Slack digest for #dev - 2020-09-29

[Apache Pulsar Slack <ap...@gmail.com>]

2020-09-28 11:13:28 UTC - Lari Hotari: @Addison Higham thanks for the suggestion. I looked into the way it's done in BrokerService. However since ClientCnx already had a way to handle timeouts for "ordinary" command requests, I decided to extend that to support also the other type of requests that didn't have timeout handling. I created PR <https://github.com/apache/pulsar/pull/8149> with that solution.

&gt; The other thing I would wonder if is there is a nice clean way to cancel all requests once we close a reader. It is a bit tricky since the requests get multiplexed across a socket, but perhaps there exists an easy way to know the outstanding request ids for a given consumer so they could then be cancelled
Yes, this is a remaining challenge. It would be expected to release all resources (including the pending requests) that are related to a specific consumer when it gets closed. This might not be even so critical after the timeouts are handled for all request types like there is in the PR that I created. At least I don't find a compelling reason for my usecase.
----
2020-09-28 11:28:06 UTC - Emil: @Emil has joined the channel
----
2020-09-28 13:36:42 UTC - el akroudi abdessamad: Hello Guys,
which design pattern is used to build the PulsarAdminTool ?
----
2020-09-29 06:22:34 UTC - Charvak Patel: @Charvak Patel has joined the channel
----