Hello,
I am having some trouble with some users connecting to my ngIRCd server
- this only seems to affect some users.
Here's an extract from my log:
> Apr 28 10:46:39 jupiter ngircd[1600]: Accepted connection 13 from
> 58.160.178.xxx:54009 on socket 10.
> Apr 28 10:46:39 jupiter ngircd[14258]: Can't resolve
> "CPE-58-160-178-xxx.hbqy1.fli.bigpond.net.au": Name or service not known
> Apr 28 10:46:39 jupiter ngircd[14258]: Possible forgery:
> 58.160.178.xxxresolved to CPE-58-160-178-xxx.hbqy1.fli.bigpond.net.au
> (which has no ip address)
> Apr 28 10:46:39 jupiter ngircd[1600]: New TLSv1 connection using
> cipher AES128-SHA on socket 13.
> Apr 28 10:46:39 jupiter ngircd[1600]: User "Skiitz!~.(a)58.160.178.xxx"
> registered (connection 13).
> Apr 28 10:46:39 jupiter ngircd[1600]: Write buffer overflow
> (connection 13, size 4036 byte)!
> Apr 28 10:46:39 jupiter ngircd[1600]: Shutting down connection 13
> (Write buffer overflow!) with 58.160.178.xxx:54009 ...
> Apr 28 10:46:39 jupiter ngircd[1600]: SSL connection 13 shutting down ...
> Apr 28 10:46:39 jupiter ngircd[1600]: User "Skiitz!~.(a)58.160.178.xxx"
> unregistered (connection 13): Write buffer overflow!
> Apr 28 10:46:39 jupiter ngircd[1600]: Connection 13 with
> 58.160.178.xxx:54009 closed (in: 0.0k, out: 11.9k).
My interpretation of the log is that the client is exceeding ngIRCd's
buffer size - can this parameter be tuned, or is this a misbehaving
(outside of RFC) client?
Terry.