Hi Filip,
On 18.05.2012 21:23, fha...@apache.org wrote:
Author: fhanik
Date: Fri May 18 19:23:12 2012
New Revision: 1340215
URL: http://svn.apache.org/viewvc?rev=1340215&view=rev
Log:
https://issues.apache.org/bugzilla/show_bug.cgi?id=52858
https://issues.apache.org/bugzilla/show_bug.cgi?id=53138
Fix broken NIO sendfile download
Fix high CPU usage due to registered OP_READ for sendfile processes
Modified:
tomcat/trunk/java/org/apache/tomcat/util/net/NioEndpoint.java
Modified: tomcat/trunk/java/org/apache/tomcat/util/net/NioEndpoint.java
URL:
http://svn.apache.org/viewvc/tomcat/trunk/java/org/apache/tomcat/util/net/NioEndpoint.java?rev=1340215&r1=1340214&r2=1340215&view=diff
==============================================================================
--- tomcat/trunk/java/org/apache/tomcat/util/net/NioEndpoint.java (original)
+++ tomcat/trunk/java/org/apache/tomcat/util/net/NioEndpoint.java Fri May 18
19:23:12 2012
@@ -1278,9 +1278,13 @@ public class NioEndpoint extends Abstrac
public boolean processSendfile(SelectionKey sk, KeyAttachment
attachment, boolean reg, boolean event) {
NioChannel sc = null;
+ if (log.isTraceEnabled()) {
+ log.trace("["+new java.sql.Date(System.currentTimeMillis()).toGMTString()+"]
Processing send file. ["+sk+"] ");
+ }
Was that intentional? I'd say the timestamp should be provided by the
log framework and not by java.sql.Date. But maybe the whole message is
just a leftover from debugging the issue.
The same for the TC 7 backport.
try {
unreg(sk, attachment, sk.readyOps());
SendfileData sd = attachment.getSendfileData();
+ //setup the file channel
if ( sd.fchannel == null ) {
File f = new File(sd.fileName);
if ( !f.exists() ) {
...
Regards,
Rainer
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@tomcat.apache.org
For additional commands, e-mail: dev-h...@tomcat.apache.org
