Comment 8 for bug 1103045
Revision history for this message
| Daniel Nichter (daniel-nichter) wrote | #8 |
© 2004
Canonical Ltd.
•
Terms of use
•
Data privacy
•
Contact Launchpad Support
•
Blog
•
Careers
•
System status
•
a60fb26
(Get the code!)
There's a gray area here. The MySQL docs say that aborted connects happen for these reasons:
* A client does not have privileges to connect to a database.
* A client uses an incorrect password.
* A connection packet does not contain the right information.
* It takes more than connect_timeout seconds to get a connect packet.
MySQL will send an error packet for the first 3, but it won't for the last one or the telnet case when the connection is closed with CTRL-] because there's no error, the client just disappeared.
So "admin command: Connect aborted" in the current output is for that last case and the CTRL-] case. In other words, it means "client disappeared while connecting", which is why the event has no error number or error string.
I did that because I think that type of event is different from those first 3 reasons for aborted connects because those reasons result in MySQL sending an error packet with and error number and string.
They're all related to "Connect" events, but some have MySQL errors and others don't. I'm rather indifferent about whether we lump them altogether as "admin: Connect" or if we have "admin: Connect aborted" separate. What do you think Joe?