What's coming in 7.4
-
OK, we pushed out 7.4.0 as unstable today. Please report any early issues here. You will also find updates to weblate, peertube once updated to 7.4.0.
As a heads up, this is quite a big update since it upgrades all the databases. Always have remote backups!
Not all features are done yet, they will be part of 7.4.1. Just wanted to get the db updates out early.
-
OK, we pushed out 7.4.0 as unstable today. Please report any early issues here. You will also find updates to weblate, peertube once updated to 7.4.0.
As a heads up, this is quite a big update since it upgrades all the databases. Always have remote backups!
Not all features are done yet, they will be part of 7.4.1. Just wanted to get the db updates out early.
@girish Good for you! Usually, but not always, "slow and steady wins the race." In other words, I like to see small, frequent releases; I abhor huge, infrequent releases. I am not alone in this. See, for example, The Art of Shipping Early and Often.
-
We have also added a first version of an OpenID Connect provider to Cloudron. Version 7.4 will ship with this to figure out how well the various integrations in the apps work out. Hopefully this will provider a smoother single-sign-on experience in the future.
So far in the tests, Nextcloud and RocketChat worked as such, but both have their own oddities in how accounts are managed, so for existing installations there will likely be no automatic upgrade path from LDAP to OpenID Connect, but lets see.
@nebulon I've done the upgrade today and was wondering about one thing (and looking at your screenshot it seems to be the same on your system). While all the endpoints are on the
mysubdomain, the discovery document is served from the base domain. Technically the discovery document could also be served from the my domain. -
@nebulon I've done the upgrade today and was wondering about one thing (and looking at your screenshot it seems to be the same on your system). While all the endpoints are on the
mysubdomain, the discovery document is served from the base domain. Technically the discovery document could also be served from the my domain. -
@fbartels it should actually also be served up by the my. domain. I wasn't quite sure what the most compatible way is.
-
Danish / German translations are done
-
One thing i have noticed since upgrading to 7.4 is that my backups now are failling:
Error uploading snapshot/app_0a8cabb4-9f38-4826-b1ff-0bd5cff5ff5b.tar.gz.enc. Message: We encountered an internal error, please try again.: cause(stat /data/bucket-name: host is down) HTTP Code: InternalErrorMy backup goes to a minio instance with a hetzner storage space mounted. The path looks like it's coming from the minio host. No changes have been done on the minio end, but I have also not yet checked it's logs. The cloudron backup list only contains one backup tagged with 7.4.0 and the error is the same for the automated and manually triggered backups.
Currently on mobile will need to check logs when at my desk later.
-
One thing i have noticed since upgrading to 7.4 is that my backups now are failling:
Error uploading snapshot/app_0a8cabb4-9f38-4826-b1ff-0bd5cff5ff5b.tar.gz.enc. Message: We encountered an internal error, please try again.: cause(stat /data/bucket-name: host is down) HTTP Code: InternalErrorMy backup goes to a minio instance with a hetzner storage space mounted. The path looks like it's coming from the minio host. No changes have been done on the minio end, but I have also not yet checked it's logs. The cloudron backup list only contains one backup tagged with 7.4.0 and the error is the same for the automated and manually triggered backups.
Currently on mobile will need to check logs when at my desk later.
-
One thing i have noticed since upgrading to 7.4 is that my backups now are failling:
Error uploading snapshot/app_0a8cabb4-9f38-4826-b1ff-0bd5cff5ff5b.tar.gz.enc. Message: We encountered an internal error, please try again.: cause(stat /data/bucket-name: host is down) HTTP Code: InternalErrorMy backup goes to a minio instance with a hetzner storage space mounted. The path looks like it's coming from the minio host. No changes have been done on the minio end, but I have also not yet checked it's logs. The cloudron backup list only contains one backup tagged with 7.4.0 and the error is the same for the automated and manually triggered backups.
Currently on mobile will need to check logs when at my desk later.
@fbartels said in What's coming in 7.4:
One thing i have noticed since upgrading to 7.4 is that my backups now are failling:
.
···
Error uploading snapshot/app_0a8cabb4-9f38-4826-b1ff-0bd5cff5ff5b.tar.gz.enc. Message: We encountered an internal error, please try again.: cause(stat /data/bucket-name: host is down) HTTP Code: InternalError
···My backup goes to a minio instance with a hetzner storage space mounted. The path looks like it's coming from the minio host. No changes have been done on the minio end, but I have also not yet checked it's logs. The cloudron backup list only contains one backup tagged with 7.4.0 and the error is the same for the automated and manually triggered backups.
Currently on mobile will need to check logs when at my desk later.
Thanks to the brave soul @fbartels for trying the update to 7.4 and letting us know how it went. Encouraged, I did the same and it has gone very smoothly. The backup takes quite a while, but it all seems to be going well now.
Well done, Cloudron team!
I am curious why Cloudron use a Kanban instead of a Gantt for tracking progress.
-
@fbartels we haven't really made any changes around backups for 7.4. Let us know what you find but this seems like some error coming from the server.
@girish said in What's coming in 7.4:
Let us know what you find but this seems like some error coming from the server.
This indeed turned out to be an issue on the minio side (backup is still running but I expect it will finish this time). The smb mount died on the minio host, so after remounting minio was operational again.
-
The night after my server auto-updated to 7.4, I got an error notification regarding email config that has no more info when clicked on it. It happened only once so far but just wanted to mention it. Other than that, 7.4 seems to be working fine for me.
Email is not configured properly -
The night after my server auto-updated to 7.4, I got an error notification regarding email config that has no more info when clicked on it. It happened only once so far but just wanted to mention it. Other than that, 7.4 seems to be working fine for me.
Email is not configured properly@humptydumpty the initial regression thread is at https://forum.cloudron.io/topic/8913/email-is-not-configured-properly-since-update-to-cloudron-7-4-0/2
-
With Auto updates on, I was force updated to 7.4.0
Can you please disable that functionality?
Unstable updates should not occur without manual involvement.
I should not have to choose between installing an unstable update or manually update all my cloudron apps, that list is quite high at this point.
-
With Auto updates on, I was force updated to 7.4.0
Can you please disable that functionality?
Unstable updates should not occur without manual involvement.
I should not have to choose between installing an unstable update or manually update all my cloudron apps, that list is quite high at this point.
@privsec said in What's coming in 7.4:
With Auto updates on, I was force updated to 7.4.0
Can you please disable that functionality?
Unstable updates should not occur without manual involvement.
I should not have to choose between installing an unstable update or manually update all my cloudron apps, that list is quite high at this point.
On a similar note, after the update, VaultWarden showed a pre-release update was available for installation. I think I am going to have that green up-arrow there till a stable release happens.
Perhaps we could have a two tone up arrow system, where neon green was for unstable releases and green for ordinary releases. We could toggle notifications for a permutation of these.
-
@privsec said in What's coming in 7.4:
With Auto updates on, I was force updated to 7.4.0
Can you please disable that functionality?
Unstable updates should not occur without manual involvement.
I should not have to choose between installing an unstable update or manually update all my cloudron apps, that list is quite high at this point.
On a similar note, after the update, VaultWarden showed a pre-release update was available for installation. I think I am going to have that green up-arrow there till a stable release happens.
Perhaps we could have a two tone up arrow system, where neon green was for unstable releases and green for ordinary releases. We could toggle notifications for a permutation of these.
@LoudLemur Yes, for vaultwarden, there seems to be some issue which is causing MySQL to crash. So, we have pushed a pre-release for those who are willing to test it. See https://forum.cloudron.io/topic/8895/mysql-not-responding-with-vaultwarden-installed
-
With Auto updates on, I was force updated to 7.4.0
Can you please disable that functionality?
Unstable updates should not occur without manual involvement.
I should not have to choose between installing an unstable update or manually update all my cloudron apps, that list is quite high at this point.
-
@LoudLemur Yes, for vaultwarden, there seems to be some issue which is causing MySQL to crash. So, we have pushed a pre-release for those who are willing to test it. See https://forum.cloudron.io/topic/8895/mysql-not-responding-with-vaultwarden-installed
-
It seems that the time display was changed in Backups & Email Log. Screenshots below. Can we please have this either be 24 hour time again or else at least add AM/PM if meant to be 12-hour format? It's confusing at a glance since it's missing AM/PM tags it is presumed to be 24-hour but that's incorrect.
--
Dustin Dauncey
www.d19.ca -
It seems that the time display was changed in Backups & Email Log. Screenshots below. Can we please have this either be 24 hour time again or else at least add AM/PM if meant to be 12-hour format? It's confusing at a glance since it's missing AM/PM tags it is presumed to be 24-hour but that's incorrect.
-
