[jira] [Created] (LOG4J2-656) Delete rolled over file

sivan created LOG4J2-656: Summary: Delete rolled over file Key: LOG4J2-656 URL: https://issues.apache.org/jira/browse/LOG4J2-656 Project: Log4j 2 Issue Type: Question Components: Appenders

[jira] [Resolved] (LOG4J2-645) Rollover date issue with rolling file appender.

[ https://issues.apache.org/jira/browse/LOG4J2-645?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] sivan resolved LOG4J2-645. -- Resolution: Fixed Fix Version/s: 2.0-rc1 This issue seems resolved. I removed the from the configurati

[jira] [Resolved] (LOG4J2-642) Mismatch in Rollover date and log entry date in the log file

[ https://issues.apache.org/jira/browse/LOG4J2-642?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] sivan resolved LOG4J2-642. -- Resolution: Fixed Fix Version/s: 2.0-rc1 This issue seems resolved. I removed the from the configurati

[jira] [Commented] (LOG4J2-645) Rollover date issue with rolling file appender.

[ https://issues.apache.org/jira/browse/LOG4J2-645?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14009302#comment-14009302 ] sivan commented on LOG4J2-645: -- This issue seems resolved. I removed the from the configura

[jira] [Commented] (LOG4J2-642) Mismatch in Rollover date and log entry date in the log file

[ https://issues.apache.org/jira/browse/LOG4J2-642?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14009305#comment-14009305 ] sivan commented on LOG4J2-642: -- This issue seems resolved. I removed the from the configura

Re: svn commit: r1597509 - /logging/log4j/log4j2/trunk/log4j-core/src/main/java/org/apache/logging/log4j/core/config/plugins/visitors/PluginVisitors.java

Clearer is better here. +1 to "registry". Gary Original message From: Matt Sicker Date:05/26/2014 19:49 (GMT-05:00) To: Log4J Developers List Subject: Re: svn commit: r1597509 - /logging/log4j/log4j2/trunk/log4j-core/src/main/java/org/apache/logging/log4j/core/config/plug

Re: Possible NPEs in org.apache.logging.log4j.core.layout.PatternLayout

Fixed in r1597669. On 25 May 2014 00:06, Matt Sicker wrote: > I added a createDefaultLayout() method to PatternLayout, but I don't think > I addressed the header/footer concern. I did, however, replace all those > calls to createLayout() with a bunch of null values as part of the > migration to

Re: svn commit: r1597509 - /logging/log4j/log4j2/trunk/log4j-core/src/main/java/org/apache/logging/log4j/core/config/plugins/visitors/PluginVisitors.java

I think you're right. On 26 May 2014 18:45, Remko Popma wrote: > Is PluginVisitorRegistry not a better name than PluginVisitors? > The convention to append "s" after another class name has many meanings, > like providing static util methods. > I think calling it a ..Registry is clearer. > > Sen

Re: PluginAttribute vs. IntPluginAttribute

The fun thing about this is that using builders, default values don't need to be specified as annotations. I'm starting to lean toward that idea more. On 26 May 2014 18:46, Gary Gregory wrote: > How would that prevent people from writing nonsensical declarations? It > seems confusing to have on

Re: svn commit: r1597494 - in /logging/log4j/log4j2/trunk: log4j-1.2-api/src/test/java/org/apache/log4j/ log4j-core/src/main/java/org/apache/logging/log4j/core/appender/db/jdbc/ log4j-core/src/main/ja

Better! Original message From: Matt Sicker Date:05/26/2014 15:27 (GMT-05:00) To: Log4J Developers List Subject: Re: svn commit: r1597494 - in /logging/log4j/log4j2/trunk: log4j-1.2-api/src/test/java/org/apache/log4j/ log4j-core/src/main/java/org/apache/logging/log4j/core/

Re: PluginAttribute vs. IntPluginAttribute

How would that prevent people from writing nonsensical declarations? It seems confusing to have one annotation with a bunch of attributes, one for each type. It feels likesome hackto me. Having specific annotations seems foolproof. Gary Original message From: Matt Sicker Date

Re: svn commit: r1597509 - /logging/log4j/log4j2/trunk/log4j-core/src/main/java/org/apache/logging/log4j/core/config/plugins/visitors/PluginVisitors.java

Is PluginVisitorRegistry not a better name than PluginVisitors? The convention to append "s" after another class name has many meanings, like providing static util methods. I think calling it a ..Registry is clearer. Sent from my iPhone > On 2014/05/26, at 12:59, mattsic...@apache.org wrote: >

[jira] [Closed] (LOG4J2-655) Add a Vagrantfile for easier testing in a GNU+Linux environment

[ https://issues.apache.org/jira/browse/LOG4J2-655?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Matt Sicker closed LOG4J2-655. -- Resolution: Done Add in r1597664. > Add a Vagrantfile for easier testing in a GNU+Linux environment >

Re: Random failure

I've removed the one that seems to fail often for you. There is one remaining similar test that starts with a sleep. That one doesn't seem to cause trouble and covers pretty much the same item so I'm ok with having just that remaining test. Sent from my iPhone > On 2014/05/27, at 1:26, Gary G

[jira] [Commented] (LOG4J2-655) Add a Vagrantfile for easier testing in a GNU+Linux environment

[ https://issues.apache.org/jira/browse/LOG4J2-655?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14009118#comment-14009118 ] Matt Sicker commented on LOG4J2-655: So far I'm adding JDK6 (openjdk-6-jdk) and Maven

[jira] [Created] (LOG4J2-655) Add a Vagrantfile for easier testing in a GNU+Linux environment

Matt Sicker created LOG4J2-655: -- Summary: Add a Vagrantfile for easier testing in a GNU+Linux environment Key: LOG4J2-655 URL: https://issues.apache.org/jira/browse/LOG4J2-655 Project: Log4j 2

Could I get access to Jenkins?

I'd like to work on setting up build profiles for Log4j2. I know Ralph has already done a bunch of set up, so this shouldn't be that difficult to extend. We could use profiles for JDK6, JDK7, and JDK8, and on different operating systems. -- Matt Sicker

Re: The PluginBuilder/Factory/Visitor/9000 refactoring is under way.

Alright, PoCs are in the trunk right now. I've only got builder classes in PatternLayout and HtmlLayout. I've also updated all the @PluginAttribute usage where defaultValue() was used before to use their appropriate typed attributes instead. On 26 May 2014 14:18, Matt Sicker wrote: > I'm workin

Re: svn commit: r1597421 - in /logging/log4j/log4j2/trunk: log4j-core/src/main/java/org/apache/logging/log4j/core/layout/Rfc5424Layout.java log4j-flume-ng/src/main/java/org/apache/logging/log4j/flume/

And here's how I'm changing it now that there's a builder class in HtmlLayout: final HtmlLayout layout = HtmlLayout.newBuilder() .withLocationInfo(includeLocation) .build(); Bam! On 25 May 2014 17:20, Matt Sicker wrote: > As I showed in the HtmlLayoutTest, I ha

Re: ConfigurablePluginBuilder vs. Builder

Done by the way. I re-used it in PluginBuilder, too, since it matches perfectly. On 26 May 2014 14:17, Matt Sicker wrote: > Sounds like a good idea to me! I'll go ahead and make the change. > > > On 26 May 2014 08:20, Gary Gregory wrote: > >> We have a new interface: >> >> public interface Con

[jira] [Commented] (LOG4J2-607) log4j2 json files do not support property substitutions

[ https://issues.apache.org/jira/browse/LOG4J2-607?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14009081#comment-14009081 ] Silviu Simen commented on LOG4J2-607: - I'm not sure if this is the same problem, but i

Re: JMH for log4j microbenchmarks

That makes it more like the LGPL which I believe would also be ok in this instance Sent from my iPhone > On May 26, 2014, at 12:16 PM, Matt Sicker wrote: > > What if it's GPL with the classpath exception? > > >> On 26 May 2014 08:16, Ralph Goers wrote: >> >> >> Sent from my iPad >> >>> O

Re: Class registration

Perhaps we can re-use the plugin infrastructure for type converters? On 26 May 2014 09:06, Gary Gregory wrote: > I am talking about (at least) a registry of type converters. > > Gary > > > On Mon, May 26, 2014 at 10:01 AM, Ralph Goers wrote: > >> Class registries? Do you mean plugins? Each jar

Re: Asking questions in commit comments.

Yikes, I forgot to forward that to the ML! Generally I ask a question in the commit message because then I have an email already sent somewhere that I can then forward or add on to. On 26 May 2014 09:00, Ralph Goers wrote: > Thanks Gary, I would have probably missed this. > > The answer to the

Re: PluginAttribute vs. IntPluginAttribute

We can choose which attribute of the annotation to use based on the type the annotation is attached to. That wouldn't be too hard. On 26 May 2014 08:34, Gary Gregory wrote: > On Mon, May 26, 2014 at 9:02 AM, Remko Popma wrote: > >> >> >> >> On Mon, May 26, 2014 at 9:58 PM, Gary Gregory wrote: >

Re: The PluginBuilder/Factory/Visitor/9000 refactoring is under way.

I'm working on PoC's for those things before I go and make widespread changes to everything. I'll follow up with one in an hour or two. On 26 May 2014 08:27, Gary Gregory wrote: > On Sun, May 25, 2014 at 11:44 PM, Matt Sicker wrote: > >> 1. I'm removing @PluginDefault and instead using an addi

Re: ConfigurablePluginBuilder vs. Builder

Sounds like a good idea to me! I'll go ahead and make the change. On 26 May 2014 08:20, Gary Gregory wrote: > We have a new interface: > > public interface ConfigurablePluginBuilder { > > /** > * Builds the plugin object after all configuration has been set. This > will use default val

[jira] [Commented] (LOG4J2-653) Add plugin builder classes as alternative to factory methods.

[ https://issues.apache.org/jira/browse/LOG4J2-653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14009039#comment-14009039 ] Matt Sicker commented on LOG4J2-653: I'm not speaking of injection for tests, but we c

Re: JMH for log4j microbenchmarks

What if it's GPL with the classpath exception? On 26 May 2014 08:16, Ralph Goers wrote: > > > Sent from my iPad > > On May 26, 2014, at 5:31 AM, Gary Gregory wrote: > > On Sun, May 25, 2014 at 5:25 PM, Gary Gregory wrote: > >> Over in Commons VFS we are not adding a pom dep to a jar because it

Re: Early Access builds for JDK 9 b13, JDK 8u20 b15 and JDK 7u60 b15 are available on java.net

I only tested JDK9b13. On 26 May 2014 05:23, Rory O'Donnell Oracle, Dublin ireland < rory.odonn...@oracle.com> wrote: > Hi Matt, > > Thanks for the update, are you testing all JDKs ? > > Rgds, Rory > On 05/23/14 10:33 PM, Matt Sicker wrote: > > Our project (log4j) built fine and tests still pa

Re: Random failure

My machine is "fast" but it is often "busy". Is there a way to make the test more deterministic? Gary On Mon, May 26, 2014 at 11:55 AM, Remko Popma wrote: > It seems that this test frequently fails spuriously on slow or busy > machines. > I will remove the test. > > > On Mon, May 26, 2014 at 1

[jira] [Commented] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008930#comment-14008930 ] Remko Popma commented on LOG4J2-636: I suspect the configuration reload might be causi

Re: Random failure

It seems that this test frequently fails spuriously on slow or busy machines. I will remove the test. On Mon, May 26, 2014 at 11:05 PM, Gary Gregory wrote: > With the latest from trunk, I get this error: > > Failed tests: > CachedClockTest.testLessThan17Millis:32 diff too large: 75 > > Tests r

Re: Don't use import sun.misc.BASE64Decoder in tests

Thank you, Gary. On Tue, May 27, 2014 at 12:25 AM, Gary Gregory wrote: > Fixed in SVN, they were only two call sites. > > Gary > > > On Mon, May 26, 2014 at 9:52 AM, Gary Gregory wrote: > >> Please use Commons Codec instead as a test-only dependency. >> >> Gary >> >> -- >> E-Mail: garydgreg...@g

log4j-perf maven layout

WRT the log4j-perf maven layout: It seems to me all of this Java code should be in src/test, not src/main. Gary -- E-Mail: garydgreg...@gmail.com | ggreg...@apache.org Java Persistence with Hibernate, Second Edition JUnit in Action, Second Edition

Re: Don't use import sun.misc.BASE64Decoder in tests

Fixed in SVN, they were only two call sites. Gary On Mon, May 26, 2014 at 9:52 AM, Gary Gregory wrote: > Please use Commons Codec instead as a test-only dependency. > > Gary > > -- > E-Mail: garydgreg...@gmail.com | ggreg...@apache.org > Java Persistence with Hibernate, Second > Edition

[jira] [Resolved] (LOG4J2-651) Log4j 2 throws ArrayIndexOutOfBoundsException

[ https://issues.apache.org/jira/browse/LOG4J2-651?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Gary Gregory resolved LOG4J2-651. - Resolution: Fixed Fix Version/s: 2.0-rc2 Fixed in trunk as reported by the user. > Log4j

[jira] [Commented] (LOG4J2-651) Log4j 2 throws ArrayIndexOutOfBoundsException

[ https://issues.apache.org/jira/browse/LOG4J2-651?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008882#comment-14008882 ] Nikita Koval commented on LOG4J2-651: - In trunk this bug was fixed. Thank you! With b

Re: Class registration

I am talking about (at least) a registry of type converters. Gary On Mon, May 26, 2014 at 10:01 AM, Ralph Goers wrote: > Class registries? Do you mean plugins? Each jar should have its own > serialized Map of plugins. > > Sent from my iPad > > On May 26, 2014, at 6:39 AM, Gary Gregory wrote:

Random failure

With the latest from trunk, I get this error: Failed tests: CachedClockTest.testLessThan17Millis:32 diff too large: 75 Tests run: 630, Failures: 1, Errors: 0, Skipped: 28 Gary -- E-Mail: garydgreg...@gmail.com | ggreg...@apache.org Java Persistence with Hibernate, Second Edition

Re: Class registration

Class registries? Do you mean plugins? Each jar should have its own serialized Map of plugins. Sent from my iPad > On May 26, 2014, at 6:39 AM, Gary Gregory wrote: > > There seems to be one or more class registries flying by in recent commits. > > How do classes in non-core module get registe

Re: Asking questions in commit comments.

Thanks Gary, I would have probably missed this. The answer to the question is it makes sense to me. An array should always be fully populated whereas it is generally OK if a single valued item is null. For example, you might have no filters but if you do have filters none of them can be null

Don't use import sun.misc.BASE64Decoder in tests

Please use Commons Codec instead as a test-only dependency. Gary -- E-Mail: garydgreg...@gmail.com | ggreg...@apache.org Java Persistence with Hibernate, Second Edition JUnit in Action, Second Edition Spring Batch in Action

[Bug 56562] RollingFileAppender creating junk files

https://issues.apache.org/bugzilla/show_bug.cgi?id=56562 --- Comment #1 from Gary Gregory --- Hello Anil, Could you use Log4j 2 (rc1 or trunk)? Version 1.2 is nearing EOL. If using 1.2, at least use 1.2.17, which is the current version. Thank you, Gary -- You are receiving this mail because:

Class registration

There seems to be one or more class registries flying by in recent commits. How do classes in non-core module get registered? Gary -- E-Mail: garydgreg...@gmail.com | ggreg...@apache.org Java Persistence with Hibernate, Second Edition JUnit in Action, Second Edit

Re: PluginAttribute vs. IntPluginAttribute

On Mon, May 26, 2014 at 9:02 AM, Remko Popma wrote: > > > > On Mon, May 26, 2014 at 9:58 PM, Gary Gregory wrote: > >> A recent commit contains: >> >> >>> +@PluginAttribute(value = "blocking", defaultValue = "true") >>> final boolean blocking, >>> >> >> Using a String default for a non

Asking questions in commit comments.

Matt, In this case, I think you need to ask these questions on the ML with an appropriate subject instead of buried in a commit message. Gary -- Forwarded message -- From: Date: Mon, May 26, 2014 at 2:02 AM Subject: svn commit: r1597517 - /logging/log4j/log4j2/trunk/log4j-core/s

Re: The PluginBuilder/Factory/Visitor/9000 refactoring is under way.

On Sun, May 25, 2014 at 11:44 PM, Matt Sicker wrote: > 1. I'm removing @PluginDefault and instead using an additional attribute > on @PluginAttribute. > Great. Please see my other thread about default primitive values and @IntPluginAttribute, and so on. 2. I'm refactoring the logic in the gian

ConfigurablePluginBuilder vs. Builder

We have a new interface: public interface ConfigurablePluginBuilder { /** * Builds the plugin object after all configuration has been set. This will use default values for any * unspecified attributes for the plugin. * * @return the configured plugin instance. */

Re: JMH for log4j microbenchmarks

Sent from my iPad > On May 26, 2014, at 5:31 AM, Gary Gregory wrote: > >> On Sun, May 25, 2014 at 5:25 PM, Gary Gregory wrote: >> Over in Commons VFS we are not adding a pom dep to a jar because it's >> license is not ASF compatible, which is ASF policy from what I understand >> and how it

[jira] [Comment Edited] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008824#comment-14008824 ] Adrian Wilford edited comment on LOG4J2-636 at 5/26/14 1:08 PM:

[jira] [Commented] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008824#comment-14008824 ] Adrian Wilford commented on LOG4J2-636: --- Yes, I did do that. I changed it from trace

[jira] [Comment Edited] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008821#comment-14008821 ] Adrian Wilford edited comment on LOG4J2-636 at 5/26/14 1:04 PM:

[jira] [Commented] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008822#comment-14008822 ] Remko Popma commented on LOG4J2-636: I see that the configuration has a monitorInterva

[jira] [Commented] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008821#comment-14008821 ] Adrian Wilford commented on LOG4J2-636: --- this is the line of code that generates the

Re: PluginAttribute vs. IntPluginAttribute

On Mon, May 26, 2014 at 9:58 PM, Gary Gregory wrote: > A recent commit contains: > > >> +@PluginAttribute(value = "blocking", defaultValue = "true") >> final boolean blocking, >> > > Using a String default for a non-string attribute seems like a punch in > the eye ;-) This really feels

[jira] [Commented] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008818#comment-14008818 ] Adrian Wilford commented on LOG4J2-636: --- p.s. I don't know how to "shut down the log

PluginAttribute vs. IntPluginAttribute

A recent commit contains: > +@PluginAttribute(value = "blocking", defaultValue = "true") > final boolean blocking, > Using a String default for a non-string attribute seems like a punch in the eye ;-) This really feels unnatural. Currently, I cannot say: @PluginAttribute(value = "b

[jira] [Commented] (LOG4J2-636) IOException: Stream Closed RollingRandomAccessFile

[ https://issues.apache.org/jira/browse/LOG4J2-636?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008817#comment-14008817 ] Adrian Wilford commented on LOG4J2-636: --- Here is the config file:

Re: JMH for log4j microbenchmarks

On Sun, May 25, 2014 at 5:25 PM, Gary Gregory wrote: > Over in Commons VFS we are not adding a pom dep to a jar because it's > license is not ASF compatible, which is ASF policy from what I understand > and how it has been explained on the ML. Ii sounds like the same problem > here. > > Gary > > >

Re: JMH for log4j microbenchmarks

Strictly speaking, you only need to add your key if you are going to be acting as the RM for a release. Adding the key to the KEYS file is something you can do yourself, it's just an edit and commit of the file. Gary On Sun, May 25, 2014 at 7:39 PM, Matt Sicker wrote: > Speaking of code signin

Re: Early Access builds for JDK 9 b13, JDK 8u20 b15 and JDK 7u60 b15 are available on java.net

Hi Matt, Thanks for the update, are you testing all JDKs ? Rgds, Rory On 05/23/14 10:33 PM, Matt Sicker wrote: Our project (log4j) built fine and tests still passed. On 23 May 2014 03:05, Rory O'Donnell Oracle, Dublin Ireland mailto:rory.odonn...@oracle.com>> wrote: Hi Nick, Earl

[jira] [Commented] (LOG4J2-653) Add plugin builder classes as alternative to factory methods.

[ https://issues.apache.org/jira/browse/LOG4J2-653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008678#comment-14008678 ] Ralph Goers commented on LOG4J2-653: I am more interested in what the impact is to the

[jira] [Commented] (LOG4J2-653) Add plugin builder classes as alternative to factory methods.

[ https://issues.apache.org/jira/browse/LOG4J2-653?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14008665#comment-14008665 ] Remko Popma commented on LOG4J2-653: I get how injection mechanisms work, but I'm not