You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@zookeeper.apache.org by "Brent (Jira)" <ji...@apache.org> on 2022/01/06 20:32:00 UTC

[jira] [Commented] (ZOOKEEPER-4291) ZooInspector has very high latency when operating against remote clusters

    [ https://issues.apache.org/jira/browse/ZOOKEEPER-4291?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17470182#comment-17470182 ] 

Brent commented on ZOOKEEPER-4291:
----------------------------------

A quick update since I know this ticket was filed forever ago... 

I am actively working on this and have a prototype going that seems to greatly cut down on network calls (and thereby improve performance), but I've still got a lot of testing and debugging to do.

> ZooInspector has very high latency when operating against remote clusters
> -------------------------------------------------------------------------
>
>                 Key: ZOOKEEPER-4291
>                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-4291
>             Project: ZooKeeper
>          Issue Type: Improvement
>          Components: contrib
>    Affects Versions: 3.6.2, 3.8.0, 3.7.1
>            Reporter: Brent
>            Priority: Major
>
> The ZooInspector tool performs well when connecting to a Zookeeper cluster on the same host it is being run on or to a very close host (network-wise).  However, when run against a Zookeeper cluster that is remote (such as one running in the cloud), the tool exhibits very high latency.  In my experience, at worst it can takes up to 20-30 seconds between the time a node is expanded and the time the UI is updated. 
> Upon brief inspection, this seems to be due to network calls to Zookeeper being made from the rendering thread (though this is not entirely clear). 
> I will (hopefully) be attempting to create a pull request to remedy this issue, but I wanted to report it first.



--
This message was sent by Atlassian Jira
(v8.20.1#820001)