I'm having trouble with a an obscene amount of aborted_connects on my
MySQL database servers.  I have 3 dedicated database servers, one of them is
a master database with 2 slaves, and the other 2 are stand-alone
database servers.  All connections are being made by PHP pages on Apache web
servers.

Most database connections come in through a private pocket network.

Network 1 consists of 2 machines on a private LAN using a switch. The NICs
on both servers are in full duplex mode.  Both servers are running Linux;
the web server is running VA Linux 2.2.18pre11-va2.0smp #1 SMP and the
database
server is running
VA Linux 2.4.5-beta4va3.17smp-piii #1 SMP.

The database server is a dual PIII with 2GB of RAM running the MySQL 3.23.52
binary.
A snapshot shows that aborted_connects are at 149605 when total connections
are 2094839.

The aborted_connects grow so quickly, so fast, that the server requires a
"flush-hosts" command to allow database connectivity.

Network 2 consists of 2 database servers and 4 web servers that communicate
on a private LAN using a hub.  All 6 of these servers are running FreeBSD
4.6.2; NICs are in full duplex mode.

Stand alone server #1 is a PII with 512MB of RAM.  A snapshot shows that
aborted_connects are at 17789 when total connections are 35490.

Stand alone database server #2 is a PIII with 768MB of RAM.  A snapshot
shows that aborted_connects are at 113537 when total connections are 190162.

Both servers are generally connected to through the private network.  When
the connections come in over the Internet the aborted_connects sky rocket
very quickly.


---------------------------------------------------------------------
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/           (the list archive)

To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php

Reply via email to