0000965: com.caucho.vfs.ClientDisconnectException: connection reset by peer error

Viewing Issue Advanced Details [ Jump to Notes ]

[ View Simple ] [ Issue History ] [ Print ]

Category

Severity

Reproducibility

Date Submitted

Last Update

0000965

[Resin]

major

sometimes

02-28-06 08:07

04-13-06 09:28

Reporter

annmarie

View Status

public

Assigned To

ferg

Priority

normal

Resolution

fixed

Platform

Status

closed

Projection

none

OS Version

ETA

none

Fixed in Version

3.0.19

Product Version

Product Build

Summary

0000965: com.caucho.vfs.ClientDisconnectException: connection reset by peer error

Description

We are running Resin 3.0.7 on Microsoft Windows Server 2003 with 2GB of RAM. We've been running the same applications for years without error. Recently after about 2 hours of high volume we've been receiving tons of these errors in the debug.log

[19:26:47.162][22] com.caucho.vfs.ClientDisconnectException: connection reset by peer

When this happens, http requests can not get through to the server except sporadically. On the client end, the callling application is recording the following responses from the Resin Server:

Call-complete failed with error code:404,Not Found at 16:46:51
and
Call-complete failed with error code:500,EOF instead of reponse status line at 20:30:23

These requests are to the same urls that always work.

CPU usage and memory seems fine, we have verbose logging on.

We set MaxUserPort to 65534 in the windows registry but it did not fix the problem. <ignore-client-disconnect> in resin.conf is set to true.

This problem is causing huge repercussions on our systems.

Please advise how to troubleshoot/fix?

Thank you,
Ann Marie

Steps To Reproduce

Additional Information

Attached Files

Relationships

Notes
(0000909) annmarie 02-28-06 08:53	Please scratch this - Call-complete failed with error code:404,Not Found at 16:46:51. This error was only received when we were rebooting the server. Call-complete failed with error code:500,EOF instead of reponse status line at 20:30:23 is what is returned. Also, certain ports remained in a "CLOSE_WAIT" state even after 10 min. The handle count went up to 4000, yet after a reboot was at 1900.

(0001006) mate 04-03-06 23:49	Seems to be a duplicate of 914, see http://bugs.caucho.com/view.php?id=914 [^]

Issue History
Date Modified	Username	Field	Change
02-28-06 08:07	annmarie	New Issue
02-28-06 08:53	annmarie	Note Added: 0000909
02-28-06 09:11	satheesh	Issue Monitored: satheesh
02-28-06 09:13	satheesh	Issue End Monitor: satheesh
04-03-06 23:49	mate	Note Added: 0001006
04-12-06 00:45	mate	Issue Monitored: mate
04-13-06 09:28	ferg	Assigned To	=> ferg
04-13-06 09:28	ferg	Status	new => closed
04-13-06 09:28	ferg	Resolution	open => fixed
04-13-06 09:28	ferg	Fixed in Version	=> 3.0.19

Mantis 1.0.0rc3[^]

36 total queries executed.
29 unique queries executed.