0005258: periodic HTTP 503 error occured in Apache+Resin

Viewing Issue Simple Details [ Jump to Notes ]

[ View Advanced ] [ Issue History ] [ Print ]

Category

Severity

Reproducibility

Date Submitted

Last Update

0005258

[Resin]

block

always

10-23-12 19:56

10-25-12 10:44

Reporter

lvchenggang

View Status

public

Assigned To

ferg

Priority

normal

Resolution

fixed

Status

closed

Product Version

3.1.10

Summary

0005258: periodic HTTP 503 error occured in Apache+Resin

Description

We have a web app using Apache(2.2.16)+Resin(3.1.10)+Nginx. almost every 2 days apache can not connect to resin and get a HTTP 503 response. we look up the resin log and find the following error message:

[08:31:07.138] {hmux-192.168.1.224:7800-636$1512575371} JniSelectManager[] file descriptor overflow fd=16735 max=16384 TcpConnection[id=hmux-192.168.1.224:7800-636,socket=JniSocketImpl$385288876[46915213517776,fd=16735],port=ClusterPort[address=192.168.1.224,port=7800]]

[08:31:07.139] {hmux-192.168.1.224:7800-636$1512575371} Tcp[apollo,636] failed keepalive (select)

can anyone help us ?
thank!

Additional Information

Attached Files

Relationships

Notes
(0006071) ferg 10-25-12 10:44	This is fixed in the 4.0.x branch. For 3.1, you can disable the keepalive-select manager with keepalive-select-enable=false: <resin...> <cluster id=""> <keepalive-select-enable>false</keepalive-select-enable> .. The <keepalive-select-enable> can also be put in the <server> or <server-default> blocks.

Issue History
Date Modified	Username	Field	Change
10-23-12 19:56	lvchenggang	New Issue
10-25-12 10:44	ferg	Note Added: 0006071
10-25-12 10:44	ferg	Assigned To	=> ferg
10-25-12 10:44	ferg	Status	new => closed
10-25-12 10:44	ferg	Resolution	open => fixed
10-25-12 10:44	ferg	Fixed in Version	=> 4.0.32

Mantis 1.0.0rc3[^]

29 total queries executed.
26 unique queries executed.