Larry A wrote:
> we also have started seeing this TCP time out error.
Doing a telnet port 25 to the destination mail server will
> work for us. In trying to track one of these down,
the destination mail server administrator said that their
network
> does not accept ICMP packets. Could SurgeMail be
checking the destination with a ping before opening port 25?
Their
> DNS entries all seem correct.
Give me an example to test. But in brief, no surgemail is
just opening a normal ordinary tcp connection, so if it
fails
it means there is a network error or the remote end refused
the connection.
If you can telnet ok but can't send via surgemail, and you
are sending to the same host (being sure to use the mx
hosts), then that would be most odd. Send me the msg*.rec
logs showing this and I'll see if I can reproduce it from
another server.
The only other situation I know of is if surgemail tries to
open a bunch of connections to a host really fast then it
might trigger a DOS evasion from the remote end, which looks
just like a failed connection. The logs should help to
indicate if this might have been the case. Send me all
the msg*.rec entries a few minutes either side of the
problem.
ChrisP.
>
> Larry
> =============================================
> ----- Original Message -----
> From: Support ChrisP <surgemail-support netwinsite.com>
> To: surgemail-list netwinsite.com
> Cc: mike mailoutinteractive.com
> Subject: Re: [SurgeMail List] Which timeout setting?
> Date: Thu, 20 Dec 2007 13:12:52 +1300
>
>> Changing the timeout won't help with that error,
that error means that a network issue exists, or firewall,
or the
>> remote server is faulty or down.
>>
>> Test it manually like this from your server:
>> telnet remote.server.ip.address 25
>> And see if it connects, then if it does resend the
message and see if surgemail can send it.
>>
>> If telnet fails then the network issue exists or
the remote end is not running
>>
>> ChrisP
>>
>>
>> Mike Bridge wrote:
>>> Hi-
>>>
>>>
>>> I was trying to adjust the timeout for
delivering mail, because I'm
>>> getting errors like this for a particular
server:
>>>
>>>
>>> 19 12:13:41 [3340191] Later 207.176.155.55
<http://207.176.155.55>
a>
>>> <...> <....> 0 "recent slow
open age=82 delay=180 399 TCP Read failed
>>> (Socket Timeout after 180 seconds) 180
sec"
>>>
>>> 180 seconds is too long for me to wait---which
of these (or something
>>> else) will adjust this setting?
>>>
>>>
>>> g_send_open_timeout - SMTP link open
timeout
>>>
>>> Timeout, in seconds when opening an SMTP link.
>>>
>>>
>>> g_send_timeout - Send timeout
>>>
>>> Timeout, in seconds when sending mail, default
is 540 (9 minutes)
>>>
>>>
>>> Thanks!
>>>
>>> -Mike
>>>
>>>
>>>
>>>
>
|