HAS_ICONS=$(echo $HAS_ICONS)
+  HAS_GLIB_SCHEMA=$(echo $HAS_GLIB_SCHEMA)
 }
 afterwork() {
   [ "$HAS_MODULES" ] && waitsplash $(gettext "Depmod...") && depmod -a
@@ -328,6 +332,8 @@
       sleep 0.2
     done
   fi
+  [ "$HAS_ICONS" ] && gtk-update-icon-cache -f /usr/share/icons/hicolor/
+  [ "$HAS_GLIB_SCHEMA" ] && /usr/bin/glib-compile-schemas /usr/share/glib-2.0/schemas/
   [ "$GUI" ] && sleep 2
 }
 mergepin() {

Posted: Tue 07 Feb 2012, 18:40
by playdayz
shinobar, Would you say that 1.2 or 1.3 would be better for Lucid 5.2.8? One person (Tman) thought 1.2 might be better. Thanks.

Updating sfs_load

Posted: Wed 08 Feb 2012, 01:16
by shinobar
playdayz wrote:shinobar, Would you say that 1.2 or 1.3 would be better for Lucid 5.2.8?
In any case 1.3 would be better, i suppose.
v1.3 has many fixes from v1.2, and i still have not found new bugs in v1.3.

sfs_load-1.3 on-the-fly

Posted: Wed 08 Feb 2012, 01:42
by RSH
Hi,

i can confirm this "strange behavior" of sfs_load-1.3 like Tman described. But i did not take it as a bug, because i do use my home workstation 100% in live mode (without pupsave) and have been warned before loading the sfs. By using no pupsave at all there is no trouble after reboot. It is a LazY Puppy build in feature.

So, though that strange behavior: to me it seems all fine, but would be better if it would unload the sfs at all times. :)

RSH