I have two mx records with a pref of 10 and 20. 10 - back end server and
20 - front end server.
Also i have a cluster back end which is receving mail from the front end
with the pref of 20.
When i send email from yahoo or google. the mail usually gets stuck in the
old back end server queue but eventually go through. Also sending email to
the outside i dont get that imediatly either.
To add to this, i cannot receive mail when user A from old server sends mail
to userB on new cluster back end server. But vice versa works fine.
This is really weird. I am not using any smart host and just using regular
dns lookups.
Also on the front end server queue, i get lots of serv...@domain.com
messages queued up.
Could this be a dns issue? I am running both internal dns and external is
managed by isp. All my exchange servers have static ips and have A-records
+ ptr in the internal dns server. No mx records for internal.
please help me
Thank you
Update you networkdrivers to the latest version (especially if you have
Broadcom cards).
Then change some registry settings for tcp/ip:
http://support.microsoft.com/?kbid=912222
To turn off TCP/IP Chimnet settings:
Through the registry:
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\Tcpip\Parameters
The global registry entries are as follows:
. EnableTCPChimney
Type: REG_DWORD
Values: 1 (enabled) 0 (disabled))
. EnableRSS
Type: REG_DWORD
Values: 1 (enabled) 0 (disabled))
. EnableTCPA
Type: REG_DWORD
Values: 1 (enabled) 0 (disabled))
The EnableTCPChimney global registry value enables the TCP Chimney
functionality in the stack. Similarly, the EnableRSS and EnableTCPA global
registry values enable the RSS functionality and the TCPA functionality in
the stack.
Restart the server after done this setting.
See for backgroup information this article:
http://msexchangeteam.com/archive/2007/07/18/446400.aspx
--
Regards,
Rene Frenger
"nobody" <nob...@nobody.com> wrote in message
news:u6qXbH97...@TK2MSFTNGP05.phx.gbl...