Scribd
Upload
52 views

PH-OS Connection

The document describes troubleshooting steps for a Process Historian (PH) that was unable to detect one of two OS servers (OSS1A and OSS1B) after installing PH/IS. The recommended procedure includes pre-check steps to verify networking, configuration, and permissions. It also provides steps to back up and extend the duration of OS server archives, reset the PH database, and remove PH/SFCache to facilitate synchronization between the PH and OS servers.

Uploaded by

Akash Gor
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
52 views

PH-OS Connection

The document describes troubleshooting steps for a Process Historian (PH) that was unable to detect one of two OS servers (OSS1A and OSS1B) after installing PH/IS. The recommended procedure includes pre-check steps to verify networking, configuration, and permissions. It also provides steps to back up and extend the duration of OS server archives, reset the PH database, and remove PH/SFCache to facilitate synchronization between the PH and OS servers.

Uploaded by

Akash Gor
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Link : https://support.industry.siemens.

com/forum/qa/en/posts/process-historian-ph-os-connection-
troubleshooting/269102

A new PH case study here, and i would like to share here for future reference. 

Scenario

There are one set of OS server (OSS1A and OSS1B) and one PH standalone (Dell EMC). 

Before Changing PH to PH/IS , everything was healthy. After installed PH/IS into the PH , the system went
into problem. Then, multiple activities have been executed such as reinstall windows, restore backup ,
recreate Historian database, but the end result was , the plant structure in PH multiproject state can only
detect single OS , which is OSS1B. PH Server state on client is green in color.  

The database can be reset again as now is commissioning phase. 

OS = Windows Server 2016

PCS 7 = PCS 7 V9.0 SP x . 

Recommended procedure to troubleshoot the event.

/***Pre-Check***/
1. ensure Microsoft Message Queuing is installed on the PH server.
2. ensure the user has adminstrator right. ( authorized to write data in the database, and the user should be used as
a PH ready User(check via services.msc) on the OS.
3. Ping OS server 1 and OS Server 2 from PH server via "COMPUTER NAME" or "HOST NAME". For example, Ping
PCS7OSS01.
4. In the ES, ensure under OS Server ( WinCC application ) OS object properties , "transfer to external archive
server" is checked.
5. Ensure ProcessHistorian.cfg available in OS Server.
6. Ensure Installed Software is PH 2014 SP3 Up6 in PH and OS Servers (PH Ready). If update is needed , read the
PH readme in the update package. https://support.industry.siemens.com/cs/ww/en/view/109780528
7. Ensure the system is having proper security right configured. Goto START -> Siemens Automation->
Security Controller -> repeat settings. (to be executed in all PCS 7 stations include PH)

8. SIMATIC HMI right must be given and PHready must be properly configured. 

9. Ensure time synchronization are done. The time in PH must be the same with OS Server. NTP time
synchronization for OS server - PH must be done!

/***Backup OS Server Archives and Extend the duration***/


1. Deactivate the wincc runtimes on both OS Servers
2. Please note theat WinCC "remembers" which segments has been created, need to be attached, need to be
synchronized. To remove such information, one need to perform a reset archives on both servers. But note: this will
delete the archive information! So , if really want to execute this actions, create a backups , and consider the
consequences.
3. to backup, at the OS Server, go to OS folder , look for Archive Manager
2. copy all the archives to a backup location. (to restore, you may put it back to "Common Archiving. But please note
that archive data stored here will never be removed automatically. But it will count into the overall number of
segments.
and let the system synchronize for you or add backup via trend control view (Online Trend Control -> configuration
dialog -> toolbar -> add backup)
3. Do the same things at the OS Server Standby.

/***At the engineering station***/


1. Go to OS Server wincc explorer
2. at Tag logging fast and slow, extend the maximum segment to longer period . For example, from 2 months to 6
months.
3. at alarm logging, IF the message is important to you, extend the maximum segment to longer period. For example,
from 2 months to 6 months. Please ensure you have sufficient D drive space.
4. Download both OS Servers.
(IF you want to save time, you may repeat the above steps in local OS Servers.But i recommend that you do a full
OS download, as there are some unknown synchronization for OS servers might not finished yet)

_______________________________________________
BEGIN Here ->
Full steps for PH reset.
1. use the database wizard to recreate the database. Configure them properly based on your hardware size. MUST
separate multiple disc for different purpose, such as Transaction log, tag, alarm, etc.
2. follow the PH manual

SFCache / PHCache Removal


To facilitate the sync between the PH and OS , go through the steps below
1. Deactivate the runtime on the standby OS.
2. Rename the “Processhistorian.cfg” in the standby OS to “Processhistorian.cfg.old” (In WinCC project folder of the
OS server)
3. Restart the standby OS.
4. Activate the runtime (on the standby OS).
5. Wait for the redundancy check to be completed.
6. Deactivate the runtime on the master OS.
7. Rename the “Processhistorian.cfg” in the master OS to “Processhistorian.cfg.old” (In WinCC project folder of the
OS server)
8. Restart the master OS.
9. Activate the Runtime (on the master OS).
10. Copy SFCache folder to D drive before delete as a backup
11. Location of PHReady Cache and SFCache are same “C:\ProgramData\Siemens\SFCache”. PHReady Cache
means like a folder name that end with Queue. For instance, AlgCSDataQueue. Delete all from this folder
12. Delete the PH Ready Cache on both OS Servers. Delete the S&F Cache on both OS Servers.
13. Rename the “Processhistorian.cfg.old” on the standby OS back to “Processhistorian.cfg”.
14. Rename the “Processhistorian.cfg.old” on the master OS back to “Processhistorian.cfg”.
15. Deactivate the runtime on the standby OS.
16. Activate the runtime on the standby OS.
17. Wait for the redundancy check to be completed.
18. Deactivate the runtime on the master OS.
19. Activate the Runtime on the master OS.
20. Wait for the S&F recovery to be completed.

EXTRA INFORMATION
________________
/***Example of Messages ***/
//General rule: Number of messages/s * 4000 bytes * 60 s/min * 60 min/h * 24 h/day * 31 days/month * y months.
// A rate of 1 message/sec is assumed in the example:

//The approximate maximum size for all segments across a period of 2 months is derived as follows:
//20 GB (1 message/s * 4000 bytes / message * 60 s/min * 60 min/h * 24 h/day * 31 days/month * 2 month)

//The resultant value for each segment is a daily rate of:


//approximately 330 MB (1 message/s * 4000 bytes / message * 60 s/min * 60 min/h * 24 h/day * 1 day)

_________________
/***Example of TagLogging Fast ***/
//You are planning on implementing a period of 2 months. You have calculated a mean rate of 750 archive values/s
for TagLogging Fast.

//The maximum size for all segment is derived as follows:

//For a memory requirement of 16 bytes per process value:


//ca. 60 GB (750 [archive values/s] * 16 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 30 [days/month] * 2
[months])

//For a memory requirement of 6 bytes per process value:


//ca. 22 GB (750 [archive values/s] * 6 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 30 [days/month] * 2
[months])

//Daily value per segment:

//For a memory requirement of 16 bytes per process value:


//ca. 1 GB (750 [archive values/s] * 16 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 1 [day])

//For a memory requirement of 6 bytes per process value:


//ca. 370 MB (750 [archive values/s] * 6 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 1 [day])

//Configure your message archives accordingly for the "TagLogging Fast" archive configuration.
/***Example of TagLogging Slow ***/

You are planning on implementing a period of 2 months. You have calculated a mean rate of 100 archive values/s for
TagLogging Slow.

The maximum size for all segment is derived as follows:

For a memory requirement of 16 bytes per process value:


ca. 8 GB (100 [archive values/s] * 16 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 30 [days/month] * 2
[months])

For a memory requirement of 6 bytes per process value:


ca. 3 GB (100 [archive values/s] * 6 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 30 [days/month] * 2 [months])

Daily value per segment:

For a memory requirement of 16 bytes per process value:


ca. 130 MB (100 [archive values/s] * 16 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 1 [day])

For a memory requirement of 6 bytes per process value:


ca. 50 MB (100 [archive values/s] * 6 [bytes/value] * 60 [s/min] * 60 [min/h] * 24 [h/day] * 1 [day])

Configure your message archives accordingly for the "TagLogging Slow" archive configuration.

Thank you. Hope this is helpful for anyone who face similar PH issue in the future. 

Please give me 5 stars Rating if my post is helpful for you. :)


Thank you for Rating!

PCS 7 Engineer
Jaz Tai , Excel Marco Singapore
PCS 7 Complete Overview
PCS 7 Technical Documentation

You might also like