Content
View differences
Updated by Pavel Balashou over 2 years ago
\- Introduce healthy/unhealthy property for storages.
- It can be a set of columns related to a storage which are able to answer at least the following questions:
- Is the storage healthy?
- If not healthy, then since when?
- If not healthy, then what is the problem?
- If not healthy, then have admins been notified already?
\- In case of the job badly behaviour set the state to be unhealthy and notify admin(s).
\- If something goes wrong with the job
- It is an expected issue that admins can resolve => we mark storage as unhealthy and notify admins with probable causes and steps how to fix it regularly till the storage is healthy again.
- It is not an expected issue that can or cannot be resolved by admins => we fail the job with an exception, mark it as unhealthy and anyway notify admins about the problem.
\- When an unhealthy storage gets back to be healthy?
- When the storage synchronization passes without unexpected behavior we mark it as healthy again.
- Do we need to notify admins about cured storages?
How to notify admins about ill storages?
- Through email
- Not on every synchronization failure, but on schedule. For instance: on first transition to unhealthy state and then once per day.
- One possible option is to have a separate cron job which once per day notifies about every unhealthy storages.
* [ ] Have an admin informed via email about storage getting unhealthy and storage getting back healthy
* [ ] Expose storage health status in OpenProject UI.
**As** a admin
**I want to** _\[enter objective\]_
**so that** _\[enter desired result\]_
**Acceptance criteria**
* _\[enter acceptance criteria\]_
* _..._
- It can be a set of columns related to a storage which are able to answer at least the following questions:
- Is the storage healthy?
- If not healthy, then since when?
- If not healthy, then what is the problem?
- If not healthy, then have admins been notified already?
\- In case of the job badly behaviour set the state to be unhealthy and notify admin(s).
\- If something goes wrong with the job
- It is an expected issue that admins can resolve => we mark storage as unhealthy and notify admins with probable causes and steps how to fix it regularly till the storage is healthy again.
- It is not an expected issue that can or cannot be resolved by admins => we fail the job with an exception, mark it as unhealthy and anyway notify admins about the problem.
\- When an unhealthy storage gets back to be healthy?
- When the storage synchronization passes without unexpected behavior we mark it as healthy again.
- Do we need to notify admins about cured storages?
How to notify admins about ill storages?
- Through email
- Not on every synchronization failure, but on schedule. For instance: on first transition to unhealthy state and then once per day.
- One possible option is to have a separate cron job which once per day notifies about every unhealthy storages.
* [ ] Have an admin informed via email about storage getting unhealthy and storage getting back healthy
* [ ] Expose storage health status in OpenProject UI.
**As** a admin
**I want to** _\[enter objective\]_
**so that** _\[enter desired result\]_
**Acceptance criteria**
* _\[enter acceptance criteria\]_
* _..._