(old)Puppy Linux Discussion Forum

always receive the message.

:s

mama21mama wrote:

FATAL: could not load /lib/modules/2.6.30.5/modules.dep: no such file or directory.

my problem

Possibly You have an incorrect "modules.dep" file. Run "depmod".

Reaperz wrote:Any one ever get this message?
"Intel chipset detected. However, 915resolution was unable to determine the chip set type."
After running the Everest utility in Windows, this is what it found for the chipset:
North Bridge - Intel Whitney i810
South Bridge - Intel 82801AA

If screen resolution is correct, ignore this message or delete "/etc/resolutionfix"
file.

Yet one bug in /etc/rc.d/rc.update script at line 235:
xmesssage "Dependencies may be missing..................etc..."
____^^^ xmessage with 3 "s"-es

PANZERKOPF wrote:Yet one bug in /etc/rc.d/rc.update script at line 235:
xmesssage "Dependencies may be missing..................etc..."
____^^^ xmessage with 3 "s"-es

That bug is still there in the latest Woof! Have now fixed it, so fix will be in Quirky 1.1.

I have now added a supplemental package (modem_ti_qc_ipwi_updates-6.1.pet) in an attempt to support all of the modems (actually, their driver modules) using the ttyUSBn device name that I know of. If you know I have missed one, please tell me. The dotpet is with the others that upgrade modem support, on page 16, here: http://www.murga-linux.com/puppy/viewto ... 235#388235

The support for the IPWireless PCMCIA modems has been using the wrong device name, but I have corrected that, to ttyIPWp0, etc.

I need the hear from anyone trying those modem types with pupdial, so I can know whether further attention is needed to make them work. Thanks for whatever help you can provide.
Richard

Just had a problem with a full install 4.3.1

The usb power has turned itself off after a crash (brownout)
fsck came up ok except for the usual blocks errors

you start the pc, it's all ok
boot into the one partition, mouse light goes off, usb probing runs and it just stops at the <prompt> #
I can manually start xwin ok and with the direct kbd/mouse all works

shutdown and reboot
it all starts again (like in a loop)

any sugestions folks before I blow it all away

thanks
scsijon

topic : http://www.murga-linux.com/puppy/viewto ... 963#260963
seems to relate

Have you tried removing the flag file in /tmp? IIRC it is named /tmp/bootcnt.txt.

?if your talking to me pizzasgood, then no, and it's a zero length file created when I just booted up, will give it a go though.

i've just found I can have a full and frugal install on the one partition and am using that to run with (2 grub entries though)

thanks
scsijon

Yet one ounce about rc.update :
There are string (in checking procedure) which calls "/usr/local/petget/findinstalledpkgs.sh"
Puppy 4.3.1 hasn't such a file.
Also the xmessage cannot work (even after correcting), says "cannot open display 0:0". A possible reason of that behaviour is chrooted environment, under which this script is running.

scsijon wrote:?if your talking to me pizzasgood, then no, and it's a zero length file created when I just booted up, will give it a go though.

that didn't work, eventually backed up all of root, had a list of what I had added (each machine and partition I have has their own record) so reformatted. I had to change & format to ext3 and back to ext2 for it to let me use the partition so something must have corrupted somewhere in the partition settings or the partition itself.

Reinstalled from scratch and everything came ok, copied root back, all is ok (so far).

scsijon

I have a network interface that has been treated a couple of different ways by the Network Wizard.

The Network Wizard on Puppy 4.3.1 identifies the interface as
pci: D-Link System DFE-690TXD CardBus PC Card.

Looking at TuxMobil I can see that I'm going to be limited to systems that use the 2.4 kernel.

When the interface was treated as eth1 it works.

When the interface was treated as wlan0 using ndiswrapper, it does not work.

How can I make the network wizard forget about the wlan0 alternative, and go back to the eth1 alternative? (I don't know what I did that made it switch.)

Added June 2, 2010: Somehow the interface reverted to eth1 again, and now I have network connectivity again. It would still be nice to know what made the interface change and then change back.

no, problem back again, (usb power off) and the error seems to be:
waiting for modules to complete loading usb 1 2 3 4 5 6 (etc)

seems likely that something is not loading?

Does anyone know what driver turns on / off the power to the usb?
and
where in what script starts it?

please

scsijon

found another topic relating to the problem
http://www.murga-linux.com/puppy/viewto ... 963#260963
could a dev, when they have some time , have a look please.

Not a bug but incorrect working only.
There is a script "gv" in /usr/bin directory which is designed to view the postscript files with epdfview. It converts a ps file into pdf and sends converted file to epdfview via pipe.
The problem is that some converted files are detected by epdfview as encrypted, even they are not encrypted. I don't know a reason of this problem but if we use a temporary file instead of pipe then this problem not occurs.
I attach modified file here.

Starting Roaring Penguin can lead to "Address Not Found" errors while browsing and the "Too many levels of symbolic links" error message because of the circular link created when Roaring Penguin makes /etc/resolv.conf a symlink to /etc/ppp/resolv.conf, which is already a symlink to /etc/resolv.conf.

Previously I added a post to an existing thread (started by warpinwolf two years ago) which discussed this bug. Perhaps that was not the proper place to post. Since the name of that thread mentioned Dingo it is possible that it was considered ancient history and was overlooked. So I'm adding this post to the 4.3.1 thread to make it clear that the bug still exists in Puppy 4.3.1.

For details and a suggested fix (which is actually a fix previously made by Barry that somehow got lost on the way from Puppy 2 to Puppy 3), see my post at Dingo, PPPOE, No DNS.

I have found a fix for modems with the ID 19d2:2000 that should allow them to switch into modem state. I put it into a small package, attached. This should be helpful for aracnachid's Vodafone Vodem Zte K3565z USB modem and Whitesnow's Onda MM515UP modem. To give due credit, I derived this fix from information in the USB ModeSwitch Forum.
Richard

look at http://www.murga-linux.com/puppy/viewtopic.php?t=48019 from patriot
contains a number of video drivers INCLUDING a vesa driver that seems to work properly, no funny screen sizes or shifting right or down.

absolute magic

thanks
scsijon

Not exactly a bug... but I have a old vid card that doesn't work unless I choose VESA, instead of the XORG.

It's an 8MB Matrox G200.

Because I have to use VESA, video playback is unacceptably slow. Before I disabled it, I had better playback with my MoBo's 1MB integrated video chipset...

Reaperz wrote:Not exactly a bug... but I have a old vid card that doesn't work unless I choose VESA, instead of the XORG.
It's an 8MB Matrox G200.
Because I have to use VESA, video playback is unacceptably slow. Before I disabled it, I had better playback with my MoBo's 1MB integrated video chipset...

Try following steps:
1) Check file /usr/X11R7/lib/xorg/modules/drivers/mga_drv.so.
2) Open /etc/X11/xorg.conf file with text editor.
3) Change string Driver "vesa" to Driver "mga" in section "Device".
4) Restart X server.

Subject: filemnt won't mount two files with same base name.

To prevent mounting the same file twice (and the the dangers inherent in doing so), filemnt compares the name of the file with the names of files already mounted. Once upon a time, it appears that filemnt simply compared the entire pathnames (which include the directory names), but because one or more of the directories in the pathname could be symlinks, two different pathnames could lead to the same file. So, at least in Puppy 4.3.1, filemnt was improved to compare only the base name of the file.

That improvement prevented mounting the same file twice, even if one of the directory names was a symlink. So far, this is good.

The bug is that if you happen to have two different files with the same base name, as would be the case if, for instance, you were comparing one custom pup-431.sfs with another. You get something similar to the following xmessage which is quite puzzling because it simply isn't true.

Sorry /mnt/sr0/pup-431.sfs is already mounted and in-use by Puppy

Also, if the filename itself (not a directory name) is a symlink, then filemnt will allow the file to be mounted twice.

So the current test gives both false positives and false negatives.

I am attaching a version of filemnt in which I use the stat utility to uniquely identify a file by its device and inode values. This should eliminate both problems.

I have also expanded the error message that appears when you do try to mount a file that is already mounted. Now it gives not just the pathname of the file you tried to mount, but also the mount point of the mounted file and the pathname that was used when it was mounted.

(Note: I am reporting this in the Puppy 4.3.1 thread because 4.3.1 is the Puppy that I have. The bug could certainly exist in older and newer puppies.)