Page 7 of 15

Audio CD

Posted: Fri 11 Jan 2013, 02:57
by gulk
tlchost wrote: It is my understanding that older puppies did show an icon for music cds. Is there any way to incorporate that feature in 5.4?
I think in the "Desktop->Desktop Drive icon manager", there is an option to check that is unchecked by default:
"Tick box for auto-launch 'handler' when media plugged"

It might or not create an icon on the desktop, but at least it will launch your default player with the music CD.
tlchost wrote: Slacko is great...thanks heaps for your efforts!
Indeed, this is the best Puppy I've tried so far. The attention to details is very impressive, considering the wide array of hardware it supports right out of the box.

PS: I had my mom over for Christmas; she came with her laptop and I thought, what the heck, let's try to introduce her to Linux using Slacko 5.4, without believing she would buy into it... Guess what, she's hooked now. She is extremely computer challenged and doesn't know the difference between an application and an OS... Well, you guys made it so easy to transition to your distribution, there is no excuse to not show it off anymore!!!

PPS: I made my mom run on a USB flash drive, but that prevent the laptop from going into "suspend" mode, because of the following in the /etc/acpi/actions/suspend.sh script:

Code: Select all

# do not suspend if usb media mounted
USBS=$(probedisk2|grep '|usb' | cut -d'|' -f1 )
for USB in $USBS
do
	mount | grep -q "^$USB" && exit
done
I had to manually comment these lines out, and now the laptop goes to sleep and resumes from sleep without issue, by clicking the hard power button, then pressing the "susupend" button.
Would you be able to detect if the Puppy runs from a USB stick and not apply this check in that case?

PPPS:
in the /etc/acpi/actions/acpi_poweroff.sh script, there is a typo:

Code: Select all

    <label>Susupend</label>
should be

Code: Select all

    <label>Suspend</label>

I'm very impressed. You bested yourself once more, 01Micko et al.
Gulk

Posted: Mon 14 Jan 2013, 11:05
by Master_wrong
during install into usb flashdisk there is syslinux error
possibly unsafe /tmp/ permissions
fix
chmod 1777 /tmp
then run installer

Posted: Mon 14 Jan 2013, 11:52
by 01micko
Perms are already 1777 on /tmp. You may have installed a bad package.

Posted: Mon 14 Jan 2013, 13:18
by SFR
@Master_wrong & @01micko

That's strange, but in my case /tmp sometimes has 1777 and sometimes 777.
Eg. in my install (pmedia=ataflash on internal SATA HDD) it had 777 by default and I couldn't make it to preserve 1777 after reboot.
(/initrd/pup_ro2/tmp has 777 as well, perhaps that's why?)
I simply added chmod 1777 /tmp to rc.sysinit to achieve this.

On the other hand, when I switch to pmedia=atahd, /tmp suddenly becomes 1777 and, in addition, it appears as a mount point.
(I checked this a few minutes ago in VBox, 5.4-PAE)

Also, pristine boot without a savefile also gives /tmp 777.

IIRC I noticed this also in 5.3.3, but I didn't report it, because usually I have "luck" of experiencing "it's_just_me" kind of issues...

@Mick:
Any progress regarding savefile deadlocks? Have you been able to recreate the problem, at all?

HTH
Greetings!

Posted: Mon 14 Jan 2013, 14:23
by Master_wrong
@sfr and @micko

I'm testing the firefox version, pristine boot/no save file. trying to install to a usb flasdisk.

thanks

Posted: Mon 14 Jan 2013, 20:14
by 01micko
Ah, @ SFR and master_wrong, I think I found the source of this issue. Thanks!

SFR: I can't reproduce the savefile deadlocks! I'm wondering if you can capture a log. Say, boot pfix=ram after the issue occurs and grab /var/log/messages from the affected savefile.

Thanks

Re: Two jwmconfig2 bugs and suggested fixes reported elsewhere

Posted: Mon 14 Jan 2013, 20:36
by 01micko
npierce wrote:01micko,

You're welcome.
01micko wrote:Actually I am doing a full rewrite of all the stuff in jwmconfig2, Joe is releasing a new version soon (he's up to r635 now) and some things have changed.
Thanks for keeping jwmconfig2 up to date with JWM (as well as all the other work you do supporting Slacko).
I think I have to reverse that decision (full rewrite) because the i18n guys have just spent a lot of time translating these files.I will just fix what breaks and try to include some backwards compatibility.

Posted: Mon 14 Jan 2013, 20:42
by 01micko
gulk,

There is a reason I have disabled suspend on USB installs, this is a very generic script and it's designed to avoid RAM crash because USB installs run entirely in RAM until the scheduled save occurs, but if it works for you then that's fine, maybe someone else will find that useful. Thanks for the typo fix.

Users may find that installing acpid from Slackware repo may fix suspend issues as they have many rules for many different machines. YMMV though as I have no need to test. Reports needed. Use at own risk.

Posted: Mon 14 Jan 2013, 23:02
by SFR
01micko wrote:SFR: I can't reproduce the savefile deadlocks! I'm wondering if you can capture a log. Say, boot pfix=ram after the issue occurs and grab /var/log/messages from the affected savefile.
Ok, to avoid some influences from my daily savefile contents, here's what I did:
1. Booted with pfix=ram and didn't change any settings.
2. Rebooted, creating 1GB slackosave.2fs on NTFS partition (sda2).
3. After reboot:
a). cat /dev/urandom > /root/bigfile until ~400MB
b). Clicked 'Save' icon (reminder: I have pmedia=ataflash in menu.lst) and I had "luck" on the first try: after ~150MB it started its deadlock->save_few_MB->deadlock->save_few_MB->... cycle.
After a while all opened ROX windows stopped responding, but I had terminal opened, so I killed cp process, and after ~1.5 min. everything went back to normal.

But like I said earlier, sometimes it's conversely - at the first time everything's ok, and then after 4th-5th try the deadlock appears - can't predict...

I attached /var/log/messages, but nothing new appeared there after clicking 'Save'.

BTW, regarding the "direct" method to recreate the issue (cat /dev/zero > /initrd/pup_ro1/root/bigfile):
I noticed that if the first attempt continuously goes good, it's worth to change output filename a couple of times (bigfile1, bigfile2, etc.), then even get back to one of older names.
That increases the chance to catch the deadlock.

If I could recreate the problem in 5.3.3 I would suspect my hardware then, but anyway I'll try it out also on my Acer, one of these days...
[Update: Acer AO 722, similar scenario (just bigger, 4GB savefile) - deadlock on the second try, nothing new in .../messages.]

For the time being, the echo 1 > .../dirty_(background)_ratio workaround in 9/10 cases still saves my days...

EDIT:
Almost forgot - I don't know if this matters, but some time ago I noticed the following difference between 5.3.3 and 5.4 in mount parameters:
In 5.4 is:

Code: Select all

/dev/loop1 on /initrd/pup_ro1 type ext2 (rw,noatime,errors=continue,user_xattr,acl)
In 5.3.3 there's no user_xattr neither acl.
I wasn't able to "erase" them, so I did remount with nouser and noacl, but it didn't make any difference really.
(my knowledge regarding mount parameters is minimal, btw)

Thank you for your interest &
Greetings!

Posted: Tue 15 Jan 2013, 23:59
by gulk
01micko wrote:gulk,

There is a reason I have disabled suspend on USB installs, this is a very generic script and it's designed to avoid RAM crash because USB installs run entirely in RAM until the scheduled save occurs, but if it works for you then that's fine, maybe someone else will find that useful.
I understand the reason now. Then maybe the script should give the option to force save (with optional check box, like "remember for next time") to savefile, then allow to suspend. If you think there is a need, I can look into this. If not, I'll happily live with these lines commented out.
01micko wrote: Users may find that installing acpid from Slackware repo may fix suspend issues as they have many rules for many different machines. YMMV though as I have no need to test. Reports needed. Use at own risk.
I tried the acpid from Slackware repo, but that didn't work any better.

One more note: I installed Skype SFS from the quick SFS installer and all worked out of the box, including webcam and sound. Bravo, definitely the best puppy I used so far.

Posted: Wed 16 Jan 2013, 11:21
by 01micko
Thunor has released gtkdialog-0.8.3 recently. I have packaged it for slacko BUT it's versioned as gtkdialog-0.8.4. Don't worry! It's the same as the stable version 0.8.3, just built from SVN. It is needed for the latest Pburn and upcoming Pmusic-3.1 by zigbert.

See Main Post!

Cheers!

Posted: Wed 16 Jan 2013, 20:48
by bark_bark_bark
the new gtkdialog works just fine. I tested it with the new pBurn and pMusic.

Posted: Sat 19 Jan 2013, 18:15
by zigbert
I am trying to extend my Slacko 5.4, and met the following issue:

- installing inkscape from ppm and all the detected dependencies.
- since gtkmm is already installed statically ppm does not detect this missing dep until the last verifying - Puppy has the .a and .la libs, but misses the .so
- I just search up gtkmm in ppm and tries to install the dynamic version of gtkmm, but this fails. I guess it is because it is marked (ALREADY INSTALLED).

How can I bypass this?


Sigmund

Posted: Sat 19 Jan 2013, 20:14
by 01micko
zigbert wrote:- installing inkscape from ppm and all the detected dependencies.
Inkscape was broken until 2 days ago. It should automatically now get it's dependencies gsl and aspell however you should update PPM and when doing so reject all repos that are puppy except slacko14. Also, PPM has a bug where it repeats dependencies from ALL available repos! Including wary. Please uncheck all duplicate deps from other repos and ensure that just one dep is marked. Yeah, not very user friendly :(
zigbert wrote:- since gtkmm is already installed statically ppm does not detect this missing dep until the last verifying - Puppy has the .a and .la libs, but misses the .so
- I just search up gtkmm in ppm and tries to install the dynamic version of gtkmm, but this fails. I guess it is because it is marked (ALREADY INSTALLED)
I just tried to download gtkmm and succeeded (it comes in from Salix repo). After it is installed it says cairomm, atkmm (and others) are missing, I guess you get them manually. The reason they are not detected as deps is the same as the reason you found ..(ALREADY INSTALLED). Yeah, not very user friendly :( . (I should perhaps rename the static versions to gtkmm_S or something).

Posted: Sat 19 Jan 2013, 21:18
by zigbert
Thank you Mick for the explanation.
I have to agree with you that ppm is not the most logical interface I have seen.


Sigmund

Posted: Sat 19 Jan 2013, 22:12
by 8-bit
I have a Compaq Presario with an AMD Athlon X2 4800 Dual-Core processor, 3 gigs of ram, and Nvidia GeForce 6150 SE Graphics.
I am running Slacko 5.4 final with comp-fusion.
I am getting a random suspend of both input and output both locally and on the net.
The suspend of IO is at random times and of random length.
When left alone, it eventually recovers.
Has this been looked into?
What, if anything can I do to remedy this?

Also, it seems to be restricted to that frugal install of Slacko 5.4 and does not show up in Lupu 520, Puppy 431, and Fatdog 6.11.

When I am typing a reply to a post on the forum, I get instances of the display of input text not keeping up with the keyboard, the mouse not selecting a clicked on item, and videos stopping or skipping.

If when typing, I continue typing, the display eventually catches up with the text the keyboard buffer contains.

So where do I go from here?

Posted: Sun 20 Jan 2013, 01:39
by 01micko
Hello 8-bit
I am running Slacko 5.4 final with comp-fusion.
Upgrade your video driver.

It's hardly fair to compare to other puppies which wont be running compiz!!! Compiz comes with appropriate warnings... ymmv.

Also, compiz often requires tweaking to xorg.conf and other settings. You'll have to search yourself. Compiz is provided as is.

Posted: Sun 20 Jan 2013, 01:46
by 8-bit
I am already running an upgraded video driver for my Nvidia card that works fine with all the other Puppy OSes I have installed.
But as you say, comp-fusion has it's own set of problems.
So for bling, comp-fusion can be used.
But for a everyday reliable OS, one would stay away from the bling.

But I am still curious as to if others have suspend problems when running comp-fusion with Slacko!

Posted: Sun 20 Jan 2013, 20:01
by SFR
8-bit wrote:But I am still curious as to if others have suspend problems when running comp-fusion with Slacko!
8-bit, I'm encountering similar symptoms, but not related to Compiz (you can read about it in my earlier posts here).
Do those suspends occur under JWM or any other WM too?

_____________

@ Mick:
Two tiny problems with features that probably aren't used by most of users, but they're still bugs...

1.
Does anyone have noticed that 'right-click Trash icon > View Summary' displays empty tree, despite the fact that there are files inside?
/usr/local/apps/Trash/AppRun, lines 189-:

Code: Select all

du -sb /$HOME/.Trash/* |sort -rn | sed -e 's/[^ ]*\t//'| while read line
 do cat "$line/Info" | tr '\n' '|' | sed 's/Item = /\n/'; done \
| sed 's/Name = //' | sed 's/Path = //' | sed 's/Size = //' | sed 's/Deletion Date = //' \
| sed 's/Type = //' | sed 's/Extension = //' | sed 's/Had desktop shortcut? = //' \
| sed 1d > /tmp/trashitems
That's still ok, but the next line:

Code: Select all

cat /tmp/trashitems | sed -e 's/</?/g' | sed -e 's/>/?/g' > /tmp/trashitems
doesn't work as it supposed to, I guess.

Proposed fixes:
- To append those two sed commands to the above "long sed parade"
- Actually Gtkdialog won't mind if there will be < > chars in cat'ted string between <input>cat /tmp/trashitems</input> tags, so maybe delete the line completely..? (works for me)

2.
ROX-Filer Panels: right-click a panel -> Panel Options won't bring dialog window.
Cause: lack of /usr/local/apps/ROX-Filer/Templates.ui file
I took one from some arch package (can't recall from where now) and it's fine now.

Greetings!

Posted: Sun 20 Jan 2013, 21:11
by bark_bark_bark
On any WM I experience Suspend troubles. I have to shut the whole thing down (forcefully) in order to be able to use it again.