I am having debugging a problem I am seeing with mysqldump.

While using mysqldump to make backups of my database, I will
intermitently get the following error :

mysqldump: Error 2013: Lost connection to MySQL server during query
when dumping table....

I am running the pre-compiled 64 bit linux 5.1.20 mysql binaries (
specifically, the Linux non RPM packages ) on a machine with 2 gigs of
memory.

The error has been seen (so far) in two tables. The first table is a
myisam table, which takes up 15 gigs on disk, and has 534858129 rows.
The second table is innodb, takes up 1.3 gigs on disk, and has 2055624
rows. The row the timeout error occurs on is always different.

The error seems to show up regardless of database usage (sometimes it
will show up when we are running intensive queries, other times it
will show up when no queries are being run).

The log-warnings option is enabled with a value of 2, but nothing
shows up in the error log when the timeout occurs.

I have set the interactive_timeout setting to a ridiculously high
number (172800), but the timeouts still occur.

I was experiencing this problem in version 5.1.18, and the upgrade to
5.1.20 has not eliminated the problem.

How should I go about debugging this problem?

I've created a bug for this problem (where there is more information
on the table schema, command line options used with mysqldump, etc) -
http://bugs.mysql.com/bug.php?id=30809

--
Aiton

The wait_timeout and interactive_timeout global variables are set to
172800, and max_allowed_packet is set to 30M, and I am still getting
the problem

> I am having debugging a problem I am seeing with mysqldump.
>
> While using mysqldump to make backups of my database, I will
> intermitently get the following error :
>
> mysqldump: Error 2013: Lost connection to MySQL server during query
> when dumping table....
>
> I am running the pre-compiled 64 bit linux 5.1.20 mysql binaries (
> specifically, the Linux non RPM packages ) on a machine with 2 gigs of
> memory.
>
> The error has been seen (so far) in two tables. The first table is a
> myisam table, which takes up 15 gigs on disk, and has 534858129 rows.
> The second table is innodb, takes up 1.3 gigs on disk, and has 2055624
> rows. The row the timeout error occurs on is always different.
>
> The error seems to show up regardless of database usage (sometimes it
> will show up when we are running intensive queries, other times it
> will show up when no queries are being run).
>
> The log-warnings option is enabled with a value of 2, but nothing
> shows up in the error log when the timeout occurs.
>
> I have set the interactive_timeout setting to a ridiculously high
> number (172800), but the timeouts still occur.
>
> I was experiencing this problem in version 5.1.18, and the upgrade to
> 5.1.20 has not eliminated the problem.
>
> How should I go about debugging this problem?
>
> I've created a bug for this problem (where there is more information
> on the table schema, command line options used with mysqldump, etc) -
> http://bugs.mysql.com/bug.php?id=30809
>
> --
> Aiton
>
> --
> MySQL General Mailing List
> For list archives: http://lists.mysql.com/mysql
> To unsubscribe: http://lists.mysql.com/mysql?unsub=xiaozhiwen@gmail.com
>
>
>


--
Aiton