Hello

We have 2 Exchange 2013 CU7 servers.

Sometime I get an error message on phones that says it cannot log in.

I can see from event log that

[Eas] The number of outstanding requests for guard TargetBackend("exch01.top.domain.no") has exceeded the max limit 150. Current request will be rejected.

Anyone know what limit its refering to? And should it be set higher?

Regards

Hi , don't know for sure ,  but i guess you can find it somewhere in the throttling policy. 

You might wanna have look into that one : From Exchange Powershell : Get-Throttling-Policy

And see if there is something with a maximum of 150 which you can refer to EAS.

You can check the following things

Please check for any errors logged in RPC logs for the same in all the cas servers

Also look for any additional events triggered for activesync in APP log
Possibly set the event log level to expert for activesync and look for any errors
get-eventloglevel MSExchange ActiveSync\Requests | set-eventloglevel -level 5

Hi Kristinansen,

I agree with Wint.

Please verify whether this caused by Throttling Policy. I find some related parameters for your reference:

1. EasMaxConcurrency : It specifies how many concurrent connections an Exchange ActiveSync user can have against a server running Exchange 2013 at one time.

2. EasMaxDevices : It specifies a limit to the number of Exchange ActiveSync partnerships that a user can have at one time.

More details to see: http://technet.microsoft.com/en-us/library/dd351045(v=exchg.150).aspx

If it is the Throttling Policy issue, please run following command example to re-set the limit.

Set-ThrottlingPolicy <ThrottlingPolicyName> -EasMaxConcurrency 300

More details to see: http://technet.microsoft.com/en-us/library/dd298094(v=exchg.150).aspx

 

Hi Mavis,

I have the same Problem (new Exchange 2013 Installation CU7). The value at the DefaultThrottlingPolicy and GlobalThrottlingPolicyis as follows: EASMaxConcurrency: 10 / EASMaxDevices: 100 I found nowhere a default value of 150. Where does the value 150? Under EASMaxConcurrency the value is 10, not 150. What do you recommend?

Thanks for your Support.

Marco

This is the same for mee, EASMaxConcurrency is 10 on both throttlig policys, the number 150 I cant find anywhere.

Getting the same. Also an Exchange 2013 CU7 Environment.

Same Default Throttling for Concurrent Devices and Devices. 10 and 100.

Doing a Get-ThrottlingPolicy | fl No mention of 150.

I'm seeing the same on a CU7 environment. The load did not increase and users do not report actual issues.

We are also seeing the same issue. CU7, no reference in ThrottlingPolicy to 150. 

I am seeing this error on the Microsoft Web Application Proxy

The HTTP response from the backend server was not received within the expected interval. Expected interval: 300 seconds.

Details:
Transaction ID: {f4be1653-43e9-0000-02d0-bef4e943d001}
Session ID: {f4be1653-43e9-0000-02d0-bef4e943d001}
Published Application Name: Exchange Microsoft-Server-ActiveSync
Published Application ID: 9ad23d25-1d98-7a66-3671-6cd7aabcab29
Published Application External URL: https://mailserver.address.com/Microsoft-Server-ActiveSync/
Published Backend URL: https://mailserver.address.com/Microsoft-Server-ActiveSync/
User: <Unknown>
User-Agent: Apple-iPhone5C2/1104.257
Device ID: <Not Applicable>
Token State: NotFound
Cookie State: NotFound
Client Request URL: https://mailserver.address.com/Microsoft-Server-ActiveSync?User=user@address.com&DeviceId=ApplF2LL1VVXDTWF&DeviceType=iPhone&Cmd=Ping
Backend Request URL: https://mailserver.address.com/Microsoft-Server-ActiveSync?User=user@address.com&DeviceId=ApplF2LL1VVXDTWF&DeviceType=iPhone&Cmd=Ping
Preauthentication Flow: PassThrough
Backend Server Authentication Mode: PassThrough
State Machine State: BEBodyWriting
Response Code to Client: <Not Applicable>
Response Message to Client: <Not Applicable>
Client Certificate Issuer: <Not Found>

have anyone fixed this?

Thanks Edd

We had this exact same problem, and after a supportcase with Microsoft, we got a solution that worked for us:

On all CAS servers:

In \Exchange Server\V15\FrontEnd\HttpProxy\sync\web.config, add this key in the <appSettings> node.

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

In \Exchange Server\V15\FrontEnd\HttpProxy\rpc\web.config, add this key in the <appSettings> node.

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

After this change recycle the MSExchangeSyncAppPool and MSExchangeRPCProxyAppPool in IIS manager.

After this, we stopped getting Event ID 2002, and all Connection_Dropped_Event_List_Full stopped appearing in Httperror logs.

This was a "new find", and the technician that had worked this out was going to inform the developers. He recommended a minimum of 2000 (default: 150), but in our case he instructed 5000. (for 25000 users)

Hopefully this will work for more people.

• Proposed as answer by kjetilnor Friday, February 20, 2015 1:23 AM

We had this exact same problem, and after a supportcase with Microsoft, we got a solution that worked for us:

On all CAS servers:

In \Exchange Server\V15\FrontEnd\HttpProxy\sync\web.config, add this key in the <appSettings> node.

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

In \Exchange Server\V15\FrontEnd\HttpProxy\rpc\web.config, add this key in the <appSettings> node.

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

After this change recycle the MSExchangeSyncAppPool and MSExchangeRPCProxyAppPool in IIS manager.

After this, we stopped getting Event ID 2002, and all Connection_Dropped_Event_List_Full stopped appearing in Httperror logs.

This was a "new find", and the technician that had worked this out was going to inform the developers. He recommended a minimum of 2000 (default: 150), but in our case he instructed 5000. (for 25000 users)

Hopefully this will work for more people.

• Proposed as answer by kjetilnor Friday, February 20, 2015 1:23 AM

We had this exact same problem, and after a supportcase with Microsoft, we got a solution that worked for us:

On all CAS servers:

In \Exchange Server\V15\FrontEnd\HttpProxy\sync\web.config, add this key in the <appSettings> node.

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

In \Exchange Server\V15\FrontEnd\HttpProxy\rpc\web.config, add this key in the <appSettings> node.

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

After this change recycle the MSExchangeSyncAppPool and MSExchangeRPCProxyAppPool in IIS manager.

After this, we stopped getting Event ID 2002, and all Connection_Dropped_Event_List_Full stopped appearing in Httperror logs.

This was a "new find", and the technician that had worked this out was going to inform the developers. He recommended a minimum of 2000 (default: 150), but in our case he instructed 5000. (for 25000 users)

Hopefully this will work for more people.

• Proposed as answer by kjetilnor Friday, February 20, 2015 1:23 AM

It Works! Thank you for your helpful contribution. Best regards Marco

Thank you Kjetilnor,

I just updated to CU7 this past weekend and your post has cleared up my error logs. I do appreciate you sharing the fix before it filtered its way through the Microsoft channels. 

John

Just an FYI,

This issue is also present in CU8.

Confirmed that this issue still exists in CU9.

Does anyone know if there are associated perfmon counters that can be monitored that correspond to these settings?

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

    <add key="HttpProxy.ConcurrencyGuards.TargetBackendLimit" value="5000" />

Thanks!!  Tom