[jira] [Commented] (LOG4J2-1125) Reuse StringBuilder to improve performance for PatternLayout

[ https://issues.apache.org/jira/browse/LOG4J2-1125?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14936463#comment-14936463 ] Dmitri Blinov commented on LOG4J2-1125: --- As it was reasonably pointed out in [http

Re: Jenkins build became unstable: Log4j 2.x #1424

Done that now. On Tue, Sep 29, 2015 at 6:04 PM, Gary Gregory wrote: > Increase the test's timeout? > > Gary > > On Tue, Sep 29, 2015 at 8:42 AM, Mikael Ståldal > wrote: > >> It works now. But there may be a timing issue, the test will fail if >> Jenkins is too slow. >> >> On Tue, Sep 29, 2015 a

[jira] [Commented] (LOG4J2-954) log4j-slf4j-impl 2.1 leads to errors

[ https://issues.apache.org/jira/browse/LOG4J2-954?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14936777#comment-14936777 ] Anton Sherstiuk commented on LOG4J2-954: I had a similar problem. Here's a workaro

[jira] [Commented] (LOG4J2-1125) Reuse StringBuilder to improve performance for PatternLayout

[ https://issues.apache.org/jira/browse/LOG4J2-1125?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938781#comment-14938781 ] Remko Popma commented on LOG4J2-1125: - [~dmitri_blinov] Any kind of pooling has to be

[jira] [Created] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

Remko Popma created LOG4J2-1142: --- Summary: ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers Key: LOG4J2-1142 URL: https://issues.apache.org/jira/browse/LOG4J2-114

[jira] [Commented] (LOG4J2-1125) Reuse StringBuilder to improve performance for PatternLayout

[ https://issues.apache.org/jira/browse/LOG4J2-1125?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938909#comment-14938909 ] Remko Popma commented on LOG4J2-1125: - I raised LOG4J2-1142 to track this issue. > R

[jira] [Updated] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Remko Popma updated LOG4J2-1142: Reporter: Dmitri Blinov (was: Remko Popma) > ThreadLocals in Layout implementations should be non-

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938933#comment-14938933 ] Ralph Goers commented on LOG4J2-1142: - Remko, Can you just create a pool in each Patt

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938935#comment-14938935 ] Gary Gregory commented on LOG4J2-1142: -- Why not just do a {{threadLocal.remove()}} w

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938941#comment-14938941 ] Ralph Goers commented on LOG4J2-1142: - You can't do it when the appender is stopped.

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938952#comment-14938952 ] Remko Popma commented on LOG4J2-1142: - Gary, {{ThreadLocal.remove()}} only removes th

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14938991#comment-14938991 ] Remko Popma commented on LOG4J2-1142: - Ralph, it may be possible. The pool would have

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14939050#comment-14939050 ] Gary Gregory commented on LOG4J2-1142: -- Yeah, the pool seems complicated in comparis

Bad PatternLayout site docs

The site docs at https://logging.apache.org/log4j/2.x/manual/layouts.html#PatternLayout shows this conversion pattern twice: enc{pattern} encode{pattern> and enc{pattern} encode{pattern} Something's not right! :-( -- E-Mail: garydgreg...@gmail.com | ggreg...@apache.org Java Persistence with

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14939080#comment-14939080 ] Ralph Goers commented on LOG4J2-1142: - In a webapp create an object and place it in a

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14939087#comment-14939087 ] Gary Gregory commented on LOG4J2-1142: -- And then Tomcat calls {{remove()}} on all th

Pattern layout and Markers

Topic 1 Let's say I want my logs to look like this: Date [thread] [loggerName] [MarkerName] Message But most of the time I do not have a marker so I end up with a lot of lame looking log entries with []. Instead I'd rather see nothing. I could use a replace like this (not tested): %replace{[%

Re: Pattern layout and Markers

Topic 1 This is quite a common use case. I have no problem with %equals but %notEmpty{[%marker]} would be nicer for this use case - it would mean don’t print anything if the variables in the pattern are empty strings. Ralph > On Sep 30, 2015, at 5:15 PM, Gary Gregory wrote: > > Topic 1 > >

[jira] [Commented] (LOG4J2-1141) resources/log4j2.xml is not loaded if a subproject contains another config file

[ https://issues.apache.org/jira/browse/LOG4J2-1141?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14939208#comment-14939208 ] LC commented on LOG4J2-1141: Here are the relevant {{StatusLogger}} traces from the logs with

[jira] [Updated] (LOG4J2-1141) resources/log4j2.xml is not loaded if a subproject contains another config file

[ https://issues.apache.org/jira/browse/LOG4J2-1141?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] LC updated LOG4J2-1141: --- Description: {code:title=build.gradle} buildscript { repositories.mavenCentral() dependencies.classpath 'org

[jira] [Updated] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ralph Goers updated LOG4J2-1142: Attachment: LOG4J2-1142.patch Remko, I have created a patch that uses an object pool instead of thr

[jira] [Updated] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ralph Goers updated LOG4J2-1142: Attachment: (was: LOG4J2-1142.patch) > ThreadLocals in Layout implementations should be non-sta

[jira] [Updated] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ralph Goers updated LOG4J2-1142: Attachment: LOG4J2-1142.patch > ThreadLocals in Layout implementations should be non-static to prev

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14939346#comment-14939346 ] Ralph Goers commented on LOG4J2-1142: - Gary, please take a look at the patch I create

Jenkins build became unstable: Log4j 2.x #1427

See - To unsubscribe, e-mail: log4j-dev-unsubscr...@logging.apache.org For additional commands, e-mail: log4j-dev-h...@logging.apache.org

Re: logging-log4j2 git commit: Add org.apache.logging.log4j.core.layout.PatternLayoutTest.testRegexEmptyMarker().

I guess this unit test needs comments at the top. There are tests that are sensitive to the line number of where they are being logged. So adding or removing lines will break those tests. Ralph > On Sep 30, 2015, at 10:03 PM, ggreg...@apache.org wrote: > > Repository: logging-log4j2 > Updated

Re: Pattern layout and Markers

I was in the middle of addressing the easier of my topics by creating a new class MarkerNamePatternConverter and unit test when I looked at our current %marker docs that say: "The name of the marker, if one is present." So I'm going to call our current implementation a bug because we do NOT use t

[jira] [Commented] (LOG4J2-1142) ThreadLocals in Layout implementations should be non-static to prevent memory leaks in web containers

[ https://issues.apache.org/jira/browse/LOG4J2-1142?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14939389#comment-14939389 ] Gary Gregory commented on LOG4J2-1142: -- I took a very brief look (it's late here, so

Re: logging-log4j2 git commit: Add org.apache.logging.log4j.core.layout.PatternLayoutTest.testRegexEmptyMarker().

Oops, my bad, fixed. Thank you, Gary On Wed, Sep 30, 2015 at 11:12 PM, Ralph Goers wrote: > I guess this unit test needs comments at the top. There are tests that > are sensitive to the line number of where they are being logged. So adding > or removing lines will break those tests. > > Ralph

Jenkins build is still unstable: Log4j 2.x #1428

See - To unsubscribe, e-mail: log4j-dev-unsubscr...@logging.apache.org For additional commands, e-mail: log4j-dev-h...@logging.apache.org