tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bugzi...@apache.org
Subject DO NOT REPLY [Bug 46209] Findbugs report on Tomcat-Pool
Date Fri, 14 Nov 2008 12:04:15 GMT
https://issues.apache.org/bugzilla/show_bug.cgi?id=46209





--- Comment #4 from Sebb <sebb@apache.org>  2008-11-14 04:04:13 PST ---
There are quite a few cases of class variables that are not final, even though
they are set in the constructor and not actually changed thereafter.

For example, 
ConnectionPool:pool
FairBlockingQueue:lock, items, waiters

Wherever possible, variables should be made final to ensure thread-safety.

There are some variables that are protected, even though there is already an
access method for them.

For example, DataSourceProxy:driver
In this case, the variable should definitely be private, as access needs to be
synchronized.

PooledConnection:counter is volatile, presumably instead of synchronising it.
However, volatile does not protect against the read-modify-write involved in
processing the counter++ statement.
Same applies to PoolProperties:poolCounter
Also, both of these are protected. I think they should be final.

It would be useful to use Javadoc to state the class thread-safety or
otherwise, and to add details of how fields are synchronised, e.g. using the
@GuardedBy annotation.

It's confusing to have aliases such as:
final ReentrantLock lock = FairBlockingQueue.this.lock;
when the code works just the same without.

Some source files still have the following copyright:
Copyright: Copyright (c) 2008 Filip Hanik


-- 
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tomcat.apache.org
For additional commands, e-mail: dev-help@tomcat.apache.org


Mime
View raw message