Lucid Puppy 5.0 - Bug Reports and Bug Fixes

[01micko]

hello COD11

As you say, the minor errors should be suppressed.

I will look into your fixes to points 2 and 5 and make a recommendation to BarryK so these issues are fixed upstream in the Woof build system, the Puppy builder.

On a rethink, I think issue 5 is now dealt with. If you are interested you may like to try the beta for the next version in this series.. luci-205

6) is probably something either in /root/.xinitrc or /usr/sbin/delayedrun that isn't installed and probably should be sent to /dev/null also upon an error being returned.

Thank you.

Cheers

[COD11]

Hi,
here is my suggestion for function "format_size_func()" in "pup_event_frontend_d" to prevent localization problems.

Stand-alone Test version:

Code: Select all

#!/bin/sh

format_size_func() {
 [ ! $SIZE ] && SIZE=0
 DN=0 # DN=1 done
 DP=0 # DP=1 comma
 [ `printf "%.1f" | grep ',' | wc -l` -ne 0 ] && DP=1
 
 TB=1073741824 # 2^30
 [ $SIZE -ge $TB ] && f_size $TB $DP "TB"
 [ $DN -eq 1 ] && return
 
 GB=1048576 # 2^20
 [ $SIZE -ge $GB ] && f_size $GB $DP "GB"
 [ $DN -eq 1 ] && return
  
 MB=1024 # 2^10
 f_size $MB $DP "MB"
 [ "$SIZE" = "0.0MB" -o "$SIZE" = "0,0MB" ] && SIZE="0"
}

f_size (){
 DN=1	
 SIZE="`LC_ALL=C dc $SIZE $1 div p`" # Fault simulation
 #SIZE="`dc $SIZE $1 div p`" # Original code
 echo "SIZE_RAW2=$SIZE" #TEST
 [ $2 -ne 0 ] && SIZE=`echo $SIZE | tr '.' ','`
 SIZE="`printf "%.1f" $SIZE`$3"
}

# TEST VALUES
#SIZE=51
#SIZE=52
#SIZE=3422334
#SIZE=356148770
SIZE=2698092316

echo "SIZE_RAW1=$SIZE"
format_size_func
echo "SIZE=$SIZE"

Best Regards
COD11

[buford]

ok i went through 14 of the current 18 pages of this thread and didnt find the fixes so here it goes.

so far running smoother than puppy 4.3 awesome

the bugs i found on my acer one netbook

vlc launches and disapears
but a bigger problem i cannot get pidgin to work it wont even launch tried many versions of it and the old purple link fix to no avail

[buford]

delete dual post

[playdayz]

01micko posted a new pidgin in one of the Lucid 5.1 threads. I will see if I can find it.

[Béèm]

01micko posted a new pidgin in one of the Lucid 5.1 threads. I will see if I can find it.

This one?

[tnah]

Hi all. first off, thanks for all the work put in to make and improve this software.

I've had a version of Linux installed in a partition for more than 10 years, so I can't claim to be a newby. However, until I found Puppy (4.1, about 18 months ago), I'd never actually used Linux, as oppsed to just getting it set up and then playing with it for a while until I couldn't find a way to do something that I needed to do.

Anyway, I've been running Puppy 4.1 on my MSI VR321 (VIA C7-M 1.6G processor, 2G RAM, 13.3" screen) laptop, mostly happily. One thing that I was not able to do was to get a good output to an external screen (I tried DRTs, LCRs, and a projector - all same problem) - the image on the external screen was too large, so that the right and bottom sides were cut off. Since this was the case for the pre-operating system start-up messages as well, and since I was also unable to fix it using a Windows installation, I had thought that it was a BIOS problem. Not being able to fix it there either, I had become tho think that it was a hardware fault - anyway, I had resigned myself to being unable to use an external screen or projector.

When Puppy 5.0.1 came along, I ran this from the CD with an external screen attached, and - to my great and happy surprise - got a good picture on it! Great - thanks!

However, the down side is that with 5.0.1 the sound doesn't work (actually, it's there, but almost inaudibly faint).

I tried ALSA sound wizard - if this changed anything, it didn't improve it.

I thought that, since I had 4.1 installed to the hard drive, I would be able to compare 5.0.1 with 4.1, and work out what the difference was, and hopefully fix the problem. So, I ran ALSA sound wizard in my 4.1 installation, to have a look at what sound chip it detected. As a consequence of doing this, I have buggered up the sound on my hard drive 4.1 installation!

For the record, ALSA on 4.1 gave the diagnosis: hda-intel 1106:3288 (rev 10), whereas running this in 5.0.1 said Via Technologies Inc . VT 8 /A [Azalia HDAC] (Via High Def). As the sticker on the computer says it's got a VIA processor, I expect that the diagnosis from 5.0.1 is closer to what is actually in there.

Another installation of 4.1 (and of 4.3.1) from CD has sound OK.

So, my line of thinking is that 5.0.1 has done, on set-up, the [wrong] thing that ALSA sound wizard did when I ran it on the 4.1 installation. I thought that I might be able to spot the crucial difference by comparing /etc/modprobe.conf files, but have not been successful, and I an\m now totally out of my deapth (as I said before, I can't claim newby status, but I'm still clueless).

There are a couple of other issues that I have with 5.0.1, but I'll describe them in another post.

Thanks if anyone can point me in the right direction, and if I get the wound problem fixed, I'll report the story back.

[smokey01]

tnah try alsamixer from a terminal window. Fiddling with some of the settings may help.

eg:

# alsamixer

[Stripe]

My epson stylus s21 printer that works fine in puppy 4.3.1 (installed as a epson stylus s20) will not work with 501 no matter what I install it as. have tried using pipslite-1.4.0.pet and installing as epson lite, cups thinks the printer is working and shows no error messages, but nothing from the printer at all.
Apart from that everything works well for me
(love to see grub for dos added as makes a frugal install easy)
Hope this helps and thanks to everybody for the effort they have put in
Well Done
Don

edited to say:
does not work with 209 either

edited again to say:
Problem with gutenprint as I have installed the 5.2.3 pet package and everything works fine now.

[tnah]

Thanks smokey01 - alsamixer in a terminal window does the job!

[rerwin]

While using grubconfig to set up grub, it stops after I enter the device for the grub files, with the message:

The partition '/dev/sda6' is mounted read-only, either unmount it or mount it read-write.

But it is actually mounted as read-write. When I enter the command that triggers that message, I see:

# echo "fooboo" > /dev/sda6/fooboo
bash: /dev/sda6/fooboo: Not a directory

which causes the statement on grubconfig line 987 to always test false:

if echo "fooboo" > $BOOT/fooboo 2>$TMP/null ; then

In googling the problem, I found this:

The 7 Deadly Linux Commands

If you are new to Linux, chances are you will meet a stupid person perhaps in a forum or chat room that can trick you into using commands that will harm your files or even your entire operating system. To avoid this dangerous scenario from happening, I have here a list of deadly Linux commands that you should avoid.

. . .

5. Code:

any_command > /dev/sda

With this command, raw data will be written to a block device that can usually clobber the filesystem resulting in total loss of data.

Apparently, the recent kernels now actively prevent that possible disaster -- and even extend that to partitions.

Barry, would you consider a different way to verify that grubconfig can write to the partition in question? It appears that one cannot now write to a /dev/ node, or at least one that is a partition.
. . .I also tried that on my 4.3.1 system and got the same result! What gives?
Thanks.
Richard

FOLLOW-UP: Semi-false alarm! The known workaround for this is to be sure the partition is not mounted, at all. I was running from a CD, but with its contents and a pupsave already on the partition, which was mounted as /mnt/home. I was fooled by the implied option (in the error message) that mounting as read-write was acceptable. When I booted from the CD with pfix=ram, the partition did not get mounted, allowing me to complete configuring grub. So, the only issue here is whether the error message is completely accurate. It could cause frustration for a new user.

Sorry for the alarm. I initially thought the problem was new in Lupu, but then found out otherwise.
Richard

[tnah]

I have lost the volume control applet on the task bar. It has (in my experience of the last couple of weeks) been very flaky (compared to that in Puppy 4xx), and, while fiddling with it, have lost it, and can't work out how to get it back. I think lost it when, after right-mouse-click, I got a menu and chose 'exit' from it.

After this happened, I tried to invoke alsamixer from a terminal windows (as I had done before, at the suggestion of smokey01, successfully to fix an earlier problem). I got "command not found".

Then, ran ALSA Sound Wizard from the menu (something that, in the past, had buggered up my sound). It went through its probing, predictably, after that, sound production and recording didn't work, but alsamixer was again invocable from the command line.

I've got sound working again, but I still haven't recovered the volume control button on the task bar. I'd like to get it back. Even better, would be to have the corresponding applet that was with Puppy 4.

Any suggestions gratefully received ...

[smokey01]

Using ROX, navigate to /usr/bin and click on retrovol

[Hattmannen]

Hi! I'm new to Puppy Linux and tried the live cd a couple of days ago. Found it to work really well and went on with a frugal install to an ext2 partition. All went well and I can now dual boot with Win XP. So far so good.
However, there seems to be a problem with freeing up memory. I didn't notice a t first and the first few boots seemed to run as it was supposed.
I've installed Firefox on my AMD Sempron 3000+ notebook with 384MiB RAM and opened up two tabs. One with a youtube video that I let play in the backgroud, the other with facebook. Before opening firefox I had aproximately 170 MiB RAM free (can't remember the exact number). After using said setup for about 10 minutes I checked the free RAM again (using the console command free) and found it to be down to only 54MiB.
Fine I though. Browser+Flash eats memory. I closed Firefox and checked the memory again. No change. Still only 54MiB of free ram.
I went on to create a 256MiB swap partition and rebooted.
Lo and behold. With nothing other than a console window opened I had now 54MiB of free RAM. The swap file is unused though (as in free), but is loaded correctly while booting.

I couldn't find any similar reports on the forum, but I'm sorry if I report a known problem.
Any suggestions to what might have gone wrong? Or is it simply a bug?
Cheers and thanks for a superb distro!

[nitrofurano]

hi - i'm having a grub error 18 when booting from hd after installed

the machine is a pentium3 @450mhz with 6gb hd, and 64mb ram (which runs puppy 5.0 live-cd perfectly)

i did a full install internal ide, and this error appeared as far i installed grub on root or mbr
where from can i find solution for this?

overally, i followed the steps from http://www.brighthub.com/computing/linu ... 4.aspx?p=2

thanks!

[smokey01]

However, there seems to be a problem with freeing up memory. I didn't notice a t first and the first few boots seemed to run as it was supposed.
I've installed Firefox on my AMD Sempron 3000+ notebook with 384MiB RAM and opened up two tabs. One with a youtube video that I let play in the backgroud, the other with facebook. Before opening firefox I had aproximately 170 MiB RAM free (can't remember the exact number). After using said setup for about 10 minutes I checked the free RAM again (using the console command free) and found it to be down to only 54MiB.
Fine I though. Browser+Flash eats memory. I closed Firefox and checked the memory again. No change. Still only 54MiB of free ram.
I went on to create a 256MiB swap partition and rebooted.
Lo and behold. With nothing other than a console window opened I had now 54MiB of free RAM. The swap file is unused though (as in free), but is loaded correctly while booting.

Hattmannen, I think you may be confusing memory with the pup_save file free space. At the bottom right hand side of your screen you will see something like the the attached graphic. The little square that displays, in my case 1.1G free, has nothing to do with memory or RAM. It is letting you know how much free space you have in your pup_save file. As you said above you installed puppy frugally. This means a pup_save file was created. All of your settings and additional software you installed are inside the pup_save file by default unless you manually placed them elsewhere. As your pup_save file diminishes you can goto Menu>Utility>Resize personal storage file to increase it. It's not a good idea to make it too big but I have had some quite large ones over the years without problems. It's generally a good idea to keep documents and photos outside of the pup_save file. I also keep my emails outside too which means I can use a number of different distros and use the same mail files. I always use seamonkey to maintain compatibility.

Hope this helps.

Smokey

[Hattmannen]

I think you may be confusing memory with the pup_save file free space. At the bottom right hand side of your screen you will see something like the the attached graphic. The little square that displays, in my case 1.1G free, has nothing to do with memory or RAM. It is letting you know how much free space you have in your pup_save file.

Hi! Sorry if I was unclear. The bottom right square saying "###M free" is the space left in the pup_save-file. I gathered as much from the forum, but thanks anyway.

To check how much free RAM I had avalible I opened up a console window and used the command free. I.e.
# free

/../After using said setup for about 10 minutes I checked the free RAM again (using the console command free) and found it to be down to only 54MiB.

Sorry if it was unclear. I only used that command when checking memory usage.
Here's the figures as I am writing now:

Code: Select all

               total           used           free            shared    buffers
Mem:       384868       340460        44408            0         7308

This is while running only Firefox and a urxvt console window.
When closing firefox free memory goes up to about 70MiB.
This confuses me since Puppy only consumes about 128MiB of RAM when I boot it with parameter pfix=ram, thus loading everything into RAM.

Thanks for the help anyway. Any other suggestions are welcome!

[smokey01]

Hattmannen, I use Seamonkey 2.0.5 mail and browser. When both are loaded I lose about 55676 of memory using the free command. I'm guessing but I'm sure the puppy files are all loaded into memory when using a frugal install, if you have enough memory. One of the reasons puppy runs so fast. I guess there would also be cached files in memory.

[Zarniwoop]

Hello!

I wanted to try Puppy 5, so I created a new Virtual Machine in VirtualBox. I've already run 4.2.1 in VirtualBox without problems, but Puppy 5 doesn't seem to work. I booted from the iso-file, and I got two errors related to the same thing. It seems that it has a problem with the cpu in the Virtual machine.

I got the error both at home using a current version of VirtualBox running on Mandriva 2009.1, and at work running version 2.1.0 of VirtualBox in Ubuntu 8.04. The last message I got at home was "PMU disabled", but it never installed.

What I do is I boot up from the iso. I get to the prompt and it starts to boot, but then gets stuck after saying that the CPU is bad/unsupported. I think at home it called the cpu "number 23".

On both PCs I have a Core2Duo CPU, but different versions of the Core2Duo. The one at home is 2.53Ghz, and at work T74000 2.16Ghz.

[Zarniwoop]

I'm trying now using Qemu instead of VirtualBox (on the same E7400 2.16Ghz cpu computer, running Ubuntu 8.04), and I get the same problem. I doesn't show the error message, but it starts booting but nothing happens (the PMU I got before was with "loglevel=7" so that might be the reason I don't get an error now).