On 14/05/18 15:18, Konstantin Kolinko wrote:
> 2018-05-14 16:45 GMT+03:00  <ma...@apache.org>:
>> Author: markt
>> Date: Mon May 14 13:45:05 2018
>> New Revision: 1831568
>> URL: http://svn.apache.org/viewvc?rev=1831568&view=rev
>> Log:
>> Fix https://bz.apache.org/bugzilla/show_bug.cgi?id=62371
>> Improve logging of Host validation failures.
>> Modified:
>>     tomcat/trunk/java/org/apache/coyote/AbstractProcessor.java
>>     tomcat/trunk/java/org/apache/coyote/LocalStrings.properties
>>     tomcat/trunk/java/org/apache/coyote/http11/Http11Processor.java
>>     tomcat/trunk/webapps/docs/changelog.xml
> Mark,
> 1) Documentation.
> The cases when UserDataHelper.CONFIG is used are listed at "System
> properties" page,
> http://tomcat.apache.org/tomcat-9.0-doc/config/systemprops.html#Logging

Tx for the heads up. I'll get that updated.

> 2) I wonder: there are many instances of Processor? One for each
> request processing thread?
> I wonder whether this implemenation of silencing with UserDataHelper
> works, as there are many copies of it. Though their total count is
> limited, so the log message will be silenced eventually.
> Looking at the history, such handling of UserDataHelper is here since
> r1489437 (5 years ago), updated at r1655973.

I noticed that too. I opted for the status quo. Thinking about it, the
issue is that if you make it static there will be a single instance
shared by all sub-classes. That isn't ideal either. I guess it could be
moved to the protocol but I haven't looked at the code to see how messy
(or not) that gets.


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

Reply via email to