(old)Puppy Linux Discussion Forum

hey micko

successfully installed 5.4.0.3 to a troublesome acer 8935g, first problem encountered is a lack of battery monitor - advice please....

regards

Has anyone else had any problems with unloading SFS files via SFS_Load that you get to through the Setup app's Install tab? I get dumped out from the desktop to a black screen with lots of stuff in white text. Only way out appears to be to hit the reset button. On rebooting, the SFS I tried to remove will still be listed.

It's possible to get round it via the somewhat inelegant method of nuking the SFS from XP (or pfix=ram boot of Puppy) and then selecting the unload option once back in the main Puppy session.

The pupsave was created in 5.4.0.3 and is 256MB (IIRC) ext2 and on the NTFS hard drive. I have yet to test to see if it also happens on the FAT32 memory stick.

Jades wrote:Has anyone else had any problems with unloading SFS files via SFS_Load that you get to through the Setup app's Install tab?

I have a problem getting them to load via the boot manager....as a sanity check, I look in /etc/rc.d/BOOTCONFIG to see what is set to set to be loaded at boot.

Since the boot manager doesn't show what's loaded, you can't use it to unload the sfs files.

Thom

Hi 01micko,

The first indication of this was the updating of the layered file system at boot up followed by the loss of my customised desktop icons. (fixed with desktop template)
Aarf first reported this behaviour with 5.4.0.2 & upup precise372 a few weeks back.
http://puppylinux.info/topic/slacko-540 ... 1-jan-2013

In my case an sfs that I had at one point loaded during testing (but no longer present in my home sub-directory) continued to show as available to load in boot manager but was in fact mounted, a copy of the relevant sfs in
another sub-directory off my home drive having loaded. Re-naming that to .xfs prevented it loading upon a reboot but it continued to show in boot manager as available to load. Attempting to load it in "load on the fly" offered the option to clear it out and upon rebooting again, normality seemed to be restored. I noted however that it still appeared in /etc/rc.d/BOOTCONFIG.save and so edited it out. (It was no longer listed in BOOTCONFIG)

Given Aarf's comments it seems like a recent new woof bug.

Bugs in bootmanager are caused by the Xdialog change, it is reverted for the next version. I don't think anything changed with bootmanager at all.

Sometimes the system can crash when unloading an sfs on the fly, I have seen it. I'd like it fixed but shinobar isn't around much these days and I haven't much time myself. Should I ditch sfs_load? No, I don't think so. You can recover with ctrl-alt-F2 and perform an immediate clean shutdown. If it doesn't get fixed then it will be documented as a "known issue".

Also, the icon bugs has been around since ever, Barry tried to fix it once and failed, that's why I made desk-setup. It's a decent work-around but also needs documenting in the "know issues".

RE: sfs unload problem

To everyone who have encountered the issue - which kernel version you use?

Here's procedure which reproduces sfs_load hang every time (at least for me) and the symptoms are: a) SFS_load hangs on "Processing... Wait a moment..." window; b) inability to shut down the system.
1. Load devx.sfs
2. Reboot
3. Unload devx.sfs -> hang

sfs_load --debug shows that the culprit is 'losetup -d'.
Also, the following line proves that some problem was foreseen for k3.2.x. but I'm on k3.4.17 and it happens too, so I simply replaced -eq with -ge and the problem has gone, hopefully without any harm for overall functionality...

BTW, it's worth to note, that in sfs_load-1.9.3 (the one in Slacko-5.3.3) all lines with 'losetup -d' are commented out.

HTH
Greetings!

Thanks again SFR!

We'll try that in the next version and this probably explains the "sometimes" in my earlier post. (k3.2x v k3.4x). Very scientific of me , eh?

Failing that we do have a fall back. FatDog's load_sfs is reported to be very reliable in other pups too, so we can go that way as well.

ally wrote:hey micko

successfully installed 5.4.0.3 to a trouble acer 8935g, first problem encountered is a lack of battery monitor - advice please....

regards

vattery, I'll make a package soon

thanks micko

I've been struggling to get puppy to run on that acer and thought it was a compatibility issue - my bad......

it's currently running 5.3.3 as it's a windows machine on ntfs so went with a 'known' product in the end, will upgrade later

much much quicker than his win7 64 install

[/u]

SFR wrote:RE: sfs unload problem

To everyone who have encountered the issue - which kernel version you use?

3.4.17-PAE (Slacko 5.4.03 and CPU is Pentium D 3.4.GHz).

SFR wrote:Here's procedure which reproduces sfs_load hang every time (at least for me) and the symptoms are: a) SFS_load hangs on "Processing... Wait a moment..." window; b) inability to shut down the system.
1. Load devx.sfs
2. Reboot
3. Unload devx.sfs -> hang

I tried a similar procedure with a JRE SFS originally downloaded from this forum. I ended up at the screen shown on the attached image. CTRL-ALT-F2 didn't do anything and I had to hit the reset button.

Hey Jades

Although symptoms you're encountering are much more "violent" than mine, could you apply the modification (that I have proposed before) to /usr/sbin/sfs_load script, to see if our issues have the same origin?
It's line 1801, after the change should look like this: Save the modified script and try to unload sfs then.

If this won't change anything, most likely there's something more...

Greetings!

Fresh manual frugal install of 5.4.0.3 on this powerful P3.Display,sound and internet connection all working and correct on initial boot.Seems to be a bit slow but I normally use my Retro builds and Ttuuxxx's 214X on this thing.Slower but still usable.

# report-video
Report Video 1.4 - Wed 20 Feb 2013 on Slacko Puppy 5.4.0.3 - Linux 3.2.33-4g i686

Chip description:
1.0 VGA compatible controller: Intel Corporation 82810E DC-133 (CGC) Chipset Graphics Controller (rev 03)
oem: Intel810(TM) Graphics Chip Accelerated VGA BIOS

X Server: Xorg Driver used: intel

X.Org version: 1.12.3
dimensions: 1024x768 pixels (321x241 millimeters)
depth of root window: 24 planes

...# report-video
Report Video 1.4 - Wed 20 Feb 2013 on Slacko Puppy 5.4.0.3 - Linux 3.2.33-4g i686

Chip description:
1.0 VGA compatible controller: Intel Corporation 82810E DC-133 (CGC) Chipset Graphics Controller (rev 03)
oem: Intel810(TM) Graphics Chip Accelerated VGA BIOS

X Server: Xorg Driver used: intel

X.Org version: 1.12.3
dimensions: 1024x768 pixels (321x241 millimeters)
depth of root window: 24 planes

...the above also recorded in /tmp/root/ as report-video,
and archived with xorg.conf and Xorg.0.log as report-video-full.gz
# glxgears
21 frames in 5.1 seconds = 4.092 FPS
27 frames in 5.2 seconds = 5.238 FPS
28 frames in 5.1 seconds = 5.542 FPS
30 frames in 5.1 seconds = 5.866 FPS
30 frames in 5.1 seconds = 5.938 FPS

-Computer-
Processor : Pentium III (Coppermine)
Memory : 253MB (148MB used)
Machine Type : Physical machine
Operating System : Puppy Linux
User Name : root (root)
Date/Time : Wed Feb 20 01:15:41 2013
-Display-
Resolution : 1024x768 pixels
OpenGL Renderer : Gallium 0.4 on softpipe
X11 Vendor : The X.Org Foundation
-Audio Devices-
Audio Adapter : ICH - Intel 82801AA-ICH

-Processor-
Name : Pentium III (Coppermine)
Family, model, stepping : 6, 8, 6 (Pentium III/Pentium III Xeon/Celeron/Core Duo/Core Duo 2)
Vendor : Intel
-Configuration-
Cache Size : 256kb
Frequency : 730.94MHz
BogoMIPS : 1462.78
Byte Order : Little Endian
No wonder it's slower, using swap with just the browser open.........

firefox-19.0 is available in the updates manager.. (see the Setup menu... you can cheat if it hasn't updated yet by deleting /root/.packages/Packages-slackware-14.0-patches and re-starting updates manager, and get it from the Utah repo, others will catch up very soon).

The cool thing is you can open a pdf from your local media, or anywhere else. Looks nice and crisp too. I have been off mozilla as of late.. but with opera moving to webkit that will surely double it's size in the near future.

Have fun!

Hi @James C

James C wrote:No wonder it's slower, using swap with just the browser open.........

This is an interesting comment. Having swap, you've done a good thing is protecting that your system wouldn't quickly lock as you increase system workload. But I am curious because of an observation I've made about 32bit PUPs I/O subsystem usage when running "pfix=RAM".

If you turn off SWAP, viacould you notice any difference in behavior of the P3 PC? Just curious of what you might observe.

Thanks in advance.

ally wrote:...first problem encountered is a lack of battery monitor - advice please....

Try this pet of pemasu's .. http://www.smokey01.com/pemasu/pet_pack ... -0.0.1.pet

If that works ok I'll get the source and compile it for the repo.

thanks micko

will report back tomorrow

01micko wrote:firefox-19.0 is available in the updates manager

Appears to be working well.......

I just compiled the latest chromium-25.0.1364.97.

25 has just gone stable. It *should* run on i586 (Athlon-XP, pentium-2) and you will notice there are far less errors logged due to being compiled against true slackware dependencies. (as with the other chromium-24.xxxx that I compiled at ibiblio and mirrors). I'll upload it soon, but as usual, ibiblio is giving troubles. EDIT: uploaded, see below

See http://www.chromium.org/developers/calendar for the current status of chromium. A handy reference.

Download

http://distro.ibiblio.org/puppylinux/pe ... 86-s14.pet

checksum <-- 3b2f13e9b6376298a3433971da99d672 chromium-25.0.1364.97-i486-s14.pet

PET REUPLOADED with fixed start script

gcmartin wrote:Hi @James C

James C wrote:No wonder it's slower, using swap with just the browser open.........

This is an interesting comment. Having swap, you've done a good thing is protecting that your system wouldn't quickly lock as you increase system workload. But I am curious because of an observation I've made about 32bit PUPs I/O subsystem usage when running "pfix=RAM".

If you turn off SWAP, via

Code: Select all

swapoff /dev/...

could you notice any difference in behavior of the P3 PC? Just curious of what you might observe.

Thanks in advance.

I already knew what would happen but........ Slacko 5.4.0.3 frugal install with 256 mb ram and swap off.

Pretty obviously got much,much slower...... almost too slow to use.
Just took about 30 seconds to open the terminal to run "free".

Went ahead and posted the above because I'm expecting a freeze-up at any moment.
Opened a video in another tab.....don't know if it's the low-ram issue but cannot play a Youtube video......repeated flash crashes.Going to turn swap back on and try the video again.