Name |
Type |
Default |
Range |
Sections |
Comment |
alivePort |
uint |
0 |
0 - 64000 |
All sections except
general |
Defines the port
on which a manager wants to receive alive messages. The alive
check mechanism is only used between managers which define an
alivePort and also have an aliveTimeout
!= 0. The alive check is done on the receiver side only
- the one which defines the alivePort.
See also aliveTimeout. |
alivePriorityClass |
bool |
0 |
0|1 |
All sections except
general |
Increases the priority
of the alive thread under Windows to TIME_CRITICAL, under Linux
and Solaris to maximum "Scheduling priority". |
aliveTimeout |
uint |
-10 |
MIN_INT - MAX_INT
|
All sections |
The
entry defines at which intervals (in seconds) the alive telegrams
of this manager must be received, when
the target manager provides an alive port. The manager itself
sends about 10 alive telegrams over this period.
If the target
manager does not receive any telegrams over this period, the target
manager closes the connection.
The value 0
deactivates the sending of alive telegrams and the monitoring
of the target manager.
Negative values
are only valid for connections between managers that run on different
computers. Positive values
are valid for all manager connections. The target of alive managers
can be any manager to which the alive manager builds a connection
and which provides an alive port. Normally this is the Event and
possibly the Data Manager. For Dist Manager also the Dist Managers
of the coupled systems. The receiving of the respective manager
can be deactivated by using the config entry "alivePort
= 0". |
allowLocalMessageCompression |
bool |
0 |
0|1 |
All sections. |
The config entry
allowLocalMessageCompression = 1 can be used to activate the message
compression on the local computer. You can set the config entry,
for example, when using the Web Client
. |
messageCompressionThreshold |
int |
0 |
0 - MAX_INT |
All sections |
This entry specifies
the threshold in bytes below which messages are not compressed.
"0" means "compress all".
Message compression always comes with
a penalty of CPU usage. Sometimes it can be more effective to
transmit small messages uncompressed because the throughput in
the network would not change. A good estimate for this entry is
the MTU, i.e. the max number of bytes which can be transmitted
in one TCP packet. The MTU depends on the transmission used and
can be as low as a few 100 bytes in case of PPP up to approx.
1200 bytes in case of Ethernet. |
connectDp |
string |
"" |
- |
All sections except
general |
Defines the data
point, which is used by a manager to notify its manager connections. |
ctrlBreakFunctionCall |
bool |
Event:
0, all other managers: 1 |
0|1 |
All sections |
Defines if nested
function calls shall be processed in one go (or until a waiting
function call is reached) (= 0) or if the script can also be interrupted
during the function code execution (= 1). |
ctrlDLL |
string |
"" |
- |
ui, ctrl |
This config entry
loads a CtrlExtension DLL / SO. First, the DLL is searched in
<proj_path>/bin. If it does not exist there, it will be
searched in the <wincc_oa_path>/bin.
If a DLL / SO should be used by the UI
and CTRL, this must be loaded in both sections. The number of
DLLs / SOs to be load is not limited.
The file extension of the DLL / SO must
not be stated, it will be determined automatically.
Note
It is recommended to load a CtrlExtention DLL / SO directly
from the Ctrl script using the #uses statement. |
dbg |
string |
"" |
- |
All sections |
Allows to activate
a dbg Flag for a Manager which can not get -dbg as a commandline
argument (mainly Pmon as Windows service) e.g.
dbg = "2,16,SOME_FLAG"
|
distSystemIds |
string |
"" |
- |
All sections except
general |
With
this config entry you can configure per manager from which distributed
system it accepts/holds the DP identification.
Example
[ctrl_3]
distSystemIds
= "1-28, 46, 280-290, 320"
Either ranges
or single numbers can be entered. Blanks are ignored.
If this config
entry is not defined, all DP identifications are accepted.
If the value
of the distributed config
entry (see Configuration
file for distributed systems) for a manager is 0, the DP identification
is not transmitted to this manager.
With the debug
level -dbg 31 you can check which identifications are accepted
or ignored. |
exitDelay |
uint |
0 |
>= 0 |
All sections |
Defines
how many seconds a manager has to wait before final exit (in this
case a log message will be returned).
If a manager
is closed by a signal, this delay should not be used (e.g. stopping
by console/pmon), otherwise it will be always used. |
ip_allow |
string |
"" |
- |
general,
data, event, ctrl, pmon, valarch, dist |
Defines IP access
lists (clients), which may access the WinCC OA
managers (TCP sockets). For further information see also IP
access lists for TCP server sockets. |
ip_deny |
string |
"" |
- |
general,
data, event, ctrl, pmon, valarch, dist |
Defines IP access
lists (clients), which may access the WinCC OA
managers (TCP sockets). For further information see also IP
access lists for TCP server sockets. |
LoadAllCtrlLibs |
bool |
0 |
0|1 |
all drivers, ctrl, ui, event |
0: Load only explicitly
defined Ctrl libraries (see LoadCtrlLibs).
1: Load all files from the scripts/libs
directory.
This keyword means that the entries LoadCtrlLibs
and UnloadCtrlLibs have no effect.
Scripts that are empty or begin with "." will not be loaded! |
LoadCtrlLibs |
string |
"" |
- |
all drivers, ctrl, ui, event |
Loads all specified
libraries (files names are separated by commas).
This entry can be used as many times as
required.
Example:
loadCtrlLibs
= "std.ctl, libCTRL.ctl, hosts.ctl, va.ctl, archiv.ctl"
A library loaded by this entry can be
removed by a subsequent UnloadCtrlLibs entry. |
UnloadCtrlLibs |
string |
"" |
- |
all drivers, ctrl, ui, event |
Does NOT load all
specified libraries (library names are separated by commas).
The keyword can be used as many times
as required.
Example:
UnloadCtrlLibs
= "std.ctl, libCTRL.ctl, hosts.ctl, va.ctl, archiv.ctl" |
maxDpNamesCount |
uint |
1000000 |
>= 0 |
All sections |
Specifies how many
DPEs the dpNames() or equivalent functions (getIdSet and similar
functions) in API may return. If the number of DPEs is exceeded
the functions do not return anything (dpNames()) or an error.
The value 0 allows an arbitrary number
of values.
Note
that this number may also be exceeded, if for a DP the
requested search object (DPE, alias, comment, etc) does not exist,
but, however, the DP matches the filter criterion. For example,
the number of DPEs (that match the filter criterion) in the function
dpGetAllAliases()
is compared with the number set in maxDpNamesCount, even if not
all DPEs have an alias. |
priorityClass |
int |
0 |
Linux:
-20
- 19 Windows:
-1 - 2 |
valarch,
ui, data, event, ctrl, redu, dist, ascii, oledb, opc |
Controls the priority
of managers under Windows and Linux.
Windows:
-1 - lower than normal
0 - Priority normal
(Default)
1 - higher than
normal
2 - Priority high
Linux:
1-19 - Priority
increase (the Manager has to be a root process)
n = 0 - none (Default)
-20--1
- Priority decrease |
queryRDBdirect |
bool |
0 |
0|1 |
ui, ctrl |
Indicates the mode
of database read queries:
0 = Standard CTRL read functions
(dpGet...()) via the WinCC OA Event Manager and the WinCC
OA Data Manager. 1 = The standard CTRL read functions
(dpGet...()) are redirected to directly contact the database
server.
Note that the two required CTRL extensions
must also be loaded in order to be able to use the RDB read functions:
CtrlDLL
= "CtrlRDBArchive" CtrlDLL = "CtrlRDBCompr"
CAUTION: If you use queryRDBdirect
= 1 and query DPEs using
the event screen, enter the DPEs or *.* for the query.
Otherwise the event screen does
not show the DPEs. |
requestDejaVu |
bool |
0 |
0|1 |
All sections except
general |
In
a redundant system value changes are processes by the active Event manager and are forwarded to
the passive Event manager.
The passive
Event manager rejects the value
changes, because it expects to receive them from the active Event manager.
As it can take
10 seconds until the passive recognizes the break-down of the
active Event manager, it may
happen that the passive Event
manager rejected value changes (where it assumes to get them from
the other Event manager) within
the 10 seconds.
DejaVu delays
the rejection of value changes. By default the Event
manager does it only for drivers (except OPC UA). For other managers
this behavior can be configured with the "requestDejaVu"
config entry. |
singleSourceConnect |
int |
0 |
0|1 |
All sections |
The
source IP addresses have to be different for redundant network
connections. The second connect is closed if it has the same source
IP address as the existing connection . If a computer establishes
a connection to a redundant network via a network interface card,
both connections have the same source IP address. In this case
the config entry singleSourceConnect has to be set to 1. |
smoothBit |
string |
"" |
"Userbit 1" - "Userbit
32" |
All driver sections |
The
entry specifies which user bits are taken into account for a smoothing
(low level old/new comparison or configured old/new comparison).
If one of these user bits (or the invalid bit) is changed, the
old/new comparison is executed although the original values would
not have changed. The entry has to be specified for each user
bit, which should be taken into account. Possible values for the
user bits are "Userbit 1",
"Userbit 2", -., "Userbit 32". |
