Extend troubleshooting guide according to findings during matrix support chat #283
Conversation
…to findings during matrix support chat
There was a problem hiding this comment.
I've been nitpicky about some small grammar things, but also about some larger presentation things, I hope this is appreciated. There are a few more grammar bugs I've noticed, but I'll pick these out once a few more people give their opinions on the overall text.
|
Thank you very much for the review @dodoradio |
eLtMosen
force-pushed
the
refine-install-troubleshoot
branch
from
e07f267
to
a132b69
Compare
| <li><h4>Verify that you used the correct image files</h4>Some devices like the Asus Zenwatch 2, the Fossil Gen4 models and the MTK6580 watches have more than one system image supplied, for different variants with slightly different hardware. Most watches have a graphical fastboot menu that shows the correct codename for the watch. Others need to be identified using the hardware differences noted above in the install section.</li> | ||
| <li><h4>Clean your watch and charger contacts</h4>Even in case the fastboot flash process succeeds with no errors, it can still silently fail due to connection problems caused by dirty contacts. Clean the pins and pads with isopropyl alcohol. Ideally using that old toothbrush you keep around for electronics cleaning or a microfiber cloth. Carefully sanding the contacts (e.g. with a nailfile) can help in case of obvious corrosion.</li> | ||
| <li><h4>Ensure a stable USB connection</h4>Put the watch and your computer on the same flat surface. Try to not move the cable connecting your computer and the watch during the whole flashing process. In case of worn or corroded contacts and pogo-pins that lost their springiness, it might be useful to attach the watch to its cradle or charging cable using duct tape or rubber bands.</li> | ||
| <li><h4>Repeat the flash process.</h4>Some users have reported needing up to five flashing attempts for a successful result.<br>There is a small chance that the image files were corrupted during download. We do not currently provide checksums for the files. However, if you download the files again under a different filename and then compare them with <code>sha256sum filename</code>, the checksums must match.</li> |
There was a problem hiding this comment.
Great idea! This makes me wonder if we should provide the sha256sum information on the install pages themselves too since we already know the checksums in advance: https://release.asteroidos.org/nightlies/beluga/MD5SUMS
(This also means that the We do not currently provide checksums for the files. claim, is incorrect 😉 )
There was a problem hiding this comment.
That is a great surprise! Coool, i thought it would be complicated to add checksums so did not ask but came up with the re-download work around.
Let's keep the We do not currently... sentence until the checksums are actually shown on the website?
Making a new issue for that:
#284
There was a problem hiding this comment.
This is so cool ! Thank you so much @eLtMosen and everybody who contributed to the reviews :D Our users will be very thankful
|
Oops, I should have squashed that git log but was too excited to merge it... Weellll.... now it's there uh |
There have been multiple instances now where users came into chat and asked for help with a bootloop. Many of them under impression they might have bricked their devices.
I take this as a sign that our current install troubleshoot might profit from some refinement. Especially since the keyword "Bootloop" is not even used as a topic heading currently.
This PR tries to better reassure users and guide them to the appropriate solution steps much more detailed than before.