merlimat closed issue #21442: [Bug] RateLimiter lock contention when use
precise publish rate limiter
URL: https://github.com/apache/pulsar/issues/21442
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go
lhotari commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1849380390
@Shawyeok There's now "PIP-322: Pulsar Rate Limiting Refactoring" (#21680)
to address this issue. The changes are in the draft PR #21681 . Please review!
Thanks
--
This is an aut
lhotari commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1823569304
> @lhotari Cool! I will study and verify your solution as soon as possible
after my vacation, which may be a few days.
Sounds good. Please check the blog post
https://cod
Shawyeok commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1822778672
> > Of course, we can also optimize the implementation of
`org.apache.pulsar.common.util.RateLimiter`. Perhaps an implementation based on
Compare-And-Swap (CAS) would perform bette
lhotari commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1822660666
> Hello, I am also learning pulsar recently, have you found a solution to
this problem？
@1625567290 Just curious to learn about your motivation to start learning
Pulsar here.
lhotari commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1820668270
> Of course, we can also optimize the implementation of
`org.apache.pulsar.common.util.RateLimiter`. Perhaps an implementation based on
Compare-And-Swap (CAS) would perform bett
Shawyeok commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1814781576
> > Sorry for late reply.
> > The current publish rate limiting is based on setting the `autoRead`
flag of the Netty channel to false. However, `autoRead` is just a hint in
Nett
Shawyeok commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1814780242
In Pulsar internal, there are multiple topics that depend on the same thread
for message writing (such as Netty's EventLoop and OrderedExecutor). If, for
some reason, these threads
1625567290 commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1812704939
> Sorry for late reply.
>
> The current publish rate limiting is based on setting the `autoRead` flag
of the Netty channel to false. However, `autoRead` is just a hint in N
Shawyeok commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1797947739
There is a related discuss in mail list:
https://lists.apache.org/thread/13ncst2nc311vxok1s75thl2gtnk7w1t
--
This is an automated message from the Apache Git Service.
To respond
Shawyeok commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1792001008
Sorry for late reply.
The current publish rate limiting is based on setting the `autoRead` flag of
the Netty channel to false. However, `autoRead` is just a hint in Netty, no
1625567290 commented on issue #21442:
URL: https://github.com/apache/pulsar/issues/21442#issuecomment-1782248694
and found nothing similar.
### Version
After enabled the `preciseTopicPublishRat
14 matches
Mail list logo
