cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Robert Stupp (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-13396) Cassandra 3.10: ClassCastException in ThreadAwareSecurityManager
Date Fri, 31 Mar 2017 15:57:41 GMT


Robert Stupp commented on CASSANDRA-13396:

[~apassiou], C* is meant to run as a standalone application using the dependencies that are
in the {{lib/}} folder. Any change to those dependencies and the way C* is started, is basically
up to the person who changes the dependencies. We can of course talk about using a different
logger implementation instead of logback and discuss the pros and cons. But that is IMO way
beyond an {{instanceof}} check.

I'm generally concerned about stability and hidden performance issues and a change to a (logger
implementation) library, which is nearly everywhere in the hot code path. Mean, we use logback
now for a really long time - but we have no test nor production experience running something
else. One example: one thing that may happen is some hidden contention in that logger library
causing weird outliers - people would complain that C* is slow but don't realize it's in this
case because of that change. That's one reason why we are so careful with library updates
especially in minor versions. All I'm saying is, that getting _all_ the consequences of such
a change is a lot of work.

> Cassandra 3.10: ClassCastException in ThreadAwareSecurityManager
> ----------------------------------------------------------------
>                 Key: CASSANDRA-13396
>                 URL:
>             Project: Cassandra
>          Issue Type: Wish
>            Reporter: Edward Capriolo
>            Priority: Minor

This message was sent by Atlassian JIRA

View raw message