[GH-ISSUE #98] [Request] Use semver rules when incrementing ambar versions #99

Closed
opened 2026-02-27 15:54:59 +03:00 by kerem · 2 comments
Owner

Originally created by @lazypower on GitHub (Nov 26, 2017).
Original GitHub issue: https://github.com/RD17/ambar/issues/98

From the 1.0.0 => 1.3.0 jump introduces non-backwards compatible changes. This is in reality a major version bump and should be labeled as a 2.0 change since it requires a full data reset on the instance.

I was caught off guard by this when issuing python ambar.py update and it happily upgraded my docker containers, but failed to inform me the indexes had changed which I observed in #92.

I attempted a rollback and found no success. This broke around 1TB of indexed and "archived" content in my CE instance which damages my trust in the ambar project progression because a breaking change was introduced at a non-major version barrier.

I'm sorry to relay this as an issue but I didn't find a mailing list to broach this topic, and I'd love to regain some trust in the ambar project, as it really is great stuff. I just need to know my data is mostly safe in the hands of the project maintainers, and that dropping -all- data wont be required without a major revision upgrade so I can watch for and ensure i only pull updates prior to that happening since the management script is the recommended method to installation/operation.

Originally created by @lazypower on GitHub (Nov 26, 2017). Original GitHub issue: https://github.com/RD17/ambar/issues/98 From the 1.0.0 => 1.3.0 jump introduces non-backwards compatible changes. This is in reality a major version bump and should be labeled as a 2.0 change since it requires a full data reset on the instance. I was caught off guard by this when issuing python ambar.py update and it happily upgraded my docker containers, but failed to inform me the indexes had changed which I observed in #92. I attempted a rollback and found no success. This broke around 1TB of indexed and "archived" content in my CE instance which damages my trust in the ambar project progression because a breaking change was introduced at a non-major version barrier. I'm sorry to relay this as an issue but I didn't find a mailing list to broach this topic, and I'd love to regain some trust in the ambar project, as it really is great stuff. I just need to know my data is mostly safe in the hands of the project maintainers, and that dropping -all- data wont be required without a major revision upgrade so I can watch for and ensure i only pull updates prior to that happening since the management script is the recommended method to installation/operation.
kerem 2026-02-27 15:54:59 +03:00
Author
Owner

@sochix commented on GitHub (Nov 28, 2017):

@chuckbutler you're right, will keep that in mind and change update script.

P.s. as for now you can check update instructions in changelog.md

<!-- gh-comment-id:347528850 --> @sochix commented on GitHub (Nov 28, 2017): @chuckbutler you're right, will keep that in mind and change update script. P.s. as for now you can check update instructions in changelog.md
Author
Owner

@lazypower commented on GitHub (Nov 30, 2017):

Thank you for the response and acknowledgement. :)

<!-- gh-comment-id:348042193 --> @lazypower commented on GitHub (Nov 30, 2017): Thank you for the response and acknowledgement. :)
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
starred/ambar#99
No description provided.