antiX live usb issue

[fehlix]

.. i wonder if something else is going to break?

It will be fine to run without fuseiso. Don't think it will break anythink if you dont have fuseiso, unless you try to run live-usb-maker on a gvfs-fuse mounted network share, which might break loop-mounts...

[rs55]

Thanks - thats great .

[dolphin_oracle]

You nailed it!
Running the cli version with --force=nofuse from antiX did the trick.

So- now I have a question:
If the only reason the live-usb was working in MX was because of a "bug" where it was asking for fuseio9660 rather than the correct name - just fuseio, is it likely this will get "fixed" in an mx update - which will cause that to also not work !!

I would like to bake this "nofuse" into the gui version - because I am now worried that using a live usb with persistence turned on and doing occasional remasters is going to fail in mx ??

I'm seeing the same only with the gui live-usb-maker-gui-antix on antiX. The cli app (live-usb-maker) works ok without --force=nofuse

We use the "gui" command when calling lum from the gui. Maybe the defaults under that switch are different from those at the standard command line.

[anticapitalista]

Other than live-usb-maker , what other apps are using fuseiso?
I ask, because , if I uninstall it ( which makes live-usb work for me), i wonder if something else is going to break?

Remove it, nothing will break (as far as we know)
Next bugfix upgrade of live-usb-maker-gui-antix will force an install of fuseiso9660 and remove fuseiso.

Thanks for your report - and fix.

[BitJam]

In my antix, I uninstalled fuseiso and installed fuseiso9660. And now the gui works as well !!

So - now - I understand it is only working because the live-usb-maker is looking to run fuseio and failing to find it !

I dont understand all the details behind this fuseiso business, but live-iso is a critical part of mx and antix, so please dont run fuseiso just to experiment . At least not in the published versions of mx and antix.

@rs55 We have had many many more failures reported due to not using fuseiso. So far we have only your failure using it.

If you don't mind, please run a test for us to see if fuseiso9660 works. Please make the following change to /usr/local/lib/cli-shell-utils/cli-shell-utils.bash. Change line 2493 to:

Code: Select all

    local prog=fuseiso9660

IOW change "fuseiso" to "fuseiso9660". Make another live-usb without using "---force=nofuse" and see if it works. Also, please doublecheck in the /var/log/live-usb-maker.log file to make sure you session (near the end of the file) has a line that says something like:

Code: Select all

Mount some-iso-file with isofuse9660

I really want to know if this fixes the problem or not. I would also like to see the output of:

Code: Select all

sudo lspci -nvv

on the problem system. Perhaps attach it as a file or email it to me at <my-user-name>@gmail.com.

If using fuseiso9660 does not fix the problem then I have one other thing I would like to try.

We were VERY sure we had fixed the fragmentation problem. We had incorporated 3 different fixes even without the --force flag. But I added --force=nofuse just in case we were wrong. It is true that I also experimented by making using fuseiso the default. The reason for this was so we could catch rare problems like yours sooner.

If we had only used fuseiso when it was needed then:

1) It probably would have been several years before we found a problem, all the while assuming fuseiso was solid.

2) It would have been inconvenient for the user who had the problem to test solutions because they would not be able to read the iso file without fuseiso!

I am delighted you like our live system! I've spent the last 8 or 9 years developing it. We are really trying to do the greatest good for the greatest number of people. Also, as soon as you told us the actual error message we were quickly able to zero in on the problem and the solution. I'm sorry you had to go through this but your efforts will help many other people in the future. Thank you for you help! We depend upon users like you to help us improve our software. I personally have four different test machines and all told our development team probably tests on 50 to 100 machines. But we can't test on every possible system so we rely on help from the users. So thank you again.

[rs55]

BitJam:
Thanks. OK I did everything you suggested. I changed the /usr/local/lib/cli-shell-utils/cli-shell-utils.bash line to.fusiso9660. Ran lum (without the --force flag) -and the usb failed with the same error.
I did check the log file and indeed it had that mount line for fuseiso9660.
I have attached the output of lscpi -nvv to this response.

You have created a fabulous operating system- and I have been involved with computers since 1972 !! ( have'nt done any engineering work in three decades though ).

Let me know if you want me to try anything else. best regards.

[anticapitalista]

Seems that the fix is to use fuseiso9660 and not fuseiso. (make sure fuseiso is removed)

I did this and both live-sub-maker (gui and cli) both work (on antiX)

[rs55]

fuseiso9660 failed in my machine - see post above.

[BitJam]

Thanks rs55!

Can you tell me the output when you run:

Code: Select all

sudo e4defrag -v vmlinuz

on the live-usb?

I started coding professionally in the early 70s. This is one of the top 10 strangest problems I've seen. Some subtle change in timing is causing fragmentation when we copy from an iso that was mounted with fuseiso.

I'm working on another fix I would like you to test.

[rs55]

I had previously uninstalled fuseiso from the system. Now I also unistalled fuseiso9660. And now the lum works great again.