C135 PCS 7 (PCS7)
6C
6 Cielo Ciencia y Tecnología
·Publicada en C135 PCS 7 (PCS7)

Troubleshooting Process Historian Connection Failures to Redundant OS Servers in PCS 7

Troubleshooting Process Historian Connection Failures to Redundant OS Servers in PCS 7

When a standalone Siemens Process Historian (PH 2014 / 2020) connects and receives data from only one OS server in a redundant pair (e.g., OSS2 works, but OSS1 fails and master switchover halts archiving), the issue is caused by a corrupted PH-Ready configuration file (ProcessHistorian.cfg) or a blocked Store and Forward (S&F) cache on the failing server. To resolve this: 1) On the failing OS server, navigate to C:\ProgramData\Siemens\SFCache and purge corrupted queue folders (AlgCSDataQueue, TagCSDataQueue); 2) In the WinCC project directory, rename ProcessHistorian.cfg to .old and restart runtime to force clean GUID regeneration; 3) Verify Microsoft Message Queuing (MSMQ) is running; and 4) Confirm that bidirectional DNS/NetBIOS hostname resolution and NTP time synchronization match between PH and both OS servers.

Table of Contents

Executive Summary

In high-reliability SIMATIC PCS 7 process architectures, the Process Historian (PH) provides long-term centralized archiving for all process values, batch records, and alarm events originating from redundant WinCC OS Servers. A critical failure mode occurs when the Process Historian Management console visualizes only one OS server (e.g., OSS2), while the redundant partner (OSS1) fails to transfer data. If an operational upset causes a master-standby switchover making OSS1 the active master, the Process Historian stops receiving all live plant data. This technical guide outlines the PH-Ready and Store & Forward (S&F) caching mechanics, details diagnostic logs, and provides the verified recovery procedure to establish bidirectional communication with both redundant OS servers.

Architecture and Fundamental Mechanics

Process Historian Client-Server Architecture and Caching Mechanics

+-----------------------+                         +-----------------------+
| OS Server 1 (OSS1)    |                         | OS Server 2 (OSS2)    |
| WinCC Project Dir:    |                         | WinCC Project Dir:    |
| ProcessHistorian.cfg  |                         | ProcessHistorian.cfg  |
| [CORRUPTED / BLOCKED!]|<=== WinCC Redundancy ===| [HEALTHY BINDING]     |
+-----------|-----------+                         +-----------|-----------+
            | MSMQ / TCP 4224                                 | MSMQ / TCP 4224
            | [COMMUNICATION HALTED]                          | [DATA FLOWING OK]
            v                                                 v
+-------------------------------------------------------------------------+
|                 Siemens Process Historian Server (PH)                   |
|                                                                         |
|   - Microsoft SQL Server Enterprise Instance                            |
|   - Process Historian Management Console:                               |
|     * OSS2 Status: CONNECTED (Green)                                    |
|     * OSS1 Status: DISCONNECTED / MISSING (Red / Not Detected)          |
|                                                                         |
|   FAILURE SYMPTOM:                                                      |
|   - Redundancy switchover: OSS1 becomes Master                          |
|   - Archiving to PH completely freezes!                                 |
+-------------------------------------------------------------------------+

Primary Root Causes for One-Sided Connections

  1. Corrupted ProcessHistorian.cfg: On each OS server, ProcessHistorian.cfg stores the unique pairing token, GUID, and encryption certificates connecting that specific WinCC station to the PH database. If an OS server was duplicated, renamed, or restored from an old backup, this file contains invalid credentials, causing the PH server to reject its data packets.
  2. Store and Forward (S&F) Queue Lockup: When network interruptions occur, the OS Server caches unsent archives in C:\ProgramData\Siemens\SFCache via Microsoft Message Queuing (MSMQ). If an MSMQ packet becomes malformed, the S&F queue deadlocks, preventing both live and cached data transmission.
  3. Name Resolution and Time Drift: The PH infrastructure communicates strictly via computer hostnames (NetBIOS / DNS), not IP addresses. Furthermore, if time drift between OSS1 and the PH exceeds 1,000 ms, the PH database engine rejects incoming archive records due to timestamp plausibility violations.

Step-by-Step Implementation Procedure

Step 1: Pre-Flight Network, Service, and User Verification

  1. From the Process Historian server, open Command Prompt and test hostname pinging to both servers: ping OSS1 and ping OSS2.
  2. From both OS servers, ping the historian: ping PH_SERVER.
  3. In Windows Services (services.msc) on both OS servers, verify that Message Queuing and CCAPHServer (PH-Ready Service) are in status Running.
  4. Verify that the Windows user account configured for PH-Ready has administrative privileges and is a member of SIMATIC HMI.
  5. Verify that NTP time synchronization is active across all stations, ensuring time drift between OS servers and PH is under 100 ms.

Step 2: Purge S&F Cache and Reset PH-Ready on the Failing Server

  1. To avoid plant interruption, perform this procedure sequentially without shutting down both servers simultaneously.
  2. On the failing server (e.g., OSS1), deactivate WinCC Runtime (if it is currently the standby server; if it is master, trigger a controlled switchover to OSS2 first).
  3. In Windows Explorer, navigate to the WinCC project directory on OSS1.
  4. Locate ProcessHistorian.cfg and rename it to ProcessHistorian.cfg.old.
  5. Open Windows Services (services.msc) and stop the service Message Queuing (MSMQ).
  6. Navigate to the Store and Forward cache directory: C:\ProgramData\Siemens\SFCache.
  7. Backup and delete all files and subfolders ending in Queue (e.g., AlgCSDataQueue, TagCSDataQueue).
  8. Restart the Message Queuing service.
  9. Rename ProcessHistorian.cfg.old back to ProcessHistorian.cfg (or leave it removed if you want WinCC to regenerate a clean configuration on connection).

Step 3: Run SIMATIC Security Control

  1. On OSS1, launch Start > Siemens Automation > Security Control.
  2. Click the button Repeat Settings to re-establish DCOM permissions and Windows firewall exceptions.
  3. Restart the OSS1 PC station to ensure all services initialize cleanly.

Step 4: Re-Establish Connection and Synchronize

  1. After reboot, launch WinCC Explorer on OSS1 and activate WinCC Runtime.
  2. Wait for WinCC Redundancy synchronization to complete (green status in Redundancy Control).
  3. On the Process Historian server, open Process Historian Management Console.
  4. Navigate to Detailed Plant Structure Information and click the Update button on the far right.
  5. Confirm that OSS1 now appears in the plant tree alongside OSS2 with active green connection status.

Verification

  1. Management Console Audit: In the Process Historian Management Console, verify that both redundant partners (OSS1 and OSS2) display green status indicators under Plant Structure.
  2. S&F Recovery Check: On OSS1, inspect C:\ProgramData\Siemens\SFCache. Confirm that queue folders remain empty or drain steadily, indicating that buffered data is successfully transferring to the PH database.
  3. Failover Archiving Test: In a test window, toggle mastership so OSS1 becomes the active master. In WinCC Information Server or WinCC Trend Control, verify that trend curves continue updating dynamically without gaps.

Limitations and Operational Precautions

  1. Historical Data Loss Risk: Deleting files in SFCache permanently deletes any unsent archive values buffered on that server during the disconnected period. If historical records are vital, copy the cache folder to a safe backup drive before purging.
  2. Software Version Parity: The Process Historian and all OS Servers must run identical update packages (e.g., PH 2014 SP3 Upd6 on the PH requires PH-Ready 2014 SP3 Upd6 on both OS servers).

Engineer FAQ

Why did Process Historian connect to only one OS server?

The failing server had a desynchronized ProcessHistorian.cfg binding token, a locked Store and Forward (S&F) MSMQ cache, or a hostname resolution failure.

Where is the Store and Forward cache stored?

In C:\ProgramData\Siemens\SFCache on the local OS Server.

What happens if the master OS switches while PH is disconnected from that server?

Process Historian stops receiving all new archive tags and alarm messages until the connection is restored or the master switches back.

What service manages the connection on the OS server?

The CCAPHServer service (Siemens Process Historian Ready Component).

Can I fix this without stopping the operating plant?

Yes, by performing the cache reset sequentially on one server at a time while the redundant partner maintains plant supervision.

Why is NTP time synchronization critical for Process Historian?

The PH database enforces strict time validation; timestamps drifting by more than 1 second from the server clock are rejected as implausible.

What does the 'Update' button in PH Management do?

It forces the Process Historian database to query active network nodes and refresh its plant structure tree.

Does deleting SFCache delete the historical database on the PH?

No. It purges only the temporary local buffer on the OS server; the central PH database is unaffected.

What network protocol does PH use for data transfer?

Microsoft Message Queuing (MSMQ) over TCP port 1801, RPC port 135, and dedicated SQL/PH ports.

Where are Process Historian error logs located?

On the OS Server inside C:\Program Files (x86)\Siemens\WinCC\Diagnostics\ and in the Windows Event Viewer under Application Logs.

Technical References

Related Articles

Related Tags

  • Process Historian
  • WinCC Redundancy
  • PH-Ready
  • Store and Forward
  • SFCache
  • MSMQ
  • OS Server
20 respuestas