Puppy 4.2 "Deep Thought" - Bugs & Fixes
Update to recent modem improvements
I have applied some lessons learned from my 3G-pupdial project, to speed up the udev rule, handle new Intel device names, and and wait a little longer for modem response to a modem test. On initial use, my Intel modem was not found after it had been detected; increasing the wait for response by a half second solved that. This dotpet is for puppy 4.2 only.
UPDATE 5/17/09: Replaced the attachment with a correction to the Intel 536 modem init script. The corrected tarball alone is available here:http://www.murga-linux.com/puppy/viewto ... 348#306565
Richard
UPDATE 5/17/09: Replaced the attachment with a correction to the Intel 536 modem init script. The corrected tarball alone is available here:http://www.murga-linux.com/puppy/viewto ... 348#306565
Richard
- Attachments
-
- modem_fix_to_4.2-3.pet
- Changes to Intel modem scripts, udev rules, modemtest. Puppy 4.2 only.
- (7.97 KiB) Downloaded 382 times
Last edited by rerwin on Sun 17 May 2009, 22:47, edited 1 time in total.
- Colonel Panic
- Posts: 2171
- Joined: Sat 16 Sep 2006, 11:09
Patchy mounting of partitions in Puppy 4.2
Hi. Sorry to have to report that detection of my partitions is very patchy, especially of sda8 (the final one).
I get this error massage when I try to do it manually;
mount; /dev/sda8 is not a valid block device
and clicking on the Pmount icon simply results in the partitions up to sda6 being mounted.
I don't have the same problem with earlier Pups such as 2.15 CE and Teenpup. Can anyone please tell me what the fix for this is? Ttuuxxx has suggested that this could be a driver error.
Best,
Colonel Panic.
I get this error massage when I try to do it manually;
mount; /dev/sda8 is not a valid block device
and clicking on the Pmount icon simply results in the partitions up to sda6 being mounted.
I don't have the same problem with earlier Pups such as 2.15 CE and Teenpup. Can anyone please tell me what the fix for this is? Ttuuxxx has suggested that this could be a driver error.
Best,
Colonel Panic.
Re: tiny core repos
I would NOT have thought by now, after all years ofmcewanw wrote:ftp://distro.ibiblio.org/pub/linux/dist ... corelinux/
dottce's are just tar.gz archives; that's what you want since you can make them straight into dotpets (as you know).
thought and design we still need or want changes in
anything about packaging!
Take this with lots of humor, but I'm serious.
Looking at the URL you posted, it seems to me using
tinycorelinux we'd be able to know and learn in short
order who the screw up packagers are, compared to
the good ones.
Do you really think Puppy is ready for people
signing off on their packages by using their name or
handle?
Before closing, can I sell you on a Firefox pet package
that requires gnome support files? Even when
Puppy doesn't have them and Firefox doesn't need
them!
- ttuuxxx
- Posts: 11171
- Joined: Sat 05 May 2007, 10:00
- Location: Ontario Canada,Sydney Australia
- Contact:
Re: tiny core repos
[quote="Bruce
Before closing, can I sell you on a Firefox pet package
that requires gnome support files? Even when
Puppy doesn't have them and Firefox doesn't need
them![/quote]
All I can say Bruce, Is that one time way back I made a couple Firefox packages that needed Gnome libs because they were simply repackage tar.gz ones with the font changes etc included, It was a bit of a learning curve to learn how to properly compile firefox for puppy, most Firefox/Seamonkey/Opera etc packages in the Software section today are just repackaged ones unless the user specifies that he/she compiled them, which I think i might of seen 2-3 times other than my stuff, they do that for the ease of pet installs, Opera of course you can't compile. but the rest you can. I would say anything from Firepup mozilla based that I released on the software section of the forum has been compiled on puppy.
Also on series 4 some applications out of the many I compiled would n't compile properly, this could be due to the faulty gtk thats in 4 series. But when that would happen, most of the time you could get Ubuntu Gutsy compiled applications to work, also I don't know of any kde base that was compiled on puppy, most of it was reused from other distros, I've probably compiled more gnome base parts than anyone who released gnomes libs on puppy, and let me tell you, it isn't fun, lol
Also sometime when say I compiled xfburn I needed dbus, but after it was compiled I was able to remove it, since it for the xfce desktop, which I didn't use but deps for you, anyways after that I had dbus installed on my pc, any programs that could use dbus automatically switch to dbus enabled, that crap because I then have to start over clean, or thing like Firefox would be build with dbus.
Hope this explains a bit.
ttuuxxx
Before closing, can I sell you on a Firefox pet package
that requires gnome support files? Even when
Puppy doesn't have them and Firefox doesn't need
them![/quote]
All I can say Bruce, Is that one time way back I made a couple Firefox packages that needed Gnome libs because they were simply repackage tar.gz ones with the font changes etc included, It was a bit of a learning curve to learn how to properly compile firefox for puppy, most Firefox/Seamonkey/Opera etc packages in the Software section today are just repackaged ones unless the user specifies that he/she compiled them, which I think i might of seen 2-3 times other than my stuff, they do that for the ease of pet installs, Opera of course you can't compile. but the rest you can. I would say anything from Firepup mozilla based that I released on the software section of the forum has been compiled on puppy.
Also on series 4 some applications out of the many I compiled would n't compile properly, this could be due to the faulty gtk thats in 4 series. But when that would happen, most of the time you could get Ubuntu Gutsy compiled applications to work, also I don't know of any kde base that was compiled on puppy, most of it was reused from other distros, I've probably compiled more gnome base parts than anyone who released gnomes libs on puppy, and let me tell you, it isn't fun, lol
Also sometime when say I compiled xfburn I needed dbus, but after it was compiled I was able to remove it, since it for the xfce desktop, which I didn't use but deps for you, anyways after that I had dbus installed on my pc, any programs that could use dbus automatically switch to dbus enabled, that crap because I then have to start over clean, or thing like Firefox would be build with dbus.
Hope this explains a bit.
ttuuxxx
http://audio.online-convert.com/ <-- excellent site
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
http://samples.mplayerhq.hu/A-codecs/ <-- Codec Test Files
http://html5games.com/ <-- excellent HTML5 games :)
- Colonel Panic
- Posts: 2171
- Joined: Sat 16 Sep 2006, 11:09
Wine - can't run
I have just completed a frugal install to a 2Gb RunDisk, and
all went well until I loaded WINE 1.16 - I used the (dot)pet
version.
The screen has gone black with just the mouse pointer on screen.
Any help for this one ?
Cheers Tui
all went well until I loaded WINE 1.16 - I used the (dot)pet
version.
The screen has gone black with just the mouse pointer on screen.
Any help for this one ?
Cheers Tui
Java kills printing in Deepthrough (4.2)
When Java is installed (by .pet, or manually from the SUN homepage), printing is killed:
CUPS is not accessible anymore (can not connect to localhost...), and also printing itself (HP Laserjet 4L on lp0) doesn't work anymore.
Some posts before in this blog I read about some trouble with printing, and tried the "libgnomeprint-2.18.5-full.pet", and it seemed to work (first installed Java, then the "libgnome.." - but only until I enabled Java in Openoffice.
Any ideas?
Mafae
CUPS is not accessible anymore (can not connect to localhost...), and also printing itself (HP Laserjet 4L on lp0) doesn't work anymore.
Some posts before in this blog I read about some trouble with printing, and tried the "libgnomeprint-2.18.5-full.pet", and it seemed to work (first installed Java, then the "libgnome.." - but only until I enabled Java in Openoffice.
Any ideas?
Mafae
My default machine is a IBM Thinkpad R50, 1.4GHz with 512MB RAM.
Before posting here I also tried on other machines like a old 650MHz 512MB DELL Laptop, and on a High-performing Dual-core with 2GB RAM, all with the same effect: As soon a Java is installed, printing doesn't work any more (and CUPS ends up in un-accessible localhost...). Also I tried different Java versions (1.6 and 1.5).
The effect with OpenOffice (Oo3 as sfs-file) is just on top, when trying to workaround with the "libgnomeprint-2.18.5-full.pet".
On my default installation (4.1.2) everything works fine.
Before posting here I also tried on other machines like a old 650MHz 512MB DELL Laptop, and on a High-performing Dual-core with 2GB RAM, all with the same effect: As soon a Java is installed, printing doesn't work any more (and CUPS ends up in un-accessible localhost...). Also I tried different Java versions (1.6 and 1.5).
The effect with OpenOffice (Oo3 as sfs-file) is just on top, when trying to workaround with the "libgnomeprint-2.18.5-full.pet".
On my default installation (4.1.2) everything works fine.
Try to manually restart CUPS with the command
Are you accessing CUPS through Menu > System > Manage Printing?
Code: Select all
/etc/init.d/cups start
sound
Hello all:
Downloaded latest version of deeper thought x2 with bug fixes. The sound does not work on my IBM laptop T22 - sound = cs46xx . The sound does not play however. I think that it may be a kernel loading problem. I as there is sone error that flashes on the screen during alsa config saying that something cs46xx is in use by something else/
I load TMX shard from the CD and the sound works from the boot with no configuration. This is why I'm thinking that it may be a kernel issue.
Thx All,
Vito
Downloaded latest version of deeper thought x2 with bug fixes. The sound does not work on my IBM laptop T22 - sound = cs46xx . The sound does not play however. I think that it may be a kernel loading problem. I as there is sone error that flashes on the screen during alsa config saying that something cs46xx is in use by something else/
I load TMX shard from the CD and the sound works from the boot with no configuration. This is why I'm thinking that it may be a kernel issue.
Thx All,
Vito
sound
Hello All Again:
I fixed the sound by copying the modules folder, modprobe.conf and the modules.conf files to overwrite the deeper thought equivalents, and rebooted; now there is sound again in puppy.
So I guess this is not a kernel problem as I had first thought, but some failure to config on boot and in subsequesnt tries by alsaconf
Thx,
Vito
I fixed the sound by copying the modules folder, modprobe.conf and the modules.conf files to overwrite the deeper thought equivalents, and rebooted; now there is sound again in puppy.
So I guess this is not a kernel problem as I had first thought, but some failure to config on boot and in subsequesnt tries by alsaconf
Thx,
Vito
java kills CUPS printing in 4.2
Tnx for the hint! Restarting CUPS manually works. Also I meanwhile experimented a bit, and the effect became more clear:Try to manually restart CUPS with the commandAre you accessing CUPS through Menu > System > Manage Printing?Code: Select all
Code: /etc/init.d/cups start
- Started with new blank 4.2 installation (frugal): CUPS works fine
- Installed "jre1.6.0_11.pet":
- CUPS was killed immediately after JRE-pet installation finished.
- Manual re-start works.
- After installation of JRE, CUPS-script does not start automatically anymore.
Cheers,
Mafae
Mafae, try this installer, yours sounds faulty. Either the.pup I posted or the later one from Ttuuxxx. If you use the one from Ttuuxxx, just download the tar.gz, extract it and run the installer. This has been 100% for me............http://www.murga-linux.com/puppy/viewtopic.php?t=41098
Spup Frugal HD and USB
Root forever!
Root forever!
@WhoDo:
Thank you for including the JQ8flat icon theme in 4.2.1.
I don't know this should say a bug.
I notice that "mount" icon is not included in the JQ8flat theme.
I made the "mount" icon for the Desktop and also made the "refresh" icon for the menu panel.
If you could, I would like you to include these icons.
Regards,
JebaJQ8
Thank you for including the JQ8flat icon theme in 4.2.1.
I don't know this should say a bug.
I notice that "mount" icon is not included in the JQ8flat theme.
I made the "mount" icon for the Desktop and also made the "refresh" icon for the menu panel.
If you could, I would like you to include these icons.
Regards,
JebaJQ8
- Attachments
-
- drives48.png
- (591 Bytes) Downloaded 2194 times
-
- refresh24.png
- (233 Bytes) Downloaded 2212 times
I'm happy to do that, JebaJQ8. Thank you for the excellent theme!jebaJQ8 wrote:I made the "mount" icon for the Desktop and also made the "refresh" icon for the menu panel.
If you could, I would like you to include these icons.
[i]Actions speak louder than words ... and they usually work when words don't![/i]
SIP:whodo@proxy01.sipphone.com; whodo@realsip.com
SIP:whodo@proxy01.sipphone.com; whodo@realsip.com
Java kills CUPS
Hi,
after some further testing, the CUPS-problem became more clear:
It has nothing to do with Java or Openoffice, it is a matter of how CUPS is accessed:
Menu>Setup>CUPS Printer Wizard
Works fine and reliable.
Menu>System>Manage Printing
Kills CUPS nearly always!
After manual restart of CUPS it works fine again.
So it seems rcrsn51 was on the right way, when he asked
after some further testing, the CUPS-problem became more clear:
It has nothing to do with Java or Openoffice, it is a matter of how CUPS is accessed:
Menu>Setup>CUPS Printer Wizard
Works fine and reliable.
Menu>System>Manage Printing
Kills CUPS nearly always!
After manual restart of CUPS
Code: Select all
../etc/init.d/cups start
So it seems rcrsn51 was on the right way, when he asked
[/b]Are you accessing CUPS through Menu > System > Manage Printing?
-
- Posts: 5464
- Joined: Fri 10 Jun 2005, 05:12
- Location: Australia
Because I have never bothered to download Puppy 4.2, I only learned about this problem with Broadcom wifi devices recently:
http://www.murga-linux.com/puppy/viewtopic.php?t=34159
This is a major error. If I had been told about this before the decision was made, I would have advised against it.
Quite a number of the extra drivers I provided are alternative drivers, to deal with specialised hardware variations or special situations. They will compete with the standard kernel drivers, so my driver packages include a modules-preference configuration (/etc/rc.d/MODULESCONFIG) which will force Puppy to use the new driver in preference to the standard one.
This will cause problems where the old driver is known to work better than the new one ... as in the case of Broadcom wifi devices in anything other than Dell or HP computers.
"b43:wl" and "b43legacy:wl" should be deleted.
"bcm43xx:wl" should be changed to "bcm43xx:ssb"
That's because the new Broadcom b43/b43legacy modules, which work in conjunction with the ssb module, should be used in preference to the deprecated bcm43xx module. The firmware in Puppy is compatible with b43/b43legacy, and is NOT compatible with bcm43xx.
The full and correct PREFLIST entry (from Puppy 4.1.2) is
Unfortunately (groan) there are also reports that Puppy 4.2 fails to automatically uncompress the b43 firmware, but hopefully this may be fixed once the preferences are corrected.
I now attach the correct MODULESCONFIG with PREFLIST entries. This is to replace trio's "No_bcm43xx_patch.pet", which is wrong.
Please use this in the next 4.2 update.
Dear oh dear. It seems that the developers of 4.2 included many, if not all, of the extra 4.1.x drivers I posted heretrio wrote:In 4.2 Final..bcm43xx is still being a problem...for those of you who wants a workaround..just edit /etc/rc.d/MODULESCONFIG
http://www.murga-linux.com/puppy/viewtopic.php?t=34159
This is a major error. If I had been told about this before the decision was made, I would have advised against it.
Quite a number of the extra drivers I provided are alternative drivers, to deal with specialised hardware variations or special situations. They will compete with the standard kernel drivers, so my driver packages include a modules-preference configuration (/etc/rc.d/MODULESCONFIG) which will force Puppy to use the new driver in preference to the standard one.
This will cause problems where the old driver is known to work better than the new one ... as in the case of Broadcom wifi devices in anything other than Dell or HP computers.
Unfortunately, that's only half-correct. As well:trio wrote:In 4.2 Final..bcm43xx is still being a problem...for those of you who wants a workaround..just edit /etc/rc.d/MODULESCONFIG (line 33) delete the red word
PREFLIST=' rt2500usb:rt73usb ath5k:ath_pci orinoco_nortel:hostap_plx orinoco_plx:hostap_plx orinoco_tmd:hostap_plx orinoco_pci:hostap_pci bcm43xx:wl b43:wl b43legacy:wl ssb:wl prism54:p54pci r8169:r8101 rtl8187:r8187 '
"b43:wl" and "b43legacy:wl" should be deleted.
"bcm43xx:wl" should be changed to "bcm43xx:ssb"
That's because the new Broadcom b43/b43legacy modules, which work in conjunction with the ssb module, should be used in preference to the deprecated bcm43xx module. The firmware in Puppy is compatible with b43/b43legacy, and is NOT compatible with bcm43xx.
The full and correct PREFLIST entry (from Puppy 4.1.2) is
Code: Select all
PREFLIST=' rt2500usb:rt73usb ath5k:ath_pci orinoco_nortel:hostap_plx orinoco_plx:hostap_plx orinoco_tmd:hostap_plx orinoco_pci:hostap_pci martian_dev:ltserial bcm43xx:ssb prism54:p54pci r8169:r8101 '
I now attach the correct MODULESCONFIG with PREFLIST entries. This is to replace trio's "No_bcm43xx_patch.pet", which is wrong.
Please use this in the next 4.2 update.
- Attachments
-
- broadcom-preflist-fix.pet
- for Puppy 4.2 only
- (1.45 KiB) Downloaded 895 times