Skip to main content
Participant
October 7, 2009
Question

FMS 3.0 stops accepting connections after 1-3 days, restart required

  • October 7, 2009
  • 1 reply
  • 1850 views

Hi,

We have a custom application that is running on FMS. Server is dedicated just for fms, there is only one application running on the server, with more instances, average traffic is 3-15 simultaneous users. The problem is, server suddenly stops accepting connections (after 1-3 days from restart), from this point it is impossible to connect to any instance, and fms server must be restarted. When the server crashes, the admin console can see all open instances, can close instances, can restart fms. There is nothing logged, that would explain this sudden failure.

Second problem is, cores are connected and disconnected too often, causing client to reconnect (takes 2-5 seconds).

Do you have any idea what is causing these problems?

Hardware: 4 x quadcore Intel(R) Xeon(R) CPU E7440 @ 2.40GHz, 32 GB ram

OS: CentOS release 5.3 (Final), kernel: 2.6.18-128.4.1.el5

cpu and mem usage: 1-5% max

FMS:

version: Adobe Flash Media Server 3.0.1 r123

license: valid, unlimited

configuration: default, changed in Application.xml:

<AutoCloseIdleClients enable="true">

- just communication throught objects, strings, int, shared objects

- no streaming used

- Mbits in/out: 50/350 (48 hours from restart)

- Msgs in/out/drop: 4.000.000/21.000.000/0 (48 hours from restart)

- no problems with connection, no lags

LOGS:

admin.log

#Fields: date time x-pid x-status x-ctx x-comment

2009-10-07 00:03:30 4816 (i)2581242 Core (8684) disconnected : _defaultRoot_:_defaultVHost_:::_1. -

2009-10-07 00:03:31 4816 (i)2581241 Connection to admin received. -

2009-10-07 00:03:31 4816 (i)2581243 Connection from core 10772 received. -

2009-10-07 00:03:31 4816 (i)2581244 Connection from core 10772 accepted. -

... repeating (10 minutes - 1 hour period)

master.log

#Fields: date time x-pid x-status x-ctx x-comment

2009-10-07 00:03:31 24962 (i)2581223 Core (8684) is no longer active. -

2009-10-07 00:03:31 24962 (w)2581256 Core (8684) _defaultRoot_:_defaultVHost_:::_1 experienced 1 failure! -

2009-10-07 00:03:31 24962 (i)2581221 Core (10772) started, arguments : -adaptor "_defaultRoot_" -vhost "_defaultVHost_" -app  -inst  -tag "_1" -conf "./conf/Server.xml" -name "_defaultRoot_:_defaultVHost_:::_1". -

... repeating (10 minutes - 1 hour period)

edge.log

#Fields: date time x-pid x-status x-ctx x-comment

2009-10-07 00:03:08 24982 (w)2641213 Connection rejected by server. Reason : [ Server.Reject ] : (_defaultRoot_, _defaultVHost_) : Application (doodletoostats) is not defined. -

2009-10-07 00:03:30 24982 (i)2581250 Edge disconnected from core (8684). -

2009-10-07 00:03:31 24982 (i)2581252 Registering core (10772). -

... repeating (1 - 3 hours period)

core.log

#Fields: date time x-pid x-status x-ctx x-comment

2009-10-07 00:03:31 10772 (i)2581237 Starting admin app on core (10772). -

2009-10-07 00:03:31 10772 (i)2581238 Core (10772) connecting to admin. -

2009-10-07 00:03:31 10772 (i)2581231 Core (10772) connected to admin. -

2009-10-07 00:03:31 10772 (i)2581234 Core (10772) connection to admin accepted. -

2009-10-07 00:03:31 10772 (i)2581246 Core (10772) sending register cmd to edge. -

...

2009-10-07 00:54:22 11329 (i)2581247 Core (11329) disconnected from edge. -

2009-10-07 00:54:22 11329 (i)2581233 Core (11329) disconnecting from admin. -

... repeating

    This topic has been closed for replies.

    1 reply

    astro44
    Inspiring
    March 10, 2010

    We have seen similar issues and the best work around is to have your clients connect on alternate cores/servers. The way to accomplish this really depends on your type of application.