Attempt failover if all opc items are bad quality, What triggers a failover – Rockwell Automation 1757-PROCESSLINX ProcessLinx OPC Server to Server Software User Manual User Manual

Page 27

Advertising
background image

Publication 1757-UM002B-EN-P - January 2003

ProcessLinx TagMap Configuration 3-7

Attempt Failover if all OPC Items are Bad Quality

This checkbox will enable or disable failover when all OPC Items are
bad Quality. This option is specifically used with ProcessLogix
redundant Server scheme but can be used with any OPC Server. The
redundant (backup) OPC Server of ProcessLogix will set all OPC Items
to 'Out-of-Service' BAD Quality when the node becomes the backup
and thus this can be used as a trigger for ProcessLinx to failover to the
redundant server. Failover for other reasons (connection time-outs,
server quit) will still automatically occur if the secondary server is
defined.

Successful Failover from one OPC90 Server to Another

A successful failover occurs when the active connected server
[Primary/Secondary] 'fails' and ProcessLinx then connects to the
configured redundant one and after which the redundant one remains
'Connected' and thus is not 'failed'. 'Server Failure' is defined below.

What Triggers a Failover

'Connected to' means that ProcessLinx can establish a link to the
configured server's OPCServer interface, add groups, add items and
have data callbacks setup. In runtime, status of the server is checked at
the periodic rate specified in the Status Validation edit box for the server
in the Pipe dialog box. The GetStatus( ) function on the OPCServer
interface is used to provide an initial decision as to whether processLinx
is connected to the server. The following are conditions that move
ProcessLinx to implement a failover (other than the OPC Server
failing).

• If the GetStatus( ) or Write( ) functions fails 5 times (meaning the

functions return a server level error or Win32 error of
RPC_C_SERVER_UNAVAILABLE) or the server status is returned as
(FAILED, NOCONFIG or SUSPENDED) then the server is marked
as 'failed' and if a redundant server is configured then ProcessLinx will
attempt to connect to that.

• If the 'Attempt failover of all OPC Items are Bad Quality' is checked in

the Pipe dialog box, then all OPC Items of every group for that server
are checked as well at the periodic rate specified in the Status Validation
edit box in the Pipe dialog box.

• If all active tags in any active group in that server are BAD Quality and

the sub-status quality is either CONFIGURATION_ERROR or
OUT_OF_SERVICE then that server is marked as 'Failed'.

Advertising