Fatdog64-710 Alpha/2 [29 April 2016] [CLOSED]

[Doglover]

Fatdog System Updater now works ty to SFR. And now the drive icons do not magically disappear. Yay!

Configuring the sound card is a snap.

specs....

Hp Pavilion circa 2014 laptop

Processor 4x AMD A6-5200 APU with Radeon(TM) HD Graphics
Memory 5570MB (555MB used)
Machine Type Physical machine
Operating System Fatdog64 [710]

Resolution 1366x768 pixels
OpenGL Renderer Gallium 0.4 on AMD KABINI (DRM 2.43.0, LLVM 3.7.0)

Audio Adapter HDA-Intel - HDA ATI HDMI
Audio Adapter HDA-Intel - HD-Audio Generic

Total Memory 5570584 kB
Free Memory 4444272 kB
MemAvailable 4851292 kB
Buffers 87052 kB
Cached 571592 kB
Cached Swap 0 kB

BIOS
Date 11/20/2013
Vendor Insyde
Version F.13
Board
Name 216F
Vendor Hewlett-Packard (www.hp.com)

2000-20ff PCI Realtek Semiconductor Co., Ltd. RTL8101/2/6E PCI Express Fast/Gigabit Ethernet controller (rev 07)
2000-20ff Module RealTek RTL-8169 Gigabit Ethernet driver
3000-3fff PCI Bus 0000:01
3000-30ff PCI Realtek Semiconductor Co., Ltd. RTL8188EE Wireless Network Adapter (rev 01)
3000-30ff Module PCI basic driver for rtlwifi
4000-40ff PCI Advanced Micro Devices, Inc. [AMD/ATI] Kabini [Radeon HD 8400 / R3 Series] (prog-if 00 [VGA controller])

This box loves Fatdog.

[SFR]

Also, these don't work in titlebar, but I just checked in Lupu-528 and none of them worked, too, so I suppose that was intentional:

• gtk-discard (?)
• gtk-goto-first
• gtk-goto-last
• gtk-go-back
• gtk-go-forward
• gtk-indent
• gtk-jump-to
• gtk-media-forward
• gtk-media-next
• gtk-media-play
• gtk-media-previous
• gtk-media-rewind
• gtk-redo
• gtk-revert-to-saved
• gtk-undelete
• gtk-undo
• gtk-unindent

Well, turns out that these icons are not being displayed in YAD (see here and my reply a few posts later).
So, I modified the patch and added them, too (except gtk-discard).
New patch attached, works fine with the stock recipe.

Greetings!

[rcrsn51]

How to Install your Printer/Scanner in Fatdog has been updated here.

This is a first draft - it needs testing from Fatdog users who own the applicable printers.

[Doglover]

Trying to get wine working but fails due to

linux32: /opt/wine-1.9.7/bin/wine: No such file or directory

Is Fatdog 32 bit compatible?

Seems to install ok, checking /opt shows the wine directory.

screenshot

[SFR]

Did you also load 32bit-fd64.sfs?
Without it you can still use it, but only in 64bit mode (via wine64).

Greetings!

[Doglover]

no (:-( oops

[Doglover]

Alpha2 now boots in under 20 seconds. Very nice and solid.

[don570]

Fatdog uses an old and buggy version of leafpad
Here is is the improved version

http://www.filewatcher.com/d/FreeBSD/di ... 89079.html

_____________________________________

[step]

@jamesbond gslapt issue

start gslap and update, look for 'poppler', you should find two versions, 0.41.0 and 0.42.0 (installed by the devx sfs). Select the 'Common' tab. Click version 0.42.0 to read the details. Now click 0.41.0 - the details don't change, gslapt keeps showing details for 0.42.0 even though 0.41.0 is selected.

[spandey]

In Fatdog does the savefile/Diectory loaded to Ram during Boot?

[mikeslr]

Hi spandey,

With a couple of exceptions, all Puppies work in the same way as bigpup suggested in response to your question on a different thread. AFAIK, the exceptions are the Quirkys Barry K's developed in the last 2 or so years and a couple of Puppies ETP developed based on those Quirkys. Those Puppies can (I think) only be installed to a USB-Stick initially, at least, using an entire partition. They install as Full Installs (not Frugal). [Perhaps, later, it may be possible to add Frugal Puppies and create a boot menu -- haven't tried]. The first post of such "divergent's" thread will usually tell you how to install, or point to directions.

The post bigpup pointed you to explained how Puppies worked shortly after Barry K developed the technique. At that time, Puppies BASE files were small, and SaveFiles (no SaveFolders, yet) were also small -- as were USB-Sticks.

Since then, with the cost of RAM and "Storage Media" getting cheaper, applications got larger and multiplied in number. Techniques were developed to take advantage of there being more RAM and Storage and enable more and larger applications to be included in the BASE files and later installed into the SaveFile/Folders. But Puppies still had to be able to work in less advantageous environments.

The technique developed enable a Puppy to NOT copy everything in either the BASE file or the SaveFile/Folder into RAM. rufwoof explains it here: http://murga-linux.com/puppy/viewtopic. ... 458#827458. This is my understanding of what he wrote: http://murga-linux.com/puppy/viewtopic. ... 192#862192. Suggest, for context, you read the entire short thread on which that post exists.

The bottom line is that a Pup's Developer provides instructions in his/her Pup's "boot-up" code as to how that Puppy will function. Those instructions are essentially the same as Barry K first envisioned modified only as rufwoof explained. You, as user --taking into consideration the amount of RAM, the power of your CPU and the storage media from which a Puppy will boot-- can over-ride those instructions by using "boot codes".

If you used Grub4Dos as bootloader, it will have written menu.lst to the boot partition/drive. One of the lines in menu.lst, as originally written, will be something like (if a frugal install to a folder named "tahr64" on a Hard-drive)

kernel /tahr64/vmlinuz psubdir=tahr64 pmedia=atahd pfix=fsck

As written, Puppy will periodically write what is currently in RAM to your SaveFile/Folder.

If you change pmedia to read pmedia=ataflash, Puppy will not write to the SaveFile/Folder but will offer you that choice of doing so at shutdown/reboot. [There will also be a way to "manually" execute a Save whenever you want, but the mechanism will depend on which Window manager you are using].

You can also change the arguments following pfix:
pfix=ram tells Puppy never to copy the SaveFile/Folder into RAM.
pfix=copy tells Puppy to copy the entire SaveFile/Folder into RAM (if possible)
pfix=nocopy tells Puppy to create links to the SaveFile/Folder, but not copy any files unnecessary to display the existence of applications linked to.

fsck, by the way, is an instruction to check integrity. So, for example, if you wanted your Frugal install on a hard-drive to act as if it were on a USB-key, check for integrity and copy the bare minimum files into RAM at bootup, the above line would be edited to read:

kernel /tahr64/vmlinuz psubdir=tahr64 pmedia=ataflash pfix=fsck,nocopy

That's a comma (,) between fsck and nocopy.

The above are not the only boot codes available.

If you used Grub or Grub2 as bootloader, boot codes would be similar. The line would begin with linux rather than kernel.

mikesLr

[gcmartin]

Hi @Spandey

Firstly, what @MikeSLR shares (and BigPUP) is accurate and offers the explanations of others to enhance the understanding of Puppy boots.

Secondly,

In Fatdog does the savefile/Diectory loaded to Ram during Boot?

Yes. Consistent with what the above has shared. This is done via the boot manager as well which, at boot time's Menu, you can modify any one of the lines, on-screen, to match a specific need you might want to use/test. For example, you "might" have a savefile/Directory for one session's services, while also having another savefile/Directory for some other specific session's services in a different location.

More on FATDOG's savefile/Directory use can be found by searching for "Advanced Parameters" on this FATDOG webpage where savefile examples exist.

Hope this helps

[spandey]

Thanks a lot to all of you. I am trying FATDOG now. Will comeback if there are any issues?

[Belham]

Hi all,

Have a question: how do I get Seamonkey 2.40 into Fatdog64-710 (Revision 2) over top of the existing 2.39? My normal mode (for any puppy project i currently use) is of updating/replacing Firefox and Chrome the easy way. For example, in my Fatdog (have both of those browsers installed along Seamonkey), I am just downloading the latest versions of the browsers straight off the web, unarchiving those, then replacing the unarchived Firefox or Chrome folder contents into the /usr/lib64/firefox folder for Firefox or into the /opt/Google folder for Chrome (of course, don't forget to add "--user-data-dir" to the last line in the google-chrome shellscript), making sure to touch nothing else, and voila', I've the latest browser version of these two running. The best is that they work with all the menu and traybar/taskbar shortcuts I've currently setup It's like 2-3 mins tops doing this.

But when I do this for Seamonkey 2.39 and bring in Seamonkey 2.40 into its /usr/lib64/, Seamonkey 2.40 won't even start, even if I go straight to its bin.file (and even reset the existing symbolic link to it).

What am I doing wrong? Do I have to compile Seamonkey 2.40 to get it into Fatdog versus what I do for Firefox and Chrome? Or does it have something to do, with say, that Seamonkey overall is so integrated (Seamonkey mail and Address Book) into Fatdog that I am overlooking doing something for updating Seamonkey?


Thanks.



EDIT: FRI, 20 MAY, 2016.....Just wanted to give an update on this. I have tried everything that i can think of to get Seamonkey 2.39 in Fatdog64 (all versions currently) updated to 2.40. I've downloaded regular Seamonkey 2.40 installs, copied directories (after changing user/group permissions), I downloaded the unofficial 64-bit builds from Seamonkey too, it also would not work when installed. I even went so far to load up Barry's Qyirky-Xerus64, copy the Seamonkey 2.40 folder directly from its /usr/lib64/ location, make sure all symlinks were reset, copy it all over to Fatdog64's /usr/lib64 directory, and 2.40 still would not start. My last gasp was giving compiling a whir after downloading Seamonkey-2.4.0 source, and even though it too seemed to install cleanly (after compiling, which gave no errors), do you think Seamonkey 2.40 would start up in Fatdog64-710a2?? Haha, heck no! I am now officially off the ends of my knowledge in getting Seamonkey 2.39 updated in any Fatdog64. Weirdly, I was able, with no problems, to do this updating of Seamonkey from 2.39 to 2.40 in all other 64-bit puppies I currently use, but ole' Fatdog has me stoooooooooped again. Long live Commander Kirk & Mr. Bond in keeping us ensigns drunk-guessing like a wallaby hitting an Alice Springs saloon! (j/k ya, Kirk & James)

[TGS]

Really liking this release so far. I've been a fatdog user for the past year or two now and really loving it. It's by far the best 'puppy' variant I've seen. Sadly in terms of my day to day usage I've stuck with the 701 release as it seems to be the most stable and least issue prone.

Unfortunately I often use VMware Workstation to test out and even use VMs as workstations, in 701 and 702 this version gives a massive range of resolutions, but in 710a for some reason the resolutions are very limited. Which ironically is a common issue amongst Linux Distros in general with VMware workstation. The resolution I use as standard 1920x1080 is not on the list at all. Normally one would just say install the VMware tools (which provides that functionality in other distros) but I'm not really sure why 701 and 702 both have the widest range of resolutions including 1920x1080 but 710a doesn't?

That is actually one of the reasons I love Fatdog so much. Many other distros have issues with resolutions and VMware, or just issues in general. Fatdog up to this point (of 710a) has not had that issue.

[WillM]

Hi Belham,
The official Seamonkey is 32 bit. I don't know if you tried the unofficial 64 bit or not so here is a link to the page for the download, it is at the bottom of the page under Contributed builds.
http://www.seamonkey-project.org/releases/#2.40

I put that in /usr/lib64, renamed the extracted tarball to seamonkey-2.40 and linked that directory to seamonkey. It worked fine.

[Rodney Byne]

Hi Belham,

After reading your annoyance earlier with the red update bar,
I managed to painlessly get rid of mine ok,
although not being a coder or compiler I can't offer
any suggestions to update from v2.39 to v2.40.
I think we are all stuck with it.

With apologies if you already know this stuff
but to get shot of the red bar:

Enter about:config into the url address bar & enter.
Press the blue "I promise etc" button.

Scroll down the displayed list to edit this first line:
app.update.auto then right click to toggle true to false.

Scroll a little further down to edit this second line:
app.update.enabled then right click to toggle false to true.

Re-launch Seamonkey to check that the red bar has gone.

That's it - job done.

Best regards.

[SFR]

x11vnc-helper

When devx.sfs is loaded, it executes itself in an infinite loop:

Code: Select all

if type wish > /dev/null; then
	exec $(sed '/Exec=/!d; s/Exec=//' /usr/share/applications/x11vnc.desktop)

Code: Select all

# sed '/Exec=/!d; s/Exec=//' /usr/share/applications/x11vnc.desktop 
x11vnc-helper
# 

Also, in 32bit-compat.sfs /usr/bin/wish-32 is a broken symlink.
The binary it should point to has an extra 'h':

Code: Select all

# ls -l /usr/bin/wish*32
lrwxrwxrwx 1 root root   10 Jan  9 10:05 /usr/bin/wish-32 -> wish8.6-32
-rwxr-xr-x 1 root root 4568 Jan  9 10:05 /usr/bin/wishh8.6-32
# 

Greetings!

[step]

Findnrun is included in Fatdog64-710, press Alt+F3 to start it. It has its own entry in the Filesystem system menu. That's version 2.1.1. I have uploaded release candidate 1 of the upcoming version 2.2.0 for testing. I would appreciate Fatdog64 users' help in testing this RC version since the final version will go into the final Fatdog64 ISO.

Please report bugs in the Findnrun thread not here. Download links in the linked thread.

Thank you in advance.

[step]

The FreeOffice suite is a sure-fire lxqt-panel killer. After installing the free suite and running textmaker16free lxqt-panel crashes. So long as textmaker is running the panel doesn't come back in spite of the daemon restarting it. dmesg log ad libitum:

Code: Select all

[23138.646957] lxqt-panel[1665]: segfault at 7fc238022000 ip 00007fc247c0f2c4 sp 00007ffec7168978 error 4 in libc-2.19.so[7fc247b91000+16c000]
[23142.411714] lxqt-panel[32755]: segfault at 7f38f0021000 ip 00007f38fcbe32c4 sp 00007ffcf88b5678 error 4 in libc-2.19.so[7f38fcb65000+16c000]
08

I installed version 756 (the shell installer) >>register first then download 756 tgz. Registration is quick (no passwords) - enter your email address and they'll send you the activation code. Install - you can choose /usr or another path (I chose /usr) and run textmaker. After the activation dialog lxqt-panel will crash. I think this suite is 32-bit. They do have a 64-bit .rpm package. I plan to test that too just to see if it triggers the panel crash.

edit: no change, the supposedly 64 bit rpm is indeed 32 bit and loads the same libs.