You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafficserver.apache.org by "Leif Hedstrom (JIRA)" <ji...@apache.org> on 2013/02/16 18:35:12 UTC

[jira] [Updated] (TS-1483) Traffic Manager when autoconf fd > 32 will cause endless spins causing traffic_cop to kill server -- Backport 3.2.x

     [ https://issues.apache.org/jira/browse/TS-1483?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Leif Hedstrom updated TS-1483:
------------------------------

    Backport to Version: 3.2.5, 3.2.3  (was: 3.2.3)
    
> Traffic Manager when autoconf fd > 32 will cause endless spins causing traffic_cop to kill server -- Backport 3.2.x
> -------------------------------------------------------------------------------------------------------------------
>
>                 Key: TS-1483
>                 URL: https://issues.apache.org/jira/browse/TS-1483
>             Project: Traffic Server
>          Issue Type: Bug
>          Components: Management
>    Affects Versions: 3.2.0
>            Reporter: Brian Geffon
>            Assignee: Brian Geffon
>             Fix For: 3.2.5
>
>         Attachments: WebIntrMain.patch
>
>
> WebIntrMain.cc uses a hard coded FD_SET Size of 32, so when the autoconf file descriptor is greater than 32 it will cause select to sleep forever since no timeout is specified. We had a few systems that would have continuous traffic_server restarts because they would (have a fd > 32) and would never accept the connection request for /synthetic.txt causing cop to kill mananger and server. But when Manager restarted is still got a high fd and it would happen again, while the whole time traffic_server is running just fine. This should be switched to FD_SETSIZE if available.
> Discovered by Thomas Jackson

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira