32bit client can\'t connect to redundant 64bit servers

[n_lev]

Hi!

We have a problem tat the moment and I was wondering if there is something I am missing.

We've got a redundant pair of WinCC OA servers - Windows Server 2012 64 bit, WinCC OA 3.11 SP1
We also have 3 clients - WinXP SP3 32bit, WinCC OA 3.11 SP1
We also have a windows 7 64bit client with he same WinCC OA

When we try to run the 64bit client, it works fine
But if we try to run any of the 32bit clients, they fail with the message like
1:51:23.484, SYS, FATAL, 63, Connection error, This manager is already connected or Data is running in emergency / backup / recovery mode or the active data is unreachable.

If I reconfigure the 32bit client to only run with one server, rather than redundant pair, it works fine, but it fails to work with a redundant pair

Am I missing something obvious?

[leoknipp]

Hello,

have you maybe defined a fixed number for the user-interface?
Also please have a look at the logfiles at the servers, maybe there you can find useful information.

Best Regards
Leopold Knipp
Senior Support Specialist

[n_lev]

Hi Leopold,

Nope, using dynamic ui number. And I've checked that the number hasn't been used

Server log just says that client connected and then that it disconnected about 100 ms later. No warning or errors

[fmulder]

Quite often, with these kind of effects, the problem was in the name resolution. Or, make sure that all the right names (take care of upper- and lowercase) are mentioned in the 'hosts' file.
Also note that a 'ping' is no proof that the names are configured correctly. The ping command will eventually do a network broadcast and will eventually find any computer 9even without an entry in the hosts file).

Just check and double-check all the entries in the hosts file

Frenk

[n_lev]

Thanks for your help, guys, problem solved now. As it turns out, when those clients were configured, someone accidentally swapped server1 and server2 I'm the "data" and "event" in the config file. Once I swapped them back, all working.

[fmulder]

Let this be a lesson.

When you seem to have problems here clients or servers connect and disconnect, then I found that in 99% of the cases, this was due to a simple flaw in the names in the config file or the name resolution of your network (hosts file )

Glad that you found the problem

[chris@blue-catalyst.co.uk]

Hi,

I have the same error message but different scenario.

We have a dual redundant servers (Win Server 2012) and 7 clients (Win7), all running version 3.12.

On each client we have a remote UI project into which we need to add a control script. The UI connects to the server fine but the control script wants to connect to localHost.

When I add the -data and -event (with either the host name of server 1 or using the IP of server 1) I get the error:

'Connection error, This manager is already connected or Data is running in emergency / backup / recovery mode or the active data is unreachable'

I have tried fixing the control manager number but had the same result. When I check the server I can see that it connected without any problems, it was assigned a spare manager number, but disconnected 100ms later.

Any thoughts?

Thanks,

Chris


Host file:
192.168.1.1 COM-SER-001 COM-SER-001-1 # SERVER 1
192.168.1.2 COM-SER-002 COM-SER-002-1 # SERVER 2
192.168.10.3 COM-SER-001-2 # SERVER 1
192.168.10.4 COM-SER-002-2 # SERVER 2
192.168.1.12 COM-WKS-003 # WORKSTATION 3

Config file:
[general]
data = "COM-SER-001-1,COM-SER-001-2$COM-SER-002-1,COM-SER-002-2"
event = "COM-SER-001-1,COM-SER-001-2$COM-SER-002-1,COM-SER-002-2"

Ctrl manager options:
AudioAlarm.ctl -data COM-SER-001 -event COM-SER-001

Server Log:
WCCILproxy (1), 2014.07.23 12:44:03.991, SYS, INFO, 222, Accept SSL connection from host ::ffff:192.168.1.12:49300.
WCCILdata (0), 2014.07.23 12:44:04.090, SYS, INFO, 4, Connected to, (SYS: 1 Ctrl -num 3 CONN: 1) @ COM-SER-001 (::1)
WCCILproxy (1), 2014.07.23 12:44:04.090, SYS, INFO, 210, The data connection from client ((SYS: 0 Ctrl -num 0 CONN: 1)) to server ((SYS: 0 Data -num 0 CONN: 1)@COM-SER-001:4897) is established.
WCCILdata (0), 2014.07.23 12:44:06.094, SYS, INFO, 39, Connection lost, MAN: (SYS: 1 Ctrl -num 3 CONN: 1), Connection closed

[chris@blue-catalyst.co.uk]

Ok,

So i managed to sort this out and thought I'd post in case anyone else looks at this in the future.

Our development platform will be setup to sync to a timer server when on site however as its not on site its not all sync'd up and the time difference between server and client was causing the problem.

I manually adjusted the clock and it started to work.

I'm still unsure why the control manager doesn't look at the config file for data and event though.

Chris

[Gertjan van Schijndel]

@Chris:
You should add the '-connectToRedundantHosts' option to the control manager or the following to the config file for connecting to redundant servers:

Code: Select all

[ctrl]
connectToRedundantHosts = 1