Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

SMTP Virtual Server Instance 1 Fails to Start

145 views
Skip to first unread message

Nik

unread,
Jan 5, 2005, 4:43:45 AM1/5/05
to
Hi

I have just built a 2 node A/P and am struggling with getting the SMTP
virtual instance to start.

I have trawled through the newsgroups and found similar issues but never a
fix.

The error that I am receiving is :

SMTP Virtual Server Instance 1 (servername): The IsAlive check for this
resource failed.
For more information, click http://www.microsoft.com/contentredirect.asp.

Event ID : 1005

I have checked in ESM and the Default SMTP server is correctly bound to the
Exchange VS IP address.

I'm kinda stuck as to were to start with looking for more info on this one,
can anyone point me in right direction ?

Cheers in advance

Nik

to mail direct remove plz.nospam. from the email address.


Evan Dodds [MSFT]

unread,
Jan 5, 2005, 3:30:38 PM1/5/05
to
Nik -

What's the error code listed for the IsAlive check failure? There are a lot
of things that can cause this, but hopefully the error code will point us in
a specific direction.

--
Evan Dodds
Microsoft Exchange Support

****** Disclaimer ******
This posting is provided "AS IS" with no warranties, and confers no rights.

Note: Please do not reply to this e-mail address. It is used for newsgroup
purposes only.

"Nik" <ni...@plz.nospam.gotadsl.co.uk> wrote in message
news:41dbb6cf$0$113$65c6...@mercury.nildram.net...

Nik

unread,
Jan 5, 2005, 5:52:27 PM1/5/05
to
"Evan Dodds [MSFT]" <ev...@online.microsoft.com> wrote in message
news:OxWAvV28...@TK2MSFTNGP09.phx.gbl...
Evan

All I am getting in the way of error codes is an event in the application
event log for MSExchangeCluster with ID 1005 and the text originally pasted.

The instance starts then fails, raises this even and then tries a few more
times between the nodes in the cluster. All other exchange services start
without issue. I have also tried to increase (significantly) the isalive
value to see if that helped, but with no joy.

Is there somewhere else I need to be looking to get the exact IsAlive check
failure error code ?

Cheers

Nik


Evan Dodds [MSFT]

unread,
Jan 6, 2005, 5:24:50 PM1/6/05
to
Nik -

Generally, you will find extended data associated with the error code in the
IsAlive failure. If you open up the event in event viewer and click the
button to copy out the event, then paste it into notepad do you see a bunch
of numbers in the section labeled "Data:"? This is the extended data and
will contain the error code that resulted in the IsAlive failure. If this
doesn't appear in the event viewer for some reason, you can also find this
data in the cluster.log file (although this is a bit more complex method to
explain in the newsgroup).

A few of the most common things I can think of that cause it to start up,
run for a minute or two, and then fail are: 1) Improper ServerBindings or
SecureServerBindings values 2) Broken Winsock 3) Inadequate system resources
to create the winsock connection for the IsAlive check, etc... could be any
of a number of things. The error code logged at the time of the failure will
hopefully point to a specific cause, or at least get you closer than the
generic IsAlive failure.

--
Evan Dodds
Microsoft Exchange Support

****** Disclaimer ******
This posting is provided "AS IS" with no warranties, and confers no rights.

Note: Please do not reply to this e-mail address. It is used for newsgroup
purposes only.

"Nik" <ni...@plz.nospam.gotadsl.co.uk> wrote in message

news:41dc6fab$0$105$65c6...@mercury.nildram.net...

Nik

unread,
Jan 7, 2005, 1:01:26 PM1/7/05
to
"Evan Dodds [MSFT]" <ev...@online.microsoft.com> wrote in message
news:ej98M6D9...@TK2MSFTNGP11.phx.gbl...

Evan

Heres what your looking for:

Event Type: Error
Event Source: MSExchangeCluster
Event Category: Services
Event ID: 1005
Date: 07/01/2005
Time: 12:21:19
User: N/A
Computer: server
Description:
SMTP Virtual Server Instance 1 (server): The IsAlive check for this resource
failed.

Data:
0000: 45 27 00 00 E'..

I figured this might be usefull also as it tends to point towards a winsock
problem.....

00000d1c.000007a8::2005/01/07-13:54:59.288 INFO Microsoft Exchange SMTP
Server Instance <SMTP Virtual Server Instance 1 (server)>: [EXRES] Thread
entered: DwMonitorSecondaryThread.

00000d1c.000007a8::2005/01/07-13:54:59.288 ERR Microsoft Exchange SMTP
Server Instance <SMTP Virtual Server Instance 1 (server)>: [EXRES]
DwCheckProtocolSocket: failed to connect socket. Error 10053.

00000d1c.000007a8::2005/01/07-13:54:59.288 ERR Microsoft Exchange SMTP
Server Instance <SMTP Virtual Server Instance 1 (server)>: [EXRES]
ExchangeCheckIsAlive: IsAlive failed, will retry in 50 msec.

00000c60.00000cac::2005/01/07-13:54:59.304 INFO [FM]
FmpHandleResourceTransition: Resource Name =
716365a1-03f1-4119-9d98-6acacca637bd [SMTP Virtual Server Instance 1
(server)] old state=129 new state=2

Hope this is what you needed.

Cheers

Nik


Evan Dodds [MSFT]

unread,
Jan 7, 2005, 8:07:51 PM1/7/05
to
Nik -

0x2745 = 10053 = definitely something Winsock related:

# for hex 0x2745 / decimal 10053 :
WSAECONNABORTED winerror.h
# An established connection was aborted by the software in
# your host machine.

So, this tells me that the cluster service was not able to sustain a
connection to port 25 to do the IsAlive check.

First thing I'd check is whether you can actually telnet to port 25 directly
and make a good connection: "telnet <the ip address of the virtual server>
25". If that doesn't work, you've definitely got a problem as the cluster
service will fail in exactly the same way your telnet attempt did. Some
Antivirus/Antispam software has been known to block access to port 25 in an
effort to prevent the SMTP service being used to relay spam, double check
that you don't have this feature enabled if you have any 3rd-party
Antivirus/Antispam software installed.

--
Evan Dodds
Microsoft Exchange Support

****** Disclaimer ******
This posting is provided "AS IS" with no warranties, and confers no rights.

Note: Please do not reply to this e-mail address. It is used for newsgroup
purposes only.

"Nik" <ni...@plz.nospam.gotadsl.co.uk> wrote in message

news:41dece74$0$10791$65c6...@mercury.nildram.net...

carsten lahme

unread,
Jan 23, 2005, 2:23:28 PM1/23/05
to
Hi all,
I had exactly the same Problem. And here is my solution:

After setting up the Cluster I went into Exchange Manager and secured the
SMTP-Server.
In our case we have an incoming and an outgoing relay, which takes the tasks
of antispam and antivirus. Therefore I only permitted the incoming
SMTP-Server access to the SMTP-Cluster.
And that's where my problem started. The Cluster tries to connect to the
SMTP-Service and is not allowed to (ip rejected) and IsAlive goes to false.
This results in Clusterswitching.
I had to give both ClusterNodeIps and the ClusterIP itself the permitions
and now it works.

May be this helps You a litle.
regards

Carsten

"Nik" <ni...@plz.nospam.gotadsl.co.uk> schrieb im Newsbeitrag
news:41dbb6cf$0$113$65c6...@mercury.nildram.net...

pfor...@gmail.com

unread,
Feb 1, 2005, 4:05:56 PM2/1/05
to

Nik

unread,
Feb 1, 2005, 4:47:48 PM2/1/05
to
Thanks for all the Help folks.

Evan pointed me in the right direction .... the underlying smtp service was
buggered .... seems to be a problem with the build .... used a standard
build and its all fine now.

Thanks again
<pfor...@gmail.com> wrote in message
news:1107291956.2...@z14g2000cwz.googlegroups.com...

0 new messages