Precise Puppy 5.4 and 5.4.2

[watchdog]

The modem was found, the memory stick icon came up on the desktop, but the modem refused to connect.

Wait for rerwin response. Only interested in solving problems and not interested in bad discussions, I personally suggest you this attempt:

Create a /etc/usb_modeswitch.conf.zte with geany with the following content:

########################################################
# ZTE MF622 (aka "Onda MDC502HS")
# ZTE MF626
# ZTE MF628+ (tested version from Telia / Sweden)
# ZTE MF633
# ZTE MF636 (aka "Telstra / BigPond 7.2 Mobile Card")
# ZTE MF637
# and probably others not listed here
#
# Contributor: Joakim Wennergren and others

DefaultVendor= 0x19d2
DefaultProduct= 0x2000

TargetVendor= 0x19d2
TargetProduct= 0x0031

MessageContent="5553424312345678000000000000061e000000000000000000000000000000"
MessageContent2="5553424312345679000000000000061b000000020000000000000000000000"

NeedResponse=1


Then run in console:

usb_modeswitch -c /etc/usb_modeswitch.conf.zte

Run in console:

pupdial

Click Choose and Probe. Insert your dialing config data and try to connect.

[666philb]

this is the AMD Catalyst™ Driver for AMD Radeon™ HD 4000, HD 3000 and HD 2000 Series

AMD has moved to the AMD Radeon™ HD 4000, AMD Radeon HD 3000, and AMD Radeon HD 2000 Series new driver support model. These updates will focus on resolving application specific issues and critical updates. The reason for the shift in support policy is largely due to the fact that the AMD Radeon HD 4000, AMD Radeon HD 3000, and AMD Radeon HD 2000 Series have been optimized to their maximum potential from a performance and feature perspective.
Applicable Products:

This article applies to the following configuration(s):
Hardware:

AMD Radeon HD 4000 Series
AMD Radeon HD 3000 Series
AMD Radeon HD 2000 Series
AMD Radeon HD Series AGP
AMD Mobility Radeon HD 4000 Series
AMD Mobility Radeon HD 3000 Series
AMD Mobility Radeon HD 2000 Series


this driver is only for the above cards (there is a driver for other more modern cards in the PPM)

this driver is for precise puppy 5.4 pae only..

download here

http://www.mediafire.com/file/n6x6okb5d ... 3.2.29.pet 55mb
MD5 e0d26f6f7b84d740cb4b0cf9d5de02fc

[666philb]

'libraw1394' is Marked as installed when it isn't

[rerwin]

One of the most frequent questions in the forum is why the modem doesn't connect. I have a G3 ZTE MF636 USB modem, of the type that looks like a memory stick, and also contains memory, and it is a couple of years old by now. It does not connect.
Why is it, that almost every new puppy have problems with registering and/or connecting this type of modems, neither as modems nor memory? ...
Anyway, PP 5.4 does not find my modem, and does not find it as memory either. It was attached after bootup. In some puppy versions, the modem first registers as a disk drive, and has to be mounted/unmounted before registering as a memory stick/modem, but that did not happen here, either.
What is more serious, after failing to find the modem, it later also failed to find an ordinary memory stick, or my external 160 Gb USB harddisk. Running a manual mount did not help.

tallboy,
Thank you for the pmodemdiag file. I see in it that the device responded to the 3 messages sent to it to cause it to switch to modem mode. The normal result is for the storage device to disappear and the modem device to appear. It disappeared, but the modem never appeared. I believe that the mode switching process does not support co-existence of both the storage and modem devices.

After concocting an experiment on the assumption that you boot with the stick plugged in, I now see that you have stated that you plug it in later. That changes my response. I found in the usb-modeswitch developer's forum a posting by someone with the same kind of issue, who "hotplugs" the device after bootup.

So my first request is for you to boot with it already plugged in. If that works, then I will look into the poster's solution and implement it if necessary.

EDIT: I re-checked the modeswitch posting and see that he fixed a script that is not in puppy. So, trying with the device already in a bootup is key, to verify it can work. The mode switcher seems to do all the right things when you plug it in, but the device does not switch!
Richard

[mavrothal]

Let me try this once more with a bit more info.
Booting from a USB stick that contains initrd.gz, vmlinuz, puppy_precise_5.4.sfs and boot. "boot" is the equivalent of grub/grub2dos for my system where I define PDEV1=sda1 (other parameters like pmedia etc can be defined but the outcome is the same)
After first boot select save to sda1 partition and after reboot sda1 indeed contains:

Code: Select all

drwxr-xr-x 13 root root      4096 Oct 28 20:45 .
drwxr-xr-x 15 root root      1024 Oct 28 04:46 ..
drwxrwxr-x  2 root root      4096 Oct 23 21:30 boot
drwxr-xr-x  9 root root      4096 Oct 29  2012 etc
drwxr-xr-x 15 root root      4096 Oct 29  2012 initrd
-rw-r--r--  1 root root   4187685 Oct 28 20:30 initrd.gz
drwxr-xr-x  5 root root      4096 Oct 29  2012 lib
-rw-r--r--  1 root root 149819424 Oct 23 14:55 puppy_precise_5.4.sfs
drwxrwxr-x  8 root root      4096 Oct 29  2012 root
drwxrwxr-x  5 root root      4096 Oct 17 01:37 usr
-rw-r--r--  1 root root   2641200 Oct 28 20:21 vmlinuz
drwx------  2 root root      4096 Oct 29  2012 .wh..wh.orph
drwx------  2 root root      4096 Oct 29  2012 .wh..wh.plnk

However precise boots again to pupmode 5.
I know that if I use the init from Lucid 5.1, precise boots as expected in pupmode 7.
Trying to figure out why, I run the initrd/init script with "set -xv" but still I can not figure out where the problem might be.
I attach the entire /initrd/tmp here (that includes a more extensive and verbal bootinit.log) in case someone can pick up the problem.

Later After looking at the data, init and woof, realized that I had removed DISTRO_IDSTRING from vmlinuz that tells puppy where the files are.
I use 2 different kernels (in different folders) and one common main.sfs at the root of the device and the PSUBDIR variable (defined by the vmlinuz IDSTRING) does not allow this set up.

[mcewanw]

retroprecise-5.4 NON PAE

Can anyone download from the puppy-precise repo?

All I get is the option from a folder.

No problem downloading/installing packages from the puppy-precise repo. Except for the fact that it's basically empty. :lol:

There's a few packages in "setup","multimedia","graphics" and "internet".

I am running Puppy Precise non PAE on my Pentium M machine. Using a frugal install on a usb ext2 formatted stick. Been running fine for a few days now.

The only issue I have also come across is that, like oldyeller, I also find that packages cannot be downloaded successfully from the puppy precise repo. I also see that there are very few packages in that particular repo anyway as James C has noted but none of these would download. Did you actually successfully download any of these James C?

It is not a major issue for, since personally I tend to download packages directly from repos rather than through PPM, but of course it is good if PPM works too.

All in all, I am finding Precise a good stable distribution.

[James C]

The only issue I have also come across is that, like oldyeller, I also find that packages cannot be downloaded successfully from the puppy precise repo. I also see that there are very few packages in that particular repo anyway as James C has noted but none of these would download. Did you actually successfully download any of these James C?

I successfully downloaded and installed Firefox from the puppy precise repo on Precise non-pae. I generally install pets that I store locally though.

[01micko]

I'm just wondering if this is the IPv4 versus IPv6 issue (oldyeller, mcewanw). What about Ubuntu repos? That would discount the IPv6 theory. (perhaps).

@mcewanw, does this occur in the slacko beta? If so then in both Precise and Slacko remove all "-4" options to wget in /usr/sbin/download_file.

NO.. it's not am IPv issue.. see next page for fix

[cthisbear]

I believe that the responses by tallboy are quite in order.

" I know that I could have dived into the puppy's innards to try
to correct some of these issues, but that should really not be
necessary after so many puppy generations, especially if this
puppy is intended for new puppy users."

Exactly.

" Nope, and that is my point.
Any new puppy should be a relatively mature product by now,
and trouble like failing to get an internet connection up and running
at first try, should simply not happen!

I think that booting, finding all hardware, and covering all possible
ways to connect to the internet, is an absolute minimum requirement
for any distro.

If some other applications don't work because of new kernels, libs
or other dependencies, so be it.
That can be explained and fixed, but only if you are online!

If I came from Windoze, and if Precise was my first attempt to
use puppylinux, the next thing happening would be that all my
friends would hear about that awful linux 'thing' that
couldn't even find the net!

I applaud all his comments....spot on.

:::::::::::

As for my responses on Broadcom B43 issues = Valid.

" When we are downloading a Puppy of this size
expectations are that maybe this time, wireless will work. '

" What is one of the biggest whinges here on Murga..
Wireless...time and time again. "

" The fact is that most newbies downloading Puppy
won't know we offer other Puppies or Puplets that have
fixed this issue. "

""""""""

Don't get me wrong.....
I couldn't be bothered trying any other Linux OS.

But I haven't tried the newer versions of Slacko either.

And many others such as Jejy69, Carolina versions,
Tman, tazoc etc.

But I don't expect to see this continual crapfest, where
people such as peebee and rerwin post fixes

and whoa and betide....nothing happens.

A follow up release needs to happen to address this.
The problem is Barry has got so many projects.
Just to look at his Blog shows what's going on....Plenty.
And this from the >>> retired Barry.
Gotta love him.

""""""

mavrothal

" Look at the time line.
Basically after a look-warm, iffy start, "release" development
started 2 months ago with 4 more betas 2RCs and the final.
Was it enough time?
Looks like was not. "

This would have been better to have been farmed out
to a project co-ordinator....an RC released for 2 weeks or so,
and then the final.

""""""

Yes I was scathing of Lucid and others etc...
but sometimes a little caustic verbosity moves things along.

playdayz final release of Puppy 5.28 05 was impressive.
3 Headed Dog version 2 was great as well.

These solid efforts inspired others to release various spinoffs.

pemasu, in my opinion should have been granted an
Official Puppy release for some his Dpup variants.

No-one likes to be flamed.
But being nice doesn't always get a Fix.

Maybe if I spent less time on the forum helping with continual
wireless issues etc, I might have been involved earlier.

""""""""

emil wrote:

" I would wish a more prominent color for text highlighting.
otherways its superb."

A frustratingly stupid issue.

Never ever in any Puppy release...
and I go back to Puppy 1.0 ........or maybe earlier
have I seen this idiotic behaviour.

::::::::::::

Now as most people including micko know,
I don't discriminate....I hate everybody.
Keep out the middleman I say.

But I love Puppy.
The world's best forum is here.

We have great fixers and fixes.
Genuinely nice and smart people.
A brilliant OS.
People who put their time and money into the project
running our Puppy servers etc.

But sometimes we have to focus and say

Could do better....
Ooops!
Lost the plot.

It happens.
If it was an oversight....
forgetting to add something...that's OK.

But charging ahead ....Need 4 Speed...
ain't good enough.

Maybe if some of you posted onWhirlpool Forums
we would get more testing done earlier.

http://forums.whirlpool.net.au/forum-re ... ?t=1794266

Despite my comments....I loves youse all.

http://marksonsparks.com/jeff-fenech/

Chris.

[artsown]

Replying to mcewanw, James C and oldyeller .... and also just
rambling

I'm evaluating both retro and PAE versions since retro seems
a bit faster and more 'solid' on my PCs.

Like James C, I had success installing Firefox from the PPM Precise
repo but in my case I tried it on the PAE version. Anything I tried
from the PPM Ubuntu repos was a disaster.

If anyone knows a source for wine that runs well on either PAE
or retro please let me know. I found wine-1.3.13-Luci52.pet
somewhere and it just sorta works on Precise, but not well.

Art

[vicmz]

Hello Art,
At the moment I'm using this Wine SFS:
http://distro.ibiblio.org/puppylinux/pe ... 6-sfs4.sfs
It works well for me in Precise 5.4 (I'm playing plants vs zombies), maybe it works for your setup, too.
Cheers

[James C]

Update...... just booted the non-pae Precise pfix=ram and now I'm also unable to download from the puppy precise repo.See screenie. No problem with Ubuntu repos though.

[James C]

No problem in the regular Precise 5.4.....

[01micko]

Ok James, that sheds some light for me, I'll come up with a hot fix shortly.

(it's to do with the "retroprecise" prefix, 99% sure)

It just so happens that I have it on the usb stick on the lappy I'm using now..

EDIT

ok, here is the DIFF

Code: Select all

--- /usr/local/petget/downloadpkgs.sh.old	2012-10-29 10:18:55.192350888 +1000
+++ /usr/local/petget/downloadpkgs.sh	2012-10-29 10:08:30.000000000 +1000
@@ -57,6 +57,7 @@
  [ "$REPO_DEFAULT_SUBSUBDIR" == "noarch" ] && OFFICIAL_REPO='quirky' #120313
  [ "$REPO_DEFAULT_SUBSUBDIR" == "gap6" ] && OFFICIAL_REPO='quirky' #120515
  [ "$DISTRO_NAME" = "Precise Puppy" ] && OFFICIAL_REPO='quirky' #120908 see /etc/DISTRO_SPECS
+ [ "$DISTRO_NAME" = "RetroPrecise Puppy" ] && OFFICIAL_REPO='quirky' #120908 see /etc/DISTRO_SPECS
  [ "$DISTRO_NAME" = "Precise Arm Puppy" ] && OFFICIAL_REPO='quirky' #120908
  [ "$DISTRO_NAME" = "Wheezy Arm Raspbian Puppy" ] && OFFICIAL_REPO='quirky' #120908
  [ "$DISTRO_NAME" = "Squeezed Arm Puppy" ] && OFFICIAL_REPO='quirky' #120908

REMEMBER.. Barry did warn that retro is unofficial...

[James C]

Success. Downloading Firefox as I type.

[tiangeng]

HP M1005
Use "Xsane image scanner",then selected "USB",but the "prescan" window not appear。

[rerwin]

Here is an experiment to utilize jemimah's improvement to dhcpcd that apparently reduces the occurrence of the loss of wireless connections where there is a weak signal. If the feature works so well with Frisbee, why not add it to the other network managers!

This requires the installation of one of the "dhcpcd-dropwait" packages I have posted on page 6:
http://www.murga-linux.com/puppy/viewto ... 566#660566

If you do not have the "dhcpcd-dropwait" package installed, networking will not work. /tmp/dhcpcd.log will document the failure as due to an invalid option.

I hope that any of you needing to use Frisbee because of the annoying disconnects with the other managers will try this experiment with either or both of the standard network managers. Others might try it to verify that wireless networking still works as usual. But my goal is to verify that random disconnects are reduced to the level you see from Frisbee.

Note that for the experiment, the delay is hard-coded as 15 seconds. Please tell me if it needs to be longer. TIA.

Technical note: I added the "-j" option to the wireless invocations to start dhcpcd in the sns rc.network and sns scripts, and the wag-profiles.sh script. Is there any place else I should modify?
Richard

[Sage]

Because I'm not a user of SM, I don't test it very often. However, after a change of HW on the bench, I used SM to run streaming TV on retro from a liveCD. Everything worked well for ~10mins, then it crashed the entire machine with long lines of code displayed. Couldn't capture these because hard reset was required. Continued with Slacko & FF. Just an observation, unable to provide any further data. Anyone else found this? Will revert to FF, Opera (already done and OK), Midori and Chrome.

[mavrothal]

This would have been better to have been farmed out
to a project co-ordinator....an RC released for 2 weeks or so,
and then the final.

I do not think is fair to "blame" Barry for that.
Is not his "fault" that no one stepped up to the plate.
He asked and as a matter of fact suggested candidates denounced the "offer" for various personal reasons.

Now regarding sufficient testing of the build, with 6-7 pupplets developing simultaneously the handful of testers that test almost everything, can test that much devoting maybe a day per pupplet.
So a "final" is destine to get more users and thus bugs.
A usual recommendation for very cautious users and any software is: "wait till update 1"
Let's not forget that Lucid went through a dozen of "finals" before it reached its maturity.

There is also an inherited problem with puppy that does not allow one puplet to take full advantage from the others. In reality the freedom to build puppy from any distro and any kernel is like building 5-6 distros in parallel! And slackware packages may not run in Ubuntu.
Currently in puppyland we have T2, Ubuntu 10.04 and 12.04, Slackware 13.7 and 14, a couple of Debian versions, some that I missed and an even greater variety of kernels.
They are destined to fix the same bugs over and over...

[nubc]

Running retroprecise 5.4 as live CD, text highlighting in SM 2.13.1 is non-existent (transparent). Weak highlighting will also impact important settings, eg, proper use of gparted. Where do I adjust the highlight color? [Selected gtk theme "Flat-grey-rounded" to fix this.]

Default wallpaper is white? upup wallpaper looks good with icons.

I see a VLC 2.0.3 package in ubuntu-precise-universe repo. Including some 67 missing dependencies. the vlc install will use 107 MB. I installed VLC with PPM, but there is no evidence that it is working after restarting X server.

Nice to have Transmission in the original app set, because my Wary 5.1.1 doesn't have any torrent app. So this CD will be worth keeping. (Had to use Puppy 2.14x to get the retroprecise ISO.)