none
hard direct fail and no available shadow servers RRS feed

  • Question

  • I was going through message tracking, external user may not be receiving emails from us, and I ran into these two errors I had not seen before, Hard Direct Fail and No Suitable Shadow Servers.

    the email in question is being forwarded from an AS400 through our Exchange 2013 through a relay connector, also we have two Exchange 2013 servers and a witness server to form a DAG group.

    does the shadow server refer to the 2nd exchange in the DAG group?

    I ran several status checks on the DAG group, and status appears to be ok.

    Monday, March 16, 2015 2:34 PM

Answers

  • Hi,

    “No suitable shadow servers” suggest server is looking for another E2K13 / Hub for redundancy. We can disable ShadowRedundancy on server.

    Set-TransportConfig -ShadowRedundancyEnabled $false

    Thanks,

    Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

     


    Simon Wu
    TechNet Community Support

    Tuesday, March 17, 2015 8:50 AM
    Moderator

All replies

  • Hi,

    “No suitable shadow servers” suggest server is looking for another E2K13 / Hub for redundancy. We can disable ShadowRedundancy on server.

    Set-TransportConfig -ShadowRedundancyEnabled $false

    Thanks,

    Please remember to mark the replies as answers if they help and unmark them if they provide no help. If you have feedback for TechNet Subscriber Support, contact tnmff@microsoft.com

     


    Simon Wu
    TechNet Community Support

    Tuesday, March 17, 2015 8:50 AM
    Moderator
  • To further validate Simon_Wu's post - We were receiving these HAREDIRECTFAIL events in get-messagetracking. 

    I checked  out Mailbox connectivity submissions (Transport) logs: C:\Program Files\Microsoft\Exchange Server\V15\TransportRoles\Logs\Mailbox\Connectivity\Submission:

    2016-02-19T14:41:26.184Z,08D32EF0D70A2544,SMTP,mailboxtransportsubmissioninternalproxy,>,Failed connection to 10.10.10.128:2525 (ConnectionRefused:0000274D)[TargetHost:exchsrv2.domain.local:2525|MarkedUnhealthy|FailureCount:1|NextRetryTime:2016-02-19T14:41:56.856Z][TargetIPAddress:10.10.10.128:2525|MarkedUnhealthy|FailureCount:1|NextRetryTime:2016-02-19T14:41:56.856Z]

    Which explained why some users we noticing very late delivery attempts on whitelisted domains. We do run greylisted but only for unknown/not trusted domains/email addresses.

    a quick:

    get-transportconfig | select ShadowRedundancyEnabled 

    showed it was $true but we've never had more than one exchange server. Running Simon_Wu's command made them stop instantly. We are running Exchange 2013 CU10 fully patched as of 2/6/2016

    Friday, February 19, 2016 3:10 PM
  • I wonder why SHADOWREDUNDANCY seems ENABLED by default even if there is only a single Exchange running.

    @Nathaniel: Did you notice any other side-effect from DISABLEing it?

    Friday, April 26, 2019 9:33 AM
  • That's why i have my bigger customers with DAG and KEMP still in 2010 ;-)

    Spent so much in in understanding the KI from Exchange 2013/2016 and what he does decide by itself.
    We made our own Health scripts before we installed first 2013 to stop such KI things the Exchange does itself. Most of the times this helps people keep it up running but it you lock down logs you get confused more than it helps sometimes.

    I think or i guess:

    * If the Exchange 2013/2016 throttles E-Mail delivery example 452 4.3.1 insufficient system resources. When Transport is not fast enough to pickup further incoming SMTP E-Mail than he tries to handle that to another server. 

    Now i think in Hybrid Deployment (1xCLOUD, 1xONPREMISE) he than checks if the customer has such a setup and has the domain active at prod.outlook.com. If you he would push wotk to the cloud member to handle. Kind of workload split which seems ok from the logic.

    You can see that in the LOGS with the mentioned HAREDIRECTFAIL.

    So all he does if the Exchange 2013/2016 is running’s at its limit (Memory, Disk Space) and Back Pressure comes into play he TRIES to OFFLOAD to Other Exchange Server in your SITE or even external to Cloud.
    SO conclusion:

    They moment you should see that event the server is at his limits. Check memory/CPU/DISK and you only should see those when this happens. So this is secondary problem which shows up because you have a primary problem.

    servers,,SMTP,HAREDIRECTFAIL,15423227559936,CH2PR07MB7208BFC66CF8D279AE56A303C9380@CH2PR07MB7208.namprd07.prod.outlook.com

    https://www.butsch.ch/post/Exchange-2010-452-431-insufficient-system-resources-(No-E-Mail-incoming).aspx



    Friday, January 10, 2020 2:34 PM