mesos-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "brian wickman (Created) (JIRA)" <j...@apache.org>
Subject [jira] [Created] (MESOS-145) mesos executor holds on to fd spawned by slave after slave death, preventing slave from restarting
Date Fri, 03 Feb 2012 04:29:53 GMT
mesos executor holds on to fd spawned by slave after slave death, preventing slave from restarting
--------------------------------------------------------------------------------------------------

                 Key: MESOS-145
                 URL: https://issues.apache.org/jira/browse/MESOS-145
             Project: Mesos
          Issue Type: Bug
            Reporter: brian wickman
            Assignee: Benjamin Hindman


The slaves spin on this error trying to start up as long as my (admittedly badly behaving)
executor holds onto the handle the MesosDriver uses for communication with the slave:

F0203 04:26:18.204121   580 process.cpp:1177] Failed to initialize, bind: Address already
in use [98]
*** Check failure stack trace: ***
    @     0x7f258750db4d  google::LogMessage::Fail()
    @     0x7f25875137b7  google::LogMessage::SendToLog()
    @     0x7f258750f3fc  google::LogMessage::Flush()
    @     0x7f258750f629  google::LogMessage::~LogMessage()
    @     0x7f258750f72d  google::ErrnoLogMessage::~ErrnoLogMessage()
    @     0x7f258752517d  process::initialize()
    @           0x4067fa  main
    @     0x7f2585dae994  __libc_start_main
    @           0x405c99  (unknown)

Ben identified the root cause as needing to fcntl(F_SETFD) on the fd prior to fork.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message