Re: log4j Configuration

2019-09-26 Thread Jerry Malcolm
Thank you so much.  This new log4j.properties file did solve the problem that the logs were not getting written.   When I finally saw the mailet log, I realized that this wasn't the real problem. Even though there was some mailet logging data now written to the log file, my own mailet log(..) o

Re: log4j Configuration

2019-09-26 Thread Tellier Benoit
Check https://github.com/linagora/james-project/pull/2718 This should solve your issue. On 20/09/2019 09:18, Jerry Malcolm wrote: > Any chance I could get someone to take a look at this problem? There's > no way for me to move forward until I can see my mailet logs.  I tried > to dig into slf4j a

Re: log4j Configuration

2019-09-19 Thread Jerry Malcolm
Any chance I could get someone to take a look at this problem? There's no way for me to move forward until I can see my mailet logs.  I tried to dig into slf4j and log4j.  But I just don't have enough background in that to figure out what is happening.  And since I don't have a full james set u

Re: log4j Configuration

2019-09-18 Thread Jerry Malcolm
Without a way to see my mailet log data, I have no way of knowing if my mailets were even called, if they succeeded, if they failed.  Right now, all I know is they went into a processor and disappeared.  This is a serious problem for me.  I really need to have a way to see this data.  If it's a

Re: log4j Configuration

2019-09-17 Thread Jerry Malcolm
I added the transport line as you suggested and sent a test email inbound.  The console showed it coming in, as does james-server.log.  But still, the only log file in the logs directory that is not empty is james-server.  Nothing is going to any of these files. On 9/17/2019 10:49 PM, Tellier

Re: log4j Configuration

2019-09-17 Thread Jerry Malcolm
I'll try that shortly.  But that doesn't explain why all of the logger output files are empty, no matter what the log4j settings are.  This was after 30 minutes of running with every log level throughout log4j.properties set to DEBUG.  Shouldn't at least some of these have some sort of log data

Re: log4j Configuration

2019-09-17 Thread Tellier Benoit
I inlined my answers... Regards, On 18/09/2019 10:32, Jerry Malcolm wrote: > Thanks for the reply.  But I did not use my old log4j.properties.  I > copied the one from the downloaded 3.3.0 zip file.  Although, it looks > almost identical to the beta5 one.  Any chance the wrong one is included > i

Re: log4j Configuration

2019-09-17 Thread Jerry Malcolm
Thanks for the reply.  But I did not use my old log4j.properties.  I copied the one from the downloaded 3.3.0 zip file.  Although, it looks almost identical to the beta5 one.  Any chance the wrong one is included in the package? I'm fine with logger-per-mailet.  However, I'm hoping there's a w

Re: log4j Configuration

2019-09-17 Thread Tellier Benoit
Maybe I could help on that one: The mailet logging had been migrated to a standard SLF4J approach. Before that, all mailet logs went though a single logger, exposed as part of the MailetContext object. Before 3.0.0 release we altered that behavior, and create one logger per mailet class for bund

Re: log4j Configuration

2019-09-17 Thread Jerry Malcolm
Not good news for me, however.  My mailet flow that worked in v3b5 is crashing now, and none of the log output I'm writing from my mailets is showing up anywhere.  I've GOT to see that log info some place in order to debug both my mailets as well as my mailet flow that is somehow no longer comp

Re: log4j Configuration

2019-09-17 Thread Garry Hurley
It’s not you. I have the same issue. I think those other log files are just there for backwards compatibility to make users feel good, since they never seem to be written to. “No news is good news” as the saying goes. Sent from my iPhone > On Sep 17, 2019, at 4:34 PM, Jerry Malcolm wrote: >