-
-
Notifications
You must be signed in to change notification settings - Fork 501
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
General | Drop Jessie support #2332
Comments
They should be kept for some time. |
@Atavic Do you have a reason for that? Is there anything that works better on a certain Jessie setup, or works on Jessie only? Of course DietPi code support for Jessie should stay longer, but the question is, if there is any reason to still ship Jessie on fresh images, which would further delay the possibility to move on with the code as well. |
No reasons except the doubt that you clarified:
Debian has this page: https://www.debian.org/releases/jessie/debian-installer/ So it could be fine to have it still available for a while, even if superseded by following versions. |
@Atavic For now it's just about preventing new users from installing a fresh Jessie DietPi image. It is not only our code support, but also that support by other 3rd party software will be dropped by times. So there are other disadvantages rising by times, outside of our control, if one chooses to install Jessie now. Removing Jessie images and install option from DietPi-PREP also is a clear message that one should not use it anymore. |
Yep agree, no benefit to Jessie over Stretch. And it increases our level of support effort which should be spent elsewhere (eg: buster). My preference:
|
Isn't the prompt warning enough? |
@Atavic We did something similar when dropping wheezy support: https://github.com/Fourdee/DietPi/tree/wheezy-final-v120
A separate branch has the advantage that we don't need to take care this anymore on future updates, when doing changes on update system or similar that somehow accidentally re-enable Jessie updates. |
One issue:
|
@Fourdee Related new issue: https://github.com/Fourdee/DietPi/issues/2561 |
amd64 packages from the Debian main repo are not available anymore, similar to what happened to the security repo some time ago: #2665 |
Issues with Jessie are growing, leading to testing and workaround effort that we can not handle much longer, e.g.: #2665 We will support Jessie for at least one version longer (until v6.24, at least) but move systems to the dedicated Jessie systems are less than 0.5%, constantly decreasing: https://dietpi.com/survey/#distro I am most sorry for Odroid C1 users, where we cannot ship a Stretch image due to major limitations kernel and firmware-wise: #2561 Related PR: #2742 |
C1:
I'll update the image as a final image with new jessie branch. |
@Fourdee |
Okay, v6.25 is the last version with Jessie support. v6.26 will have all Jessie related code removed completely. The Jessie-support branch will stay, most likely until Stretch support is about to be dropped (~2 years from now) and may receive very important bug fixes, when reported or some slight Jessie tailor when someone is in mood to do so. However again I recommend to switch to Stretch, where possible, also for security reasons. |
PR up for dropping Jessie support and related code cleanup: #2963 |
Since all remaining Stretch related bugs are solved (mainly Odroid C2 Kodi HW acceleration + desktop Xorg start), we should take out all Jessie images from official download source. As well we should disable support in DietPi-PREP, only allowing to upgrate Jessie images to Stretch.
However they could be kept in e.g. legacy sub dir, but then with a README.txt that they are not recommended and support might be dropped.
We should then keep an eye on existing Jessie install count (survey) and encourage users to switch to a Stretch image, e.g. via notification in banner and/or on next update.
I suggest we either at timestamps to survey report or simply remove older uploads to have more current numbers.
Currently there are
1.2%(EDIT: <0.5%) Jessie installs.Waiting for e.g. 3 months, checking Stretch vs Jessie counts on meanwhile uploads, then create final Jessie update and drop support from code, if numbers fall under certain percentage.
Feedback and suggestions on this are highly appreciated. We will definitely drop Jessie support by times, to free code from special handling and reduce testing effort, especially when it comes to official Buster release, expected in summer 2019.
But we want to make this change as smooth as possible for Jessie users.
The text was updated successfully, but these errors were encountered: