Theoretical "Flash Failed

balenaEtcher
1

Just a note, maybe for future consideration in Etcher itself.

Am running Debian Linux, Testing branch. Additional, under Wayland.

Every time I flash an iso, it fails - something to do with “checksum failed for range” and some number which mean absolutely nothing to me.

In the subject I said “Theoretical” - on booting the usb stick, it appears to be perfectly fine.

Never had this running with X11, so am led to conclude is to do with Wayland.

Yes, Wayland still has a few little paper-cuts still, but nothing (for me, anyway) of any importance - little irritations are expected and quite easily lived with.

However, every day it is becoming clearer that X11 is going bye-bye, to be replaced by Wayland. Good Thing, too.

So, if Wayland is to become the “new” display protocol, Balena Etcher will need to adapt accordingly.

Anyhow, just download the latest version on Balena.io, not yet set up, will see if non-issue continues to exist.

But at least, if any Linux user complains of failed flashes, in the meantime, one can simply ask: “did you try booting the usb stick?”

Romane

2

Hi Romane - thanks so much for your detailed feedback, it’s really useful and I’ve logged it into our system so the Etcher team are aware of it. Did downloading the latest release have any impact on the issue at all?

4

Good morning Jon

Yes, for myself did call it a non-issue (grinning)

No, the new version behaved the same. Falsely advised that the flash had failed, but all of the flashed USB sticks booted and ran fine.

I still have access to X11, so tomorrow (after shopping) will do a test, running the X11 system. Am expecting it to go through without error, but only by doing the test can it be claimed with any certainty that it may be because of Wayland (key word ‘may’). Will report the result -it may alter my feedback in some way. (but I wonder why a display server would cause a failure of verification? Am not a programmer or with skills in that area, as such, but am unable to see why just using logic and reasoning).

Romane

5

Tested with different USB drive. Same type as the other one. Bought at the same time as the other, part of a “bulk” purchase.

And it worked as it should, as well as with a different style of stick - no “failure” report from the verification result.

So, an even (read much) more likely reason. The flash drive is sufficiently faulty that it does result in an checksum error; only takes one bit to make a difference. so, update to my original post, not caused by the display server but by the flash drive (now residing in the appropriate bin). Now, that makes more sense and is a more reasonable cause for the “issue” than looking at the display server.

Though what I said about Wayland replacing X11 still holds true. Those applications which natively run under Wayland perform much better than those which run through xWayland (provides compatibility for applications not yet ported from X to Wayland), though Wayland is not yet anywhere near “ubiquitous” in the various Linux distributions

Romane