#878 ✓resolved
hiro

MySQL server has gone away

Reported by hiro | June 3rd, 2009 @ 02:49 AM | in 0.10.0

I met this messages:
merb : worker (port 8004) ~ (mysql_errno=2006, sql_state=HY000) MySQL server has gone away
Query: Mysql server has gone away. Please report this issue to the Datamapper project. Specify your at least your MySQL version when filing a ticket - (MysqlError)

I use this version:

mysqld --version

mysqld Ver 5.1.32-ndb-7.0.5-cluster-gpl for unknown-linux-gnu on x86_64 (MySQL Cluster Server (GPL))

Comments and changes to this ticket

  • Jerry West

    Jerry West June 3rd, 2009 @ 05:33 AM

    Wed, 03 Jun 2009 10:27:03 GMT ~ error ~ (mysql_errno=2006, sql_state=HY000) MySQL server has gone away
    Query: Mysql server has gone away. Please report this issue to the Datamapper project. Specify your at least your MySQL version when filing a ticket

    A simultaneous (separate) SSH shell running mysql on the remote db server is still alive and well:

    mysql> SELECT VERSION();
    +--------------------+ | VERSION() | +--------------------+ | 5.0.51a-3ubuntu5.4 | +--------------------+

    The client is Ubuntu 8.04.2 ruby 1.8.6 (2007-09-24 patchlevel 111) [i486-linux] with dm-core 0.9.11

  • Dan Kubb (dkubb)

    Dan Kubb (dkubb) June 3rd, 2009 @ 11:47 PM

    • Assigned user set to “Dirkjan Bussink”
  • Dirkjan Bussink

    Dirkjan Bussink June 4th, 2009 @ 04:00 AM

    Looks like you have auto reconnect disabled (the default since 5.1), could you verify your server configuration?

  • hiro

    hiro June 4th, 2009 @ 09:01 PM

    My MySQL Configuration is almost default. Maybe, auto reconnect is disabled.

    I investigate this behavior. I couldn't found the my.cnf option for enabling auto-reconnect.

  • Dirkjan Bussink

    Dirkjan Bussink June 20th, 2009 @ 01:55 PM

    • State changed from “unconfirmed” to “resolved”

    do_mysql now tries to do a full reconnect if needed. This will be part of the upcoming 0.10 release.

  • Ted Han (knowtheory)

    Ted Han (knowtheory) July 15th, 2009 @ 07:18 AM

    • Milestone set to 0.10.0
    • State changed from “resolved” to “unconfirmed”

    Hey Dirkjan,

    We're having this problem on 0.10, both on RC1, and on edge at the moment. The problem we're encountering (which we've had is very consistently across a few systems, EC2, as well a variety of local mac systems) is as follows:

    Merb boots up the spawner, and spins up however many worker processes you specify, and then at exactly the mysql timeout, regardless of how much activity the workers are generating, the spawner loses it's mysql connection AND somehow takes out all the worker connections.

    we're running 5.0 in production, and i'm running 5.1 locally.

  • Ted Han (knowtheory)

    Ted Han (knowtheory) July 15th, 2009 @ 07:36 PM

    • State changed from “unconfirmed” to “resolved”

    False alarm. Failure on our platform was caused by not being on edge.

    Connection still drops, but reconnection is successful

Please Sign in or create a free account to add a new ticket.

With your very own profile, you can contribute to projects, track your activity, watch tickets, receive and update tickets through your email and much more.

New-ticket Create new ticket

Create your profile

Help contribute to this project by taking a few moments to create your personal profile. Create your profile »

Pages