Hello everybody,
well, I would like to have some comments from the guys that have already used the WinCC OA or PVSS in redundant system.
I configured my system in WinCC OA 3.10 with redundant servers. Everything works fine but there are one specific situation that I don't understand the behaviour.
In the page where I have the access to the information of the system everything seems ok and when I force a switchover in the system using the button "Force active" the system comutes as I expected; if the server 1 is as primary and I click in "Force active" in the second the second server assumes the system ans the server 1 stays as standby or secondary.
The same behaviour occurs if, for example, I force a fault in modbus driver; after the switchover when I resarted the modbus driver the system comutes again and the server 1 is primary and server 2 is secondary because I selected ther server 1 as "Preferred".
In both situations the system is very fast comuting and retourning to the normal state.
But there are this situation where the system has a different behaviour. The situation is when I have the server 1 running as primary and the server 2 running as secondary and isolate the server 1 from the network, removing the ethernet cables form the two ethernet ports present in the server, the system switchover as I expected and everything seems ok; when I introduce the ethernet cables again in ther server 1, the system switchover again, because I marked the server 1 as "Preferred", the server 1 takes the control but in the server 2 all the WinCC OA processes stop (only the
Process Monitor - first one - stays with the state 2 and the Database Manager - second process of the list - stays with the state 1. The server 2 begins synchronizing the database with the server 1 and when it ends the synchronization the WinCC OA starts normaly (it takes 2 or 3 minutes to goes up).
This only occurs in this specific situation.
I don't know if this is the expected behaviour ... so I'm trying to confirm with somebody that developed some similar project. I hope someone answer me.
Best regards ....
Steve Inácio
Cegelec Portugal
Redundants servers + WinCC OA expected behaviour
Search
-
- Posts: 4
- Joined: Thu Dec 15, 2011 12:14 pm
Re: Redundants servers + WinCC OA expected behaviour
Hello,
if the connection between the redundant servers is lost and reestablished again the WinCC OA-project at one server is stopped and restarted to make the recovery.
This is the correct behaviour of a redundant WinCC OA system.
Please notice that the ETM Portal does not substitute the common WinCC OA support. For questions/problems related to WinCC OA please get in contact with the software vendor for your project.
Best Regards
Leopold Knipp
Senior Support Specialist
if the connection between the redundant servers is lost and reestablished again the WinCC OA-project at one server is stopped and restarted to make the recovery.
This is the correct behaviour of a redundant WinCC OA system.
Please notice that the ETM Portal does not substitute the common WinCC OA support. For questions/problems related to WinCC OA please get in contact with the software vendor for your project.
Best Regards
Leopold Knipp
Senior Support Specialist
-
- Posts: 4
- Joined: Thu Dec 15, 2011 12:14 pm
Re: Redundants servers + WinCC OA expected behaviour
Hello Leopold,
thank you for your answer.
Firstly I saw it in that way, the system recovery and return to a redundant system and should be normal to stop and restart to make the recovery, sincronize the databases but when I make this fault several times I noticed something that could be a problem.
Assuming the server 1 is primary and my preferred and the hole system is running in redundant system, if I isolate the server 1 from the network removing the ethernet cable from the server ethernet ports, the server 2 assumes the control. However, when I reconnect the server 1 to the network, and correct me if I'm wrong, should be the server 1 that should stop and restart after synchronize his database with the server 2 (this server is the one that is collecting the data from the PLC in the network) and not the oposite. My problem is that I'm not sure if I'm losing data.
I can contact WinCC OA support but when I went to Viena to make the training, Markus told us to share our questions in the portal to get a fast answer and to share the problems with the Win CC OA community. It's a problem or a question for me today but could be to a coleage tomorow, right?
Regards
thank you for your answer.
Firstly I saw it in that way, the system recovery and return to a redundant system and should be normal to stop and restart to make the recovery, sincronize the databases but when I make this fault several times I noticed something that could be a problem.
Assuming the server 1 is primary and my preferred and the hole system is running in redundant system, if I isolate the server 1 from the network removing the ethernet cable from the server ethernet ports, the server 2 assumes the control. However, when I reconnect the server 1 to the network, and correct me if I'm wrong, should be the server 1 that should stop and restart after synchronize his database with the server 2 (this server is the one that is collecting the data from the PLC in the network) and not the oposite. My problem is that I'm not sure if I'm losing data.
I can contact WinCC OA support but when I went to Viena to make the training, Markus told us to share our questions in the portal to get a fast answer and to share the problems with the Win CC OA community. It's a problem or a question for me today but could be to a coleage tomorow, right?
Regards
Re: Redundants servers + WinCC OA expected behaviour
Hello,
if the system are reconnected one system is stopped. Which one will be stopped is not defined.
With the config-entry useOfflineErrorstateInfo you can activate additional functionality for the redundant system. If this config-entry is set to 1 the system which had the better state while the sysdtem were disconnected will remain running.
For details please have a look at the WinCC OA online help.
For questions related to WinCC OA you should use the common WinCC OA support instead of the portal if you have a valid contract with ETM. ETM uses the worldwide Siemens Support as your first contact. After adding your service request into the support database the service request is forwarded to ETM and we always have a look at the issues forwarded to ETM.
For the common support there are analysis times defined, see the entry "Support & Services" --> "Customer Support" at our homepage.
At the portal we have a look time-by-time, therefore it can take several hours or days before an answer is posted by ETM.
Best Regards
Leopold Knipp
Senior Support Specialist
if the system are reconnected one system is stopped. Which one will be stopped is not defined.
With the config-entry useOfflineErrorstateInfo you can activate additional functionality for the redundant system. If this config-entry is set to 1 the system which had the better state while the sysdtem were disconnected will remain running.
For details please have a look at the WinCC OA online help.
For questions related to WinCC OA you should use the common WinCC OA support instead of the portal if you have a valid contract with ETM. ETM uses the worldwide Siemens Support as your first contact. After adding your service request into the support database the service request is forwarded to ETM and we always have a look at the issues forwarded to ETM.
For the common support there are analysis times defined, see the entry "Support & Services" --> "Customer Support" at our homepage.
At the portal we have a look time-by-time, therefore it can take several hours or days before an answer is posted by ETM.
Best Regards
Leopold Knipp
Senior Support Specialist