0001375: mod_caucho starts returning 404s on all resin handled requests

Viewing Issue Advanced Details [ Jump to Notes ]

[ View Simple ] [ Issue History ] [ Print ]

Category

Severity

Reproducibility

Date Submitted

Last Update

0001375

[Resin]

major

random

09-25-06 15:04

05-15-08 09:01

Reporter

justinwcs

View Status

public

Assigned To

ferg

Priority

normal

Resolution

fixed

Platform

Status

closed

Projection

none

OS Version

ETA

none

Fixed in Version

3.0.23

Product Version

3.0.21

Product Build

Summary

0001375: mod_caucho starts returning 404s on all resin handled requests

Description

We have several load balanced web servers running apache and mod_caucho on resin 3.0.21. When we send the servers kill -HUP during log rotation ocassionaly a server that has been up for a long period of time (>30 days) will start sending out 404 errors on all requests that resin is handling. Values mapped via plugin_ignore are handled properly by the Apache server. Sometimes if the apache server is restarted the problem goes away other times it does not go away no matter if I kill -9 the process and restart. The problem always goes away if the server is completely rebooted. The server is configured via the ResinConfigServer directive.

Steps To Reproduce

Additional Information

Server: Apache/1.3.37 (Unix) Resin/3.0.21 mod_ssl/2.8.28 OpenSSL/0.9.8b
2.6.8-2-386 0000001 Tue Aug 16 12:46:35 UTC 2005 i686 GNU/Linux

A failed request looks like this (I changed some values to placeholders 'x' ):
xxx.xx.x.x www.hogwarts.com - [xx/Sep/2006:xx:xx:xx -0700] "GET /index.jsp HTTP/1.1" 404 - "" "Mozilla/4.0 (compatible; MSIE 6.0; AOL 9.0; Windows NT 5.1; SV1; .NET CLR 1.1.4322)" "-" 0

And a File does not exist shows up in the apache error logs. The other webservers in the cluster run fine. One thing that's kind of odd is that there is no data sent back when the 404 hits.

Attached Files

Relationships

Notes
(0001528) ferg 09-25-06 19:41	As a workaround, you might look at the /tmp/*_6802 files and delete them when doing an Apache restart. That wouldn't fix the initial problem, but would let you restart Apache without rebooting the machine.

Issue History
Date Modified	Username	Field	Change
09-25-06 15:04	justinwcs	New Issue
09-25-06 16:15	justinwcs	Issue Monitored: justinwcs
09-25-06 19:41	ferg	Note Added: 0001528
11-30-06 15:59	ferg	Assigned To	=> ferg
11-30-06 15:59	ferg	Status	new => closed
11-30-06 15:59	ferg	Resolution	open => fixed
11-30-06 15:59	ferg	Fixed in Version	=> 3.0.23
11-30-06 15:59	ferg	Additional Information Updated
05-15-08 08:58	sam	Additional Information Updated
05-15-08 09:01	sam	View Status	private => public

Mantis 1.0.0rc3[^]

34 total queries executed.
28 unique queries executed.