Re: 2.6 Release

Yes, I built the site but can’t upload it right now. I will let you know when it is there. Ralph > On May 25, 2016, at 9:47 AM, Matt Sicker wrote: > > Are you going to upload it to home.apache.org ? > > On 25 May 2016 at 10:51, Apache

Re: 2.6 Release

Are you going to upload it to home.apache.org? On 25 May 2016 at 10:51, Apache wrote: > The release build is finished. I have to build the site and make it all > available for review. I may not finish that until tonight. > > Ralph > > On May 25, 2016, at 6:57 AM, Remko Popma wrote: > > Ok. Done

Re: 2.6 Release

The release build is finished. I have to build the site and make it all available for review. I may not finish that until tonight. Ralph > On May 25, 2016, at 6:57 AM, Remko Popma wrote: > > Ok. Done tweaking the announcement. Thanks for waiting! > > On Wed, May 25, 2016 at 7:49 PM, Ralph Goe

Re: 2.6 Release

Ok. Done tweaking the announcement. Thanks for waiting! On Wed, May 25, 2016 at 7:49 PM, Ralph Goers wrote: > Yes. You need to update announcement.vm > > Sent from my iPad > > On May 24, 2016, at 11:54 PM, Remko Popma wrote: > > Can I tweak the announcement text a little? > > I'd like it to say

Re: 2.6 Release

Yes. You need to update announcement.vm Sent from my iPad > On May 24, 2016, at 11:54 PM, Remko Popma wrote: > > Can I tweak the announcement text a little? > > I'd like it to say > > ... the ability to be "garbage-free" (avoid allocating temporary objects) > while logging if certain conditi

Re: 2.6 Release

On Windows, I get a test failure in core: UdpSerializedSocketServerTest>AbstractSocketServerTest.test10ShortMessages:118->AbstractSocketServerTest.testServer:161->AbstractSocketServerTest.testServer:201 Incorrect number of events received expected:<10> but was:<7> Gary On Tue, May 24, 2016 at 1

Re: 2.6 Release

Can I tweak the announcement text a little? I'd like it to say ... the ability to be "garbage-free" (avoid allocating temporary objects) while logging if certain conditions are met, (...) If no objections I'll commit this when I get home in a few hours. Sent from my iPhone > On 2016/05/25,

Re: 2.6 Release

I started the release process but the build failed. I’m going to try from the office tomorrow. Ralph > On May 24, 2016, at 2:38 PM, Gary Gregory wrote: > > FYI: > > Just a note that we have some 28 @Ignored tests. Some are documented, some > not. > > For example: > > org.apache.logging.lo

Re: 2.6 Release

FYI: Just a note that we have some 28 @Ignored tests. Some are documented, some not. For example: org.apache.logging.log4j.core.appender.rolling.RollingRandomAccessFileManagerHeaderFooterTest.testAppender(): @Ignore // FIXME org.apache.logging.log4j.core.net.server.AbstractSocketServerTest.test1

Re: 2.6 Release

OK. Hopefully I can do it tonight. Ralph > On May 24, 2016, at 9:29 AM, Remko Popma wrote: > > Thanks, Mikael! > > And I fixed LOG4J2-623 and > LOG4J2-1362 , which were > the last remaining

Re: 2.6 Release

Thanks, Mikael! And I fixed LOG4J2-623 and LOG4J2-1362 , which were the last remaining showstoppers. I think we are ready for the 2.6 release! On Tue, May 24, 2016 at 10:59 PM, Mikael Ståldal

Re: 2.6 Release

Fixed. On Tue, May 24, 2016 at 3:41 PM, Remko Popma wrote: > Good catch! Can you fix the manual? > > Sent from my iPhone > > On 2016/05/24, at 17:03, Mikael Ståldal wrote: > > The manual says that the default target for ConsoleAppender is SYSTEM_ERR, > but it seems to be SYSTEM_OUT in the code.

Re: 2.6 Release

Good catch! Can you fix the manual? Sent from my iPhone > On 2016/05/24, at 17:03, Mikael Ståldal wrote: > > The manual says that the default target for ConsoleAppender is SYSTEM_ERR, > but it seems to be SYSTEM_OUT in the code. > >> On Mon, May 23, 2016 at 6:52 PM, Gary Gregory wrote: >> An

Re: 2.6 Release

The manual says that the default target for ConsoleAppender is SYSTEM_ERR, but it seems to be SYSTEM_OUT in the code. On Mon, May 23, 2016 at 6:52 PM, Gary Gregory wrote: > Another alternative is too add, later, an option to choose the style of > the layout. Or, provide a different Layout class.

Re: 2.6 Release

Another alternative is too add, later, an option to choose the style of the layout. Or, provide a different Layout class. My only concern is to be able to validate XML events with a 'simple' schema. JSONSchema creeps in here as well. Gary On Mon, May 23, 2016 at 7:49 AM, Ralph Goers wrote: > If

Re: 2.6 Release

If I was to make the modification I would add an option to the layout to allow it to behave as it currently does. Ralph > On May 23, 2016, at 7:30 AM, Matt Sicker wrote: > > I'd support commenting out the YAML layout info in the docs. If users try out > an undocumented feature, they can expe

Re: 2.6 Release

I'd support commenting out the YAML layout info in the docs. If users try out an undocumented feature, they can expect breakage in the future. On 23 May 2016 at 09:21, Remko Popma wrote: > No. In general we can change what's in core, although we try not to break > things if possible. > > That sa

Re: 2.6 Release

No. In general we can change what's in core, although we try not to break things if possible. That said, changing an output format, especially one that's likely to be machine-processed, is not a good thing. A quick alternative is to document on the site that we will change the format for Th

Re: 2.6 Release

That really depends on our guarantee of backwards compatibility. Does it extend to layouts like this? On 23 May 2016 at 07:42, Remko Popma wrote: > Is the JSON/YAML layout fix a showstopper? > I haven't seen anyone volunteer to fix it and would not want this to hold > up the release indefinitely

Re: 2.6 Release

Is the JSON/YAML layout fix a showstopper? I haven't seen anyone volunteer to fix it and would not want this to hold up the release indefinitely... On Mon, May 23, 2016 at 3:33 PM, Ralph Goers wrote: > My computer had to be sent out for repairs but I was able to retrieve my > code signing key an

Re: 2.6 Release

Yes, let's wait with LOG4J-1390 until after the 2.6 release. On Sat, May 21, 2016 at 8:59 AM, Remko Popma wrote: > I took a look at LOG4J2-1390 > and the associated > branch. > The fix looks correct if the newest Kafka client correctly implemen

Re: 2.6 Release

My computer had to be sent out for repairs but I was able to retrieve my code signing key and install it on my other MacBook Pro. So I should be able to start the release whenever we are ready. From what I can tell we still need a fix for the YAML and Json layouts. Ralph > On May 22, 2016, at

Re: 2.6 Release

Time machine. Sent from my iPhone > On May 21, 2016, at 10:41 PM, Remko Popma wrote: > > Good stuff! > This reminded me I need to take a look at my backup strategy. :-) > >> On Sun, May 22, 2016 at 2:12 PM, Ralph Goers >> wrote: >> I fixed the license header issues. Found a bug in how proper

Re: 2.6 Release

Good stuff! This reminded me I need to take a look at my backup strategy. :-) On Sun, May 22, 2016 at 2:12 PM, Ralph Goers wrote: > I fixed the license header issues. Found a bug in how properties were > being merged into composite configurations - that was causing the problem > with ${filename}

Re: 2.6 Release

I fixed the license header issues. Found a bug in how properties were being merged into composite configurations - that was causing the problem with ${filename}. Now all I need is to get my personal MacBook Pro to be able to boot. Ralph > On May 21, 2016, at 5:57 PM, Remko Popma wrote: > >

Re: 2.6 Release

About the ThreadContext map in JSON/YAML format: towards the end I see a comment from Ralph that the current output is wrong. If you have the bandwidth I'd say go ahead and fix it. On Friday, 20 May 2016, Mikael Ståldal wrote: > What about the format for thread context map in JsonLayout and Yaml

Re: 2.6 Release

On Sat, May 21, 2016 at 7:17 AM, Remko Popma wrote: > Got it. No rush. Enjoy! > +1. Travel safe. Gary > Sent from my iPhone > > On 2016/05/21, at 23:02, Ralph Goers wrote: > > No. There is still one unit test generating the ${filename} file. I have > fixed the other one but have not committe

Re: 2.6 Release

Got it. No rush. Enjoy! Sent from my iPhone > On 2016/05/21, at 23:02, Ralph Goers wrote: > > No. There is still one unit test generating the ${filename} file. I have > fixed the other one but have not committed it yet. I will do that when I get > them both fixed. I am visiting with family th

Re: 2.6 Release

No. There is still one unit test generating the ${filename} file. I have fixed the other one but have not committed it yet. I will do that when I get them both fixed. I am visiting with family this morning and flying this afternoon so I don’t know how much I will get done today. Ralph > On May

Re: 2.6 Release

Are all the license header issues resolved? On Saturday, 21 May 2016, Ralph Goers wrote: > Sorry, I am still traveling. To make matters worse, my Mac that has my > code signing key won’t boot. I will be taking it to the Apple store on > Monday. I don’t see anything wrong with the hardware. I thi

Re: 2.6 Release

Sorry, I am still traveling. To make matters worse, my Mac that has my code signing key won’t boot. I will be taking it to the Apple store on Monday. I don’t see anything wrong with the hardware. I think something went wrong with the last OS/X update. So it is still going to be a few days. Ral

Re: 2.6 Release

So I'm thinking it may be better to leave in the Kafka workaround and release current master as it is. (Can't wait to get 2.6 out there, by the way!) :-) On Saturday, 21 May 2016, Remko Popma wrote: > I took a look at LOG4J2-1390 > and the ass

Re: 2.6 Release

I took a look at LOG4J2-1390 and the associated branch. The fix looks correct if the newest Kafka client correctly implemented KAFKA-1660 , but I would suggest it is better to leave the workaround

Re: 2.6 Release

Mikael, I finally had a chance to look at the update for the Layout page. Looks good. I like the new table with parameters for the JSON layout. On Fri, May 20, 2016 at 5:26 PM, Mikael Ståldal wrote: > I have updated the layout manual page a bit. Please have a look at it. > > On Fri, May 20, 2016

Re: 2.6 Release

I just found an issue with using the newest Kafka client library though. It seems to be slightly incompatible with older Kafka servers. I need to investigate. On Fri, May 20, 2016 at 6:18 PM, Matt Sicker wrote: > That would be cool with me as we don't have a release candidate in process > right

Re: 2.6 Release

That would be cool with me as we don't have a release candidate in process right now thanks to shitty hotel internet. ;) On 20 May 2016 at 11:01, Mikael Ståldal wrote: > I have a fix for this issue: > https://issues.apache.org/jira/browse/LOG4J2-1390 > > Do we want to include it in the release?

Re: 2.6 Release

I have a fix for this issue: https://issues.apache.org/jira/browse/LOG4J2-1390 Do we want to include it in the release? On Fri, May 20, 2016 at 10:26 AM, Mikael Ståldal wrote: > I have updated the layout manual page a bit. Please have a look at it. > > On Fri, May 20, 2016 at 10:15 AM, Mikael S

Re: 2.6 Release

I have updated the layout manual page a bit. Please have a look at it. On Fri, May 20, 2016 at 10:15 AM, Mikael Ståldal wrote: > What about the format for thread context map in JsonLayout and YamlLayout? > > https://issues.apache.org/jira/browse/LOG4J2-1362 > > https://issues.apache.org/jira/bro

Re: 2.6 Release

What about the format for thread context map in JsonLayout and YamlLayout? https://issues.apache.org/jira/browse/LOG4J2-1362 https://issues.apache.org/jira/browse/LOG4J2-623 Currently it is: { "timeMillis" : 1463730772138, "thread" : "main", "level" : "INFO", "loggerName" : "HelloWorld"

Re: 2.6 Release

In fixing the problem with ${filename} I found a bug. I am still debugging the second test that is creating that file. Given the slow internet at my hotel I may have to wait til next week. Ralph > On May 19, 2016, at 10:37 AM, Remko Popma wrote: > > I've been making minor edits to the site a

Re: 2.6 Release

I've been making minor edits to the site and docs, but please feel free to start the release whenever you are ready. On Thu, May 19, 2016 at 12:06 AM, Remko Popma wrote: > I fixed 3 license issues (GilsDosAndDontsLatencyTesting.txt, log4j-perf/ > src/main/resources/perf-WithoutAnyAppender-locati

Re: 2.6 Release

I fixed 3 license issues (GilsDosAndDontsLatencyTesting.txt, log4j-perf/ src/main/resources/perf-WithoutAnyAppender-location.xml, and src/main/resources/perf-WithoutAnyAppender.xml). On Wed, May 18, 2016 at 8:30 PM, Ralph Goers wrote: > Thanks for finding it. I will fix it. > > Ralph > > On May

Re: 2.6 Release

Thanks for finding it. I will fix it. Ralph > On May 18, 2016, at 6:02 AM, Remko Popma wrote: > > ${filename} is created by CompositeConfigurationTest, ::compositeLogger and > ::testReconfiguration > > I don't know why though. Can someone else fix that? > > I will fix the license issues for

Re: 2.6 Release

${filename} is created by CompositeConfigurationTest, ::compositeLogger and ::testReconfiguration I don't know why though. Can someone else fix that? I will fix the license issues for GilsDosAndDontsLatencyTesting.txt, > src/main/resources/perf-WithoutAnyAppender-location.xml, and > src/main/re

Re: 2.6 Release

There are 2 files with unapproved licenses in log4j-core: ${filename} and src/test/java/org/apache/logging/log4j/core/async/perftest/GilsDosAndDontsLatencyTesting.txt. Obviously the first is a build error but I am not sure if the second is meant to be there. If it is it needs an apache license

Re: 2.6 Release

No need to torture yourself with slow internet. We can wait a few days no? That would give us all time or at least myself to review the current code. But I can't look until tonight at the earliest... Gary On Tue, May 17, 2016 at 8:15 AM, Ralph Goers wrote: > The performance page has been comple

Re: 2.6 Release

I applied the patch for LOG4J2-1348. Can't see anything else remaining. I'm good to go with this. On Wed, May 18, 2016 at 12:19 AM, Matt Sicker wrote: > I hope you have unlimited data! ;) > > I'm ready for release (and ready to start working on a presentation about > the new stuff for my JUG). >

Re: 2.6 Release

I hope you have unlimited data! ;) I'm ready for release (and ready to start working on a presentation about the new stuff for my JUG). On 17 May 2016 at 10:15, Ralph Goers wrote: > The performance page has been completed. Is there anything left to do > before the 2.6 release build? If not I c