You are viewing a plain text version of this content. The canonical link for it is here.
Posted to bugs@httpd.apache.org by bu...@apache.org on 2010/10/20 19:13:25 UTC
DO NOT REPLY [Bug 50125] apache failed with any exception
https://issues.apache.org/bugzilla/show_bug.cgi?id=50125
Rainer Jung <ra...@kippdata.de> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
Resolution| |INVALID
--- Comment #1 from Rainer Jung <ra...@kippdata.de> 2010-10-20 13:13:21 EDT ---
Bugzilla is not a support forum. Plese post your questions to the Tomcat users
list, or the Apache Web Server users list.
To subscribe, send mail to
users-subscribe@tomcat.apache.org
or
users-subscribe@httpd.apache.org
Nevertheless here are some remarks that might be helpful. Please don't answer
here but take the discussion to one of the lists.
> It works ALMOST ok but when I hit ANY exception, for ex,
> java.lang.NullPointerException, http doesn't reponse anymore. We have to
> restart Tomcat+Apache
That's pretty strange, but ...
> Sorry for log in french:
... your logs speak a different language (assuming clocks are in sync):
Ordered by timestamp:
Apache receives request, responds with 500:
xyz.com:80 ip - - [14/Oct/2010:23:40:20 +0200] "GET /xyz/jsp/test.jsp?p=test
HTTP/1.1" 500 3046 "-" ""
mod_jk logs the request, it tool 0.05 seconds
[Thu Oct 14 23:40:20 2010]worker1 192.168.163.113 0.058192
and the reason for the status 500 is
14 oct. 2010 23:40:20 org.apache.catalina.core.StandardWrapperValve invoke
GRAVE: "Servlet.service()" pour la servlet jsp a généré une exception
java.lang.NullPointerException
13 seconds later someone tells Apache to shut down
[Thu Oct 14 23:40:33 2010] [notice] caught SIGTERM, shutting down
... and someone tells Tomcat to shutdown:
14 oct. 2010 23:40:33 org.apache.coyote.http11.Http11BaseProtocol pause
INFO: Suspension de Coyote HTTP/1.1 sur http-8180
but this shutdown hangs ...
Starting up Apache again
[Thu Oct 14 23:41:34 2010] [notice] Apache/2.2.9 (Debian) mod_jk/1.2.26
configured -- resuming normal operations
mod_jk loggs it initialized after the startup
[Thu Oct 14 23:41:34 2010][2281:2233493328] [info] init_jk::mod_jk.c (2830):
mod_jk/1.2.26 initialized
[Thu Oct 14 23:41:34 2010][2282:2233493328] [info] init_jk::mod_jk.c (2830):
mod_jk/1.2.26 initialized
Now Tomcat proceeds shutting down
14 oct. 2010 23:43:42 org.apache.catalina.connector.Connector pause
GRAVE: La suspension du gestionnaire de protocole a échouée
java.net.ConnectException: Connection timed out
Another request comes in, answered with 503
xyz.com:80 ip - - [14/Oct/2010:23:44:41 +0200] "GET /xyz/jsp/test.jsp?p=test
HTTP/1.1" 503 417 "-" ""
Right, mod_jk says it couldn't connect to Tomcat
[Thu Oct 14 23:44:41 2010][2286:1086212432] [info] jk_open_socket::jk_connect.c
(566): connect to 127.0.0.1:8009 failed (errno=111)
[Thu Oct 14 23:44:41 2010][2286:1086212432] [info]
ajp_connect_to_endpoint::jk_ajp_common.c (869): Failed opening socket to
(127.0.0.1:8009) (errno=111)
[Thu Oct 14 23:44:41 2010][2286:1086212432] [error]
and that's what we expect, since someone shut it down.
22 seconds later Apache gets shut down again:
[Thu Oct 14 23:45:03 2010] [notice] caught SIGTERM, shutting down
[Thu Oct 14 23:45:04 2010] [notice] Apache/2.2.9 (Debian) mod_jk/1.2.26
configured -- resuming normal operations
[Thu Oct 14 23:45:33 2010] [notice] caught SIGTERM, shutting down
If you don't know who shuts down the components all the time, maybe someone
implemented a clever log file monitoring and some automatism outside of Apache
and Tomcat does it?
Regards,
Rainer
--
Configure bugmail: https://issues.apache.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org