inconsistent SFS creation results

Issues and / or general discussion relating to Puppy

Moderator: Forum moderators

Post Reply
User avatar
JASpup
Posts: 1653
Joined: Sun Oct 04, 2020 10:52 am
Location: U.S.A.
Has thanked: 70 times
Been thanked: 89 times

inconsistent SFS creation results

Post by JASpup »

Early attempts:

1. createsfs turned the default Tahr32 Thunderbird 31 PET into a SFS that loads.

2. The same operation on Opera 43 browser created a 4kb SFS with mystic process messages.

3. pets2sfsgui created a SFS for the Opera, but the loaded SFS does not run. The PET it was made from installs/runs.

On the Whiz-Neophyte Bridge
Linux Über Alles
Disclaimer: You may not be reading my words as posted.

s243a
Posts: 501
Joined: Mon Dec 09, 2019 7:29 pm
Has thanked: 90 times
Been thanked: 37 times

Re: inconsistent SFS creation results

Post by s243a »

JASpup wrote: Fri Dec 11, 2020 1:30 pm

Early attempts:

1. createsfs turned the default Tahr32 Thunderbird 31 PET into a SFS that loads.

2. The same operation on Opera 43 browser created a 4kb SFS with mystic process messages.

3. pets2sfsgui created a SFS for the Opera, but the loaded SFS does not run. The PET it was made from installs/runs.

The most likely reason, is that the sfs didn't include all the dependencies. I think that Scottman is working on a solution for this:

in fact, that is what I plan to do with Pkg... Make it build an SFS that contains all the *.files of the packages that went into it, instead of one big, combined one - so the package and all its deps are listed as installed, once the SFS is mounted.

https://github.com/puppylinux-woof-CE/w ... -729883273

There could be potentially other issues though. SFS's are loaded in the bottom layer. Puppy could have more control over what layer sfs files are loaded in. The easypup/T.O.P.P.L.E.S.S/N.E.M.M.E.S.I.S. searies of puppies let's you load them in higher layer. Mty psandbox script also lets you do this. Another possible issue is that lower layers could be blocked by whiteout files. Puppy has a feature to prevent this, where whiteout files are cleaned but this may not always be desirable. My understanding is that this cleaning procedure is done when you upgrade puppy so that upgraded changes aren't blocked. It is safer in my opinion to start a new savefile when upgrading rather than trying to reuse an existing savefile. Perhaps the existing savefile could be used in a sandbox either with the old or new basesfs.

User avatar
JASpup
Posts: 1653
Joined: Sun Oct 04, 2020 10:52 am
Location: U.S.A.
Has thanked: 70 times
Been thanked: 89 times

Re: inconsistent SFS creation results

Post by JASpup »

s243a wrote: Fri Dec 11, 2020 2:03 pm

The most likely reason, is that the sfs didn't include all the dependencies. I think that Scottman is working on a solution for this:

You're guessing the Opera PET grabs dependencies the SFS requires us to manually include? That could be it. Usually when I install a browser it either doesn't require new dependency installs or they're tricky to include and it just won't run.

What explains why createsfs isn't making the Opera SFS but pets2sfsgui is?

I'm sort of looking for the path of least resistance, so if a tool works with a little haggling, I'll use it, but if the issue is layers and hacks, I'm probably ahead of myself.

On the Whiz-Neophyte Bridge
Linux Über Alles
Disclaimer: You may not be reading my words as posted.

Post Reply

Return to “Users”