Flag broken updater on every refresh attempt #150
Conversation
There was a problem hiding this comment.
Both this PR and Uninett/zinolib#71 address only a second part of Uninett/zinolib#68. The first part concerns EventManager and it is not fixed yet. So make sure that merging those two will not close Uninett/zinolib#68.
Also see RFC below
This I do not understand. Issue #68 is two different problems. |
If you merge this PR then Uninett/zinolib#68 will be automatically closed. See your first comment (PR description) here. Change "Closes #XXX" to "For #XXX" so that GitHub does not close Uninett/zinolib#68 automatically |
| while True: | ||
| if current_app.updater is None: | ||
| try: | ||
| connect_to_updatehandler() | ||
| except NotConnectedError as e: | ||
| raise LostConnectionError("Could not establish connection to UpdateHandler") from e | ||
| updated = current_app.updater.get_event_update() |
There was a problem hiding this comment.
| while True: | |
| if current_app.updater is None: | |
| try: | |
| connect_to_updatehandler() | |
| except NotConnectedError as e: | |
| raise LostConnectionError("Could not establish connection to UpdateHandler") from e | |
| updated = current_app.updater.get_event_update() | |
| while True: | |
| if current_app.updater is None: | |
| try: | |
| connect_to_updatehandler() | |
| except NotConnectedError as e: | |
| raise LostConnectionError("Could not establish connection to UpdateHandler") from e | |
| updated = current_app.updater.get_event_update() |
Is there a specific reason why current_app.updater is None-check should be run on every iteration? Could be moved to before the while-loop instead
There was a problem hiding this comment.
It used to be before the while loop but that was not sufficient, the connection can be lost during the loop!
For Uninett/zinolib#68 together with Uninett/zinolib#71