Skip to content
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

Upgrade to 1.0.0 (Mailu 2.0.0) #218

Open
fastlorenzo opened this issue Apr 12, 2023 · 12 comments
Open

Upgrade to 1.0.0 (Mailu 2.0.0) #218

fastlorenzo opened this issue Apr 12, 2023 · 12 comments
Labels
documentation Improvements or additions to documentation

Comments

@fastlorenzo
Copy link
Collaborator

Please read the migration guide if upgrading from a previous version of the Helm Chart as it has a few breaking changes that requires manual actions.

Starting from version 1.0.0 of the Helm Chart, only Mailu 2.0.0 is supported.
If you want to continue using Mailu <2.0.0 (not recommended), please use version 0.3.3 of the Helm Chart.

@fastlorenzo fastlorenzo added the documentation Improvements or additions to documentation label Apr 12, 2023
@fastlorenzo fastlorenzo pinned this issue Apr 12, 2023
@kooskaspers
Copy link

kooskaspers commented Apr 13, 2023

@fastlorenzo Could you elaborate a bit more on the data-migration part? Like how to migrate the data from the embedded mysql database to the mariadb/postress instance? Especially how to create a SQL dump file for recreating the database.

@fastlorenzo
Copy link
Collaborator Author

@kooskaspers if using MySQL, you can do a mysqldump on the old instance, then deploy the new version of the chart, and restore the dump.

@dominch
Copy link

dominch commented Apr 13, 2023

Just yesterday I decided to change one small thing on my release and that turns out as complete disaster :(
As You probably know all docker.io images were deleted, so redeploying now cause strange and not intuitive message:

Error response from daemon: pull access denied for mailu, 
repository does not exist or may require 'docker login': denied: requested access to the resource is denied

This does not sound ok ;)
Only one pod was ready and that was redis AFAIR (and it was not from mailu repo)
I found out that docker is missing mailu images, then info about ghcr.io, then tried to move those images to docker mirror, but there was no ghcr mailu 1.9.26 tag (only 1.9). Eventually I needed to find those in cache, retag and upload to my stage repo where I could rewrite them to docker.io so helm could start pods again
For now probably any change to helm settings will kill mailu instance, ale probably it's not possible to install anything <1.0

@fastlorenzo
Copy link
Collaborator Author

fastlorenzo commented Apr 13, 2023

@dominch did you try this: #213 (comment) (using chart version 0.3.3)

@dominch
Copy link

dominch commented Apr 13, 2023

the real problem with ghcr is that You can't easily list version tags.
I could find that there is 1.9 only there so had no idea if that is 1.9.latest or something like 1.9.0
That's why I decided to look for cached images and put them on mirror, but sure, let's upgrade it to 1.9.50

edit: i just tried to change those, it want's roundcube db password secret which I don't have on 1.9.26

@kooskaspers
Copy link

@kooskaspers if using MySQL, you can do a mysqldump on the old instance, then deploy the new version of the chart, and restore the dump.

Thanks for helping out @fastlorenzo. Well I'm using the built-in SQLite database of Mailu. So I'm not sure exactly what you mean with "Running the upgrade will remove the existing MySQL deployment and data!". I think that's only applicable for those who are using external databases.

So for the ones using the built-in SQLite database: from which pod on the old instance should I make use of the mysqldump command to create my dump? Because I'm not seeying any specific pod regarding a database:

image

@fastlorenzo
Copy link
Collaborator Author

For SQLite you can just backup your sqlite file, to restore it's as easy as copying it back :)

@UbiquitousBear
Copy link

@fastlorenzo - it doesn't look like there's any mention of sqlite in the new values.yaml, and no mention in the README or migration guide; does that mean it's not supported?

@dominch
Copy link

dominch commented Apr 14, 2023

@fastlorenzo - it doesn't look like there's any mention of sqlite in the new values.yaml, and no mention in the README or migration guide; does that mean it's not supported?

that means You have simply sqlite unless anything else was configured
sqlite means simple file

@kooskaspers
Copy link

For SQLite you can just backup your sqlite file, to restore it's as easy as copying it back :)

Works like a charm. Finally performed the upgrade from 0.3.3 to 1.2.0. Went flawless.
Thank you for the effort you put into these releases.

@fastlorenzo
Copy link
Collaborator Author

For SQLite you can just backup your sqlite file, to restore it's as easy as copying it back :)

Works like a charm. Finally performed the upgrade from 0.3.3 to 1.2.0. Went flawless.
Thank you for the effort you put into these releases.

Thanks for the feedback!

@mglants
Copy link

mglants commented Jul 11, 2023

After update getting errors like while sending PROXY protocol header to upstream

2023/07/11 06:29:52 [error] 16#16: *4640 send() failed (113: Host is unreachable) while sending PROXY protocol header to upstream, client: 192.168.51.254, server: 0.0.0.0:993, login: "[email protected]", upstream: 192.168.40.1:143

Connection to storage server failed.
Server Error: AUTHENTICATE PLAIN: * BAD internal server error errors in webmail

@nextgens nextgens unpinned this issue Oct 24, 2023
@nextgens nextgens pinned this issue Oct 24, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
documentation Improvements or additions to documentation
Projects
None yet
Development

No branches or pull requests

5 participants