tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Propes, Barry L " <barry.l.pro...@citi.com>
Subject RE: TC 4.1.36 stopped logging
Date Tue, 01 Apr 2008 16:32:02 GMT
Chris,

I've switched the params on the XML file to look more closely at this and inspect leaks.

I might even should tighten it up more, to about 15 seconds. The deal here is that we have
some 30-40 businesses sharing this Oracle server that allows hundreds and hundreds of connections..
(dare I ask why they do it this way?).

While with my login profile I'm not able to see the clock on the server (again, same question
as above) I did verify with the "server admin" that it did indeed show and was set to the
accurate, current time.

With that said about the JAR file, I'll keep that in mind and rest easy. It was a little puzzling
though how errors seem, no not seem, they are in fact late arriving to the log on the prod
server and immediately get logged to the log files on my box (the dev env).

That's partly why I thought the jar responsible for logging might have gotten hosed.

Thanks, Chris!



-----Original Message-----
From: Christopher Schultz [mailto:chris@christopherschultz.net]
Sent: Monday, March 31, 2008 5:28 PM
To: Tomcat Users List
Subject: Re: TC 4.1.36 stopped logging


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Barry,

Propes, Barry L wrote:
| all of a sudden, I've got a connection pool leak. I'll troubleshoot
| that separately from this issue, though quite honestly I'm not sure
| how it suddenly occurred, but I believe I can fix that....although
| what is one of the ideal monitoring tools?

Set:
~        removeAbandoned="true"
~        removeAbandonedTimeout="30"
~        logAbandoned="true"

on your <Resource> and you'll get stack traces in your log files when
connections are held for more than (in this case) 30 seconds. All of my
queries are nearly instantaneous, so a 30-second connection checkout is
considered a leak.

| Aside from that, there appears to be some latency in my logging...in
| other words, the errors getting logged seem to be substantially
| buffered then written to the log (not in real time) maybe hours later
| after an exception. I know this because I've purposely gone in and
| thrown or forced one so that I could see it in the logs.

Wow, that's odd. Which logging mechanism are you using? Is the clock on
your server correct?

| Is it possible my loggings jar file has gotten corrupted?

In my experience, the JAR either works or it doesn't. We had someone
check a JAR file into CVS without setting the "kb" flag (sets it to
binary) and everyone who was using win32 started getting errors all over
the place.

- -chris
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkfxZVkACgkQ9CaO5/Lv0PA9AgCfdujqHU5qa37/bxYMoj7InjU0
v28An0hhtHdv+br84r8vr6tgQEtUNE55
=1VrG
-----END PGP SIGNATURE-----

---------------------------------------------------------------------
To start a new topic, e-mail: users@tomcat.apache.org
To unsubscribe, e-mail: users-unsubscribe@tomcat.apache.org
For additional commands, e-mail: users-help@tomcat.apache.org

Mime
View raw message