zookeeper-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From zodvik <...@git.apache.org>
Subject [GitHub] zookeeper issue #307: ZOOKEEPER-2770 ZooKeeper slow operation log
Date Fri, 09 Mar 2018 07:32:26 GMT
Github user zodvik commented on the issue:

    https://github.com/apache/zookeeper/pull/307
  
    > reasons why any given read or write operation may become slow: a software bug, a
protocol problem, a hardware issue with the commit log(s), a network issue. If the problem
is constant it is trivial to come to an understanding of the cause. However in order to diagnose
intermittent problems we often don't know where, or when, to begin looking. We need some sort
of timestamped indication of the problem
    
    IMO, slow query log is useful for those systems were the response times can vary due to
type of query (long scans, improper index, poor query plan, etc.). For the type of issues
mentioned, we can expect all operations to be impacted and fill up the log. 
    
    If the intent is to get an time stamp indication of the problem, a metric reporting mechanism
will be more useful. If we have rolling interval percentiles (per command?) published either
to `stat` or `jmx`, that would give a good visibility.



---

Mime
View raw message