-
Notifications
You must be signed in to change notification settings - Fork 17
Conversation
All the endpoints are actually functioning now.
* Add server.Close to the signal processing * Demote the queue full message to the debug level * Add a channel closed channel for increased safety
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Apologies that I'm suggesting some fairly drastic changes, but I think it would allow a much simpler implementation.
@faec @cmacknz I propose we work on the server implementation in iterations and take more reasonable approach: if something does not work or works slow – we investigate and adjust the implementation. This PR is not supposed to be a final frozen version of the server and I see a lot of nitpicking going on in the review that's blocking the very first implementation. I think switching to |
I can understand if you prefer another approach, but I don't think that removing several hundred lines of synchronization code is "nitpicking"... my point was rather that the first implementation should be smaller and simpler, since all the extra synchronization machinery isn't needed for any current use cases. I don't really think Craig's suggestions are nitpicking either, as some of the issues with the conditions and channels could let a single unresponsive client back up channels for the whole server... |
That's exactly the reason why I addressed those concerns switching to Okay, I understand your point and since it's a blocker from your perspective I'll address it once I'm back and then ask for another review. Just didn't want to block any related shipper work for almost 2 weeks. |
Decision made that This can be changed in a follow up PR if it is more convenient. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good, left a few minor comments that I don't think are blocking.
Thanks for addressing all the feedback!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
One minor concern, otherwise looks good thanks!
* Return error when UUID does not match * Send even 0 value for the persisted index
All the endpoints are actually functioning now.
This PR has no validation for incoming events, this will be addressed in a follow up PR #79.
Related to #34