Udostępnij za pośrednictwem


Lync 2010 Backend-database server went down

So, when you have Front-end edition pool with SQL Server and the SQL Server is down for some reason. Then your Lync server will be in resilient mode with REGISTRAR Component primarily active to server your user

The following features won’t be available

  • Conferencing (IM, Video and Web)
  • Presence & DND based routing
  • Modify Presence or Change Call Forwarding Settings
  • Contact List
  • Response Group and Call Park

The following features will still be available due to modular REGISTRAR component built in the FE/SBA server

  • PSTN Inbound and Outbound calls
  • Intra-Site calls
  • Inter site calls (PSTN Rerouting)
  • Hold, Retrieve, Transfer
  • Authentication, Authorization
  • Voicemail Deposit (Redirect to Exchange UM in Data Center)
  • Voicemail Retrieve (through PSTN)
  • Call Forwarding, SimulRing
  • Boss-Admin, Team-call
  • Call Detail Records (CDR)
  • All 2 Party Intra Site communications
  • Audio Conferencing through PSTN
  • Contact Search

 

This is because of the Registrar component on the FE server which resides on the Front-End Server

And the local copy of the SQL Server database RTCLOCAL on each FE/SBA

On the SIP log, lync server saying that it lost connection to backend database server

 

09/28/2011|13:48:54.266 280:3C4 INFO  :: Data Received - 172.31.10.31:5061 (To Local Address: 172.31.10.211:58647) 710 bytes:

09/28/2011|13:48:54.266 280:3C4 INFO  :: SIP/2.0 503 Service unavailable

Authentication-Info: TLS-DSK qop="auth", opaque="F947CF13", srand="F77F22BC", snum="3", rspauth="d6ac9feba989ad7ce618372ad1e5fb3eb2563acc", targetname="NTLyncEE01.nwtraders.com", realm="SIP Communications Service", version=4

From: "Robert Lea"<sip:robert@nwtraders.co.nz>;tag=e3fbc1f23a;epid=ccd46068cb

To: <sip:robert@nwtraders.co.nz>;tag=F9E68E7E2D7C2F5F42D91E1AD7F8DC12

Call-ID: dfe9ff99be614b5bb612e533dbaa4af5

CSeq: 1 SUBSCRIBE

Via: SIP/2.0/TLS 172.31.10.211:58647;ms-received-port=58647;ms-received-cid=F0B00

ms-diagnostics: 4144;reason="Database communication link is down";source="NTLyncEE01.nwtraders.com"

Retry-After: 54

Server: RTC/4.0

Content-Length: 0

 

 

09/28/2011|13:48:54.266 280:3C4 INFO  :: End of Data Received - 172.31.10.31:5061 (To Local Address: 172.31.10.211:58647) 710 bytes

Comments

  • Anonymous
    January 01, 2003
    Great article.

    I had a question about client recovery time. Once the back-end is available, do clients immediately return to full mode or is there a delay between back-end returning, Lync services detecting the return and the clients being signalled that full service is restored?

    A signout/in, immediately returns the client to full mode once the back-end is restored.

    Is there a specific SIP message the Front Ends use to signal to clients that full service is restored?

    regards
    Jon