3689751 - Lumira intermittently hangs during "getting data from backend" workflow

• Occasional Lumira connectivity issue occurs during the "getting data from backend" workflow.
• The system has many Lumira dashboards based on BW
• The page hangs for a period of time, before eventually resolving itself without intervention
• In a jstack output (thread dump) of the Lumira server, it will be seen that lots of threads are waiting on the UniqueControllerStorage session map
  waiting to lock <0x000001e1fcd61c60> (a com.sap.ip.bi.webapplications.runtime.impl.controller.UniqueControllerStorage)
  
  One thread will have it locked, and is doing JCO RFC call:
   at com.sap.conn.rfc.driver.CpicDriver.nativeCpic_coxread([BI[B[I)I(Native Method)
   at com.sap.conn.rfc.driver.CpicDriver.cpic_coxread(I[B[I)I(CpicDriver.java:717)
   at com.sap.conn.rfc.driver.CpicDriver.coxread([BI[II)Lcom/sap/conn/rfc/exceptions/RfcIoRc;(CpicDriver.java:189)
   at com.sap.conn.rfc.driver.RfcTypeDirectCpic.read([BI[I)Lcom/sap/conn/rfc/exceptions/RfcIoRc;(RfcTypeDirectCpic.java:67)
   - locked <0x000001e2ba839210> (a com.sap.conn.rfc.driver.RfcTypeDirectCpic)
   at com.sap.conn.rfc.engine.RfcIoOpenCntl.ab_rfcread([BII)V(RfcIoOpenCntl.java:824)
   at com.sap.conn.rfc.engine.RfcIoOpenCntl.readRFCID()I(RfcIoOpenCntl.java:692)
  <TRUNCATED>
   at com.sap.ip.bi.webapplications.runtime.impl.controller.ControllerCollector.destroyController(Ljava/lang/String;)Lcom/sap/ip/bi/framework/base/controller/IController;(ControllerCollector.java:398)
   - locked <0x000001e1fcd61c60> (a com.sap.ip.bi.webapplications.runtime.impl.controller.UniqueControllerStorage)
• In the Lumira Server glf traces, requests will be seen be taking a long time to process for a period of time. Looking at where the time is spent, lots of JCO calls to BICS_PROV_CLOSE can be seen - all taking ~20 seconds, eg:
  2025 11 19 14:31:21.532|JCo callStarted() with clientID 2039252580, systemID <BW SYSTEM ID>, function_name BICS_PROV_CLOSE.  Passport: .
  2025 11 19 14:31:40.505|Exiting method 
• With additional JCO tracing enabled (1609534- How to trace SAPJCO For BI), more detail is seen in the Lumira Server GLF logs, eg:
  ControllerSessionTimeout <129> [2025-11-19 14:31:21.532]: [JCoRFC] Executing function BICS_PROV_CLOSE on handle [20/59953654] 
  ControllerSessionTimeout <129> [2025-11-19 14:31:40.504]: [JCoRFC] [traffic] Received 851 bytes for BICS_PROV_CLOSE
  ControllerSessionTimeout <129> [2025-11-19 14:31:40.505]: [JCoAPI] Exception occurred:
  com.sap.conn.jco.JCoException: (102) JCO_ERROR_COMMUNICATION: CPIC-CALL: CMRCV on convId: 59953654 with rc: 20
  LOCATION    CPIC (TCP/IP) on local host with Unicode
  ERROR       connection to partner '<BW HOST>:<BW PORT>' broken
  ERRNO TEXT  WSAECONNRESET: Connection reset by peer 
  
• In the BW gateway logs, filtering on the same conversation ID shows that no activity is seen on BW when Lumira tries to close the session. Later, BW times the session out by itself

• SAP BusinessObjects Business Intelligence (BI) Platform 4.x / 2025
• SAP Lumira
• Windows
• Linux / Unix 

Lumira connectivity issue, getting data from backend, JCO_ERROR_COMMUNICATION, CPIC-CALL, Azure firewall scale-in, session map locking, SAP BW communication, SAP Lumira troubleshooting, intermittent dashboard issue , KBA , BI-LUM-SRV-ADM , Lumira Server BI Platform Admin, Auth, Security , Problem