Using Log4j, its possible to write a database or nosql appender [1] if we want 
to provide a remote access to the logs. I have used database appender before 
and it is asynchronous logging without any overhead. 

Thanks
Raminder 

1. http://logging.apache.org/log4j/2.x/manual/appenders.html

On Aug 28, 2014, at 11:13 AM, Schwartz, Terri <[email protected]> wrote:

> For cipres logging we use a single file, rolled over daily, and include 
> thread-id, and when relevant, task-id in each entry using log4j's nested 
> diagnostic context.   It's usually easy enough to grep for just the task I'm 
> interested in.
> 
> Terri
> ________________________________________
> From: Miller, Mark [[email protected]]
> Sent: Thursday, August 28, 2014 8:02 AM
> To: [email protected]
> Subject: RE: Experiment log messages to different log files
> 
> Re max files, here is what I found:
> 
> It depends on the file system. ext3 suppport ~32000 subdirectories (not 
> files!) in a given directory, with ext4 it's 64000 by default. xfs has no 
> limit to my knowledge.  You should consider not putting too many files in a 
> single directory. Most software doesn't handle that well (e.g. mc will be 
> slow, many gui tools will be unusable). It's better to create a hierarchy of 
> nested folders and distribute the files in them using some algorithm (hash of 
> file name or content or any other method which will distribute the files 
> equally). That's what many mature programs (e.g. squid) do.
> 
> 
> -----Original Message-----
> From: Marlon Pierce [mailto:[email protected]]
> Sent: Thursday, August 28, 2014 7:54 AM
> To: Airavata Dev
> Subject: Experiment log messages to different log files
> 
> I'm wondering if it would be a good idea to create a separate log file for 
> each experiment.  We current direct everything to airavata.log and 
> airavata-server.out. As a consequence, log entries for different submissions 
> get interleaved, which makes tracing a particular experiment's life cycle 
> difficult.
> 
> Note there is a 32,000 file limit per directory in Linux by default (unless 
> my knowledge is obsolete), so we'd need to take this into account.
> 
> Alternatively, we could do a better job of labeling the entries in 
> airavata.log so that it was clear which experiment is associated with the 
> entry.
> 
> Other suggestions?
> 
> Marlon
> 

Reply via email to