Replies: 1 comment
-
|
This has been resolved in a post IBC PR : #135 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
I have a small observation. Not sure it is worth addressing in any shape or form, but just mentioning it to see other's opinions. Maybe it is just me not used to this paradigm...
Because of the nature how writers are conceptually detached from stream creation and deletion, it easily happens (at least to me) that readers stay connected to a flow which had been re-created. Imagine you have "well behaving writer application" which creates a flow, writes data to it for some time and when it is done, it removes the flow. Now if a reader connects there during this session, keeps running and this writer is later re-started, it now creates a flow, which has the same ID, but is technically a different flow, so previous reader won't get data from this new writer session.
Not sure what to do about it, though. Either the encouraged pattern should be for writers to never "clean up" flows after themselves (by calling
mxlDestroyFlow) and only leave that to flow garbage collection, or maybe the flow readers need to be notified (by read failure) when the flow they are reading is destroyed? Opinions?Beta Was this translation helpful? Give feedback.
All reactions