Puppy 4.3.1-Release-Candidate2 -- bug reports

Message

BarryK · #1 Post by **BarryK** » Wed 14 Oct 2009, 10:04

Blog Announcement:

http://puppylinux.com/blog/?viewDetailed=01169

Download:

http://puppylinux.com/test/puppy-4.3.1-rc2/

Of course, the usual thing. If you have been testing 4.3.1-RC1 be sure that there is no 'pup-431.sfs' left behind on the hard drive.

Ah, I just got a pm, Ecube has posted a fix for Partview. It now puts up a message if no partitions mounted. Attached. Gunzip it and copy it into /usr/local/Partview/.

eprv · #2 Post by **eprv** » Wed 14 Oct 2009, 14:43

Hi,
I like the frugal install to ext 3 with save to disk option (not save to file)
I know it was problematic in the past
but there is no warning now during installation so tried again
however attempting to do that, works until I try to load an sfs file
than it crashes during next boot:
setting up the layered file system... fail
Fatal: module pci ................... not found
................
pup_ro4 is overlapped
it ends with kernel panic

I had to write down the messages as the system could not load so it may be not exact

rerwin · #3 Post by **rerwin** » Wed 14 Oct 2009, 14:51

I have consulted with zigbert about pfind and pburn vanishing after starting up on my 400 MHz PC..

rerwin:
I changed that [pburn] value from 2 to 4 seconds and that works. I also changed the pfind value from 2 to 5 and that works, too. I would add at least 1 more second to each just to be safe. The pfind value depends on the size of the hit list

zigbert:
I have put your sleep values + 1 into the scripts. - Thanks a lot.

However, the fix has not appeared in the new pburn 3.1.2 and pfind was not updated. I am concerned that these simple but important fixes will be left out of our "prime time" Puppy 4.3.1.

Here is the affected part of /usr/local/bin/pfind:

Code: Select all

#Start loop to kill PID when window closes.
#This is a workaround for a bug in gtkdialog. Works only in Xorg.
sleep 3
while [ A != B ] && [ "`which xwininfo`" ]; do
	TMP="`xwininfo -stats -name Pfind`"
	[ ! "$TMP" ] && TMP="`xwininfo -stats -name Pfilesearch`"
	if [ ! "$TMP" ]; then
		#run second test to ensure that window wasn't just rerendering during the first check
		sleep 2    <=======This is line 12 - change to 6=====
		TMP="`xwininfo -stats -name Pfind`"
		[ ! "$TMP" ] && TMP="`xwininfo -stats -name Pfilesearch`"

It is the line 12 sleep value that should be "6". In /usr/bin/pburn, the corresponding line is 11 and the value "5" But I feel both numbers could be slightly larger to remove all the risk on slow PCs.

Can these be included at this late date, to avoid flaky-appearing behavior on the older PCs? Thank you zigbert for helping me with this, but I don't want it left out of 4.3.1.

Are there any other tools that use the same technique to work around a gtkdialog bug, that could benefit from a similar change?
Richard

toutix · #4 Post by **toutix** » Wed 14 Oct 2009, 19:07

431rc2 behaves same way as 431rc1:

-booting from usb stick stops on "Loading drivers needed to access disk drives....." with HP/Compaq mini 110c
-booting flawlessly from same usb stick on older IBM X41
-booting flawlessly from external usb CD rom drive also on 110c
-filesystem used on usb stick is ext2

Thanks for puppy!

MinHundHettePerro · #5 Post by **MinHundHettePerro** » Wed 14 Oct 2009, 21:53

431RC2, Res. 800x600:
fdd, 7+1 int. hdd part., (2 non-detected SCSI int. hdd) CD, DVD, 2x8 part. ext. USB hdd.
HotPup of 2.14R didn't stack drive icons

Ok, I don't normally run with all these partitions mounted, but still ...

No worries on my behalf, though, I know that this problem is not so prevailing with higher res's

.

/
MHHP

davids45 · #6 Post by **davids45** » Thu 15 Oct 2009, 00:28

G'day,
The untidy stacking of drive icons is also a problem for high res. monitors if there are multiple-partitioned multiple hard-drives, as well as lots of optical drives and usb drives.
As noted by MinHundHettePerro, this stack-up problem has been fixed in quite a few Puplets (eg ttuuxxx's 2 revisited & MU's NewYears), but not in the basic Pups.
One annoying aspect of this iconic stack-up is, when installing from the Live CD to such a well-endowed system, to mount the optical drive with the Live CD so as to locate the installation files, it is not always easy to see the right icon (eg sr0) in the stack.
While it's OK once you know where things are and what to do, this untidiness could be a problem for newcomers to Puppy who have the latest and greatest hardware, as well as those with older hardware and smaller screens.
David S.

Jim1911 · #7 Post by **Jim1911** » Thu 15 Oct 2009, 02:05

I have tried unsuccessfully to make a frugal installation on an ext4 partition. I was able to make a full hd installation on an ext4 partition.

Cheers,
Jim

r__hughes · #8 Post by **r__hughes** » Thu 15 Oct 2009, 02:27

Just one small glitch in an otherwise perfect new Partview (in first & sercond versions of the perl script)

My 15.01G sda1 drive is wrongly reported to be 16G
My 16,0G sda2 is correctly reported as 16G

see picture below

Kal · #9 Post by **Kal** » Thu 15 Oct 2009, 02:58

This is on the service_pack-430-to-431.

I noticed that on the symbolic links to the libraries in /usr/lib, their copies not links as they should be. This just uses more space is all (no big deal). Being space conscious, you might like to know.

Kal

PaulBx1 · #10 Post by **PaulBx1** » Thu 15 Oct 2009, 03:19

Found a strange message in dmesg, not present in Puppy 412:

Code: Select all

evdev.c(EVIOCGBIT): Suspicious buffer size 511, limiting output to 64 bytes. See http://userweb.kernel.org/~dtor/eviocgbit-bug.html

I looked at that web page and it said among other things, "You need to upgrade your userspace programs to get rid of this message."

Also noticed that while Puppy412 mounts ext3 filesystems in ordered data mode, in 431 it is in writeback data mode. Not sure if the difference is significant, or if it was intended.

Tui · #11 Post by **Tui** » Thu 15 Oct 2009, 04:46

Problem as as reported with V4.3.1-RC1 - Please read

Still there! This is being sent via Live CD v4.3.1-rc2 all in ram

So what's wrong at my side, if all is well with you guys ?

Tui

zigbert · #12 Post by **zigbert** » Thu 15 Oct 2009, 14:53

Sorry rerwin

My mistake

I have changed the sleep values in these packages.

svgt · #13 Post by **svgt** » Thu 15 Oct 2009, 15:59

I have no success with the roaring penguin, If I drive the setup all is ok.

But when I try to start the connection I get the following error:

pppoe-start_shell /usr/bin/pppoe-start: line 191: 3381 terminated
$CONNECT "$@" >/dev/null 2>&1

I use a pcmcia PC-Card Level-One FPC-0106TX Card Bus Adapter

Regards
Svgt

BarryK · #14 Post by **BarryK** » Thu 15 Oct 2009, 22:14

MinHundHettePerro wrote:431RC2, Res. 800x600:
fdd, 7+1 int. hdd part., (2 non-detected SCSI int. hdd) CD, DVD, 2x8 part. ext. USB hdd.
HotPup of 2.14R didn't stack drive icons

Ok, I don't normally run with all these partitions mounted, but still ...

No worries on my behalf, though, I know that this problem is not so prevailing with higher res's .

/
MHHP

431RC2 has new code for placing those icons. Have you compared with 431Rc1, 430, 412, or 411?

cthisbear · #15 Post by **cthisbear** » Thu 15 Oct 2009, 22:47

Barry:

Have you seen this.
Many users would love this feature.

http://www.murga-linux.com/puppy/viewtopic.php?t=47838

" If you've been missing WICD or NetworkManager, this pet is for you.

Wpa_gui runs in the background and automatically
connects you to any open wireless network it finds.
Encrypted Network?
No problem, just hit scan, double click your network,
and enter the password.
Choose file->save, and you'll be automatically connected
whenever you're in range. "

Regards........Chris.

MinHundHettePerro · #16 Post by **MinHundHettePerro** » Fri 16 Oct 2009, 08:38

BarryK wrote: 431RC2 has new code for placing those icons. I'm aware of that, hence my reporting this issue again. Have you compared with 431Rc1, 430, 412, or 411? 412, 430pre-a - 431RC2, I tried most, but not all, of the development versions, all with the same stacking feature.

/
MHHP

BarryK · #17 Post by **BarryK** » Fri 16 Oct 2009, 10:11

RC2 is now "closed".

I'm uploading the final, will announce it when finished uploading.

Denethor · #18 Post by **Denethor** » Fri 16 Oct 2009, 12:23

Excuse the language but I am using a translator!

I have a problem with puppy linux already present in the version 4.30 network!
I have a Packard Bell Dot S with Atheros 8132 PCI-E Ethernet network adapter. When I try to connect the OS recognizes me automatically atlc1 driver if I am not mistaken. Tells me that all ok is never then when I go to do the car configuration DHCP not detects me suffered the setting the makes only after a little waiting and eventually fails the network! I think it is a bug in the drivers because sometimes obliterated and in the process of recovering the same driver was able to make it work but I cannot now!

panzerpuppy · #19 Post by **panzerpuppy** » Sat 17 Oct 2009, 01:13

Puppy 4.3.1 FINAL is out.

dawg · #20 Post by **dawg** » Sun 18 Oct 2009, 14:31

Funny, I haven't even noticed this thread until now!
Maybe the mods could pin these kinda threads during the time of their actuality (and unpin them later)?
It'd make finding relevant things much easier, I believe.

And forgive me for ranting now, but is it just me, or do other people too happen to inadvertently miss stuff on phpBB forums?

Personally, I find all the "eye candy" rubbish to be more of an "eye killer" and a distraction from the relevant stuff (e.g. TEXT). Annoying.

And I guess perhaps I should make a thread and repost this, for more people to notice and consider

(old)Puppy Linux Discussion Forum

(old)Puppy Linux Discussion Forum

Puppy 4.3.1-Release-Candidate2 -- bug reports

Puppy 4.3.1-Release-Candidate2 -- bug reports

frugal install to ext3 save to disk

pfind and pburn unusable on slower PCs

boot stops when trying to load drivers

New Perl partview rounds integers up

USB - Rundisk Flash Drive