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 Historical content remains viewable.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

TMG 2010 connenctivity lost when number of denied TCP exceeds limi

973 views
Skip to first unread message

Chris Proud

unread,
Feb 1, 2010, 7:08:02 AM2/1/10
to
TMG 2010 on Windows 2008 R2 x64

Whenever I get the alert "The number of denied TCP and non-TCP packets per
second exceeded the system limit. As a result, Forefront TMG reduced the
number of records of denied packets that are written in the log." all the TCP
port connectivity verifiers fail and any other new outgoing tcp connections,
like the SMTP alert. PING verifiers don't appear to be affected neither
current connections (I think). I think it is also affecting new connections
from workstations.

It appears to happen sporadically. Some times it can occur every 5 minutes
for half an hour, sometimes longer. It cleared itself this morning but the
other day a restart seemed to do the trick. Maybe it is linked to
something/one scanning our server.

Thanks

Jens Baier

unread,
Feb 1, 2010, 2:52:55 PM2/1/10
to
Hi,

> Whenever I get the alert "The number of denied TCP and non-TCP packets per
> second exceeded the system limit. As a result, Forefront TMG reduced the
> number of records of denied packets that are written in the log." all the
> TCP
> port connectivity verifiers fail and any other new outgoing tcp
> connections,
> like the SMTP alert. PING verifiers don't appear to be affected neither
> current connections (I think). I think it is also affecting new
> connections
> from workstations.

create an execption for these IP Addresses in the Flood Mitigation settings

--
Gruss Jens
www.it-training-grote.de
www.forefront-tmg.de
https://mvp.support.microsoft.com/profile/Marc.Grote
http://blog.it-training-grote.de

Chris Proud

unread,
Feb 4, 2010, 5:57:01 AM2/4/10
to
Hi Gruss,

Thanks for you suggestion. I have already disabled flood mitigation because
it was causing all kinds of other problems! (I unticked "Mitigate flood
attached and work propagation", should I disable it anywhere else?)

What seems to be happening in my case is that ALL new connections are denied
when the global denied packets event occurs. I can't see how adding an
exception would help - the event indicates a global deny limit, not specific
to any IP. Where would I add the exception too?

Also, the event description does not indicate any kind of blocking should
occur, just that its going to stop logging the packets.

Thanks

Chris

Chris Proud

unread,
Feb 11, 2010, 11:00:01 AM2/11/10
to
Bump!

Chris Proud

unread,
Apr 21, 2010, 9:05:01 AM4/21/10
to
This has started to happen again. Its being quite persitent at the moment.

Does anyon have any ideas what might be causing the problem?

Phillip Windell

unread,
Apr 22, 2010, 2:37:23 PM4/22/10
to
You're gonna have to find the source.

They should be in the logs,..it still logs them,...it only says that it is
reducing the records, not eliminating them.

Does the alert give the source IP#?

It could be an infection on the LAN or a DoS attack from outside,...that's
why you have to find the source. The problem is not ISA,..the problem is
what is bombing the ISA.

--
Phillip Windell

The views expressed, are my own and not those of my employer, or Microsoft,
or anyone else associated with me, including my cats.
-----------------------------------------------------

"Chris Proud" <Chris...@discussions.microsoft.com> wrote in message
news:E317C5DD-3CC9-4D53...@microsoft.com...

Luis Marques

unread,
Mar 8, 2012, 9:35:24 AM3/8/12
to
Hi Chris,

did u find something? i have exactly the same problems since a few days

> On Monday, February 01, 2010 7:08 AM Chris Proud wrote:

> TMG 2010 on Windows 2008 R2 x64
>
> Whenever I get the alert "The number of denied TCP and non-TCP packets per
> second exceeded the system limit. As a result, Forefront TMG reduced the
> number of records of denied packets that are written in the log." all the TCP
> port connectivity verifiers fail and any other new outgoing tcp connections,
> like the SMTP alert. PING verifiers do not appear to be affected neither
> current connections (I think). I think it is also affecting new connections
> from workstations.
>
> It appears to happen sporadically. Some times it can occur every 5 minutes
> for half an hour, sometimes longer. It cleared itself this morning but the
> other day a restart seemed to do the trick. Maybe it is linked to
> something/one scanning our server.
>
> Thanks


>> On Monday, February 01, 2010 2:52 PM Jens Baier wrote:

>> Hi,
>>
>>
>> create an execption for these IP Addresses in the Flood Mitigation settings
>>
>> --
>> Gruss Jens
>> www.it-training-grote.de
>> www.forefront-tmg.de
>> https://mvp.support.microsoft.com/profile/Marc.Grote
>> http://blog.it-training-grote.de


>>> On Thursday, February 04, 2010 5:57 AM Chris Proud wrote:

>>> Hi Gruss,
>>>
>>> Thanks for you suggestion. I have already disabled flood mitigation because
>>> it was causing all kinds of other problems! (I unticked "Mitigate flood
>>> attached and work propagation", should I disable it anywhere else?)
>>>
>>> What seems to be happening in my case is that ALL new connections are denied
>>> when the global denied packets event occurs. I cannot see how adding an
>>> exception would help - the event indicates a global deny limit, not specific
>>> to any IP. Where would I add the exception too?
>>>
>>> Also, the event description does not indicate any kind of blocking should
>>> occur, just that its going to stop logging the packets.
>>>
>>> Thanks
>>>
>>> Chris


>>>> On Thursday, February 11, 2010 11:00 AM Chris Proud wrote:

>>>> Bump!


>>>>> On Wednesday, April 21, 2010 9:05 AM Chris Proud wrote:

>>>>> This has started to happen again. Its being quite persitent at the moment.
>>>>>
>>>>> Does anyon have any ideas what might be causing the problem?
>>>>>
>>>>> "Chris Proud" wrote:


>>>>>> On Thursday, April 22, 2010 2:37 PM Phillip Windell wrote:

>>>>>> You're gonna have to find the source.
>>>>>>
>>>>>> They should be in the logs,..it still logs them,...it only says that it is
>>>>>> reducing the records, not eliminating them.
>>>>>>
>>>>>> Does the alert give the source IP#?
>>>>>>
>>>>>> It could be an infection on the LAN or a DoS attack from outside,...that is
0 new messages