Files
unison/doc/manual/tracing.texi
Craig Dowell adb00d72d5 manual update
2010-01-28 17:34:20 -08:00

1175 lines
50 KiB
Plaintext

@node Tracing
@chapter Tracing
The tracing subsystem is one of the most important mechanisms to understand in
@command{ns-3}. In most cases, @command{ns-3} users will have a brilliant idea
for some new and improved networking feature. In order to verify that this
idea works, the researcher will make changes to an existing system and then run
experiments to see how the new feature behaves by gathering statistics
that capture the behavior of the feature.
In other words, the whole point of running a simulation is to generate output for
further study. In @command{ns-3}, the subsystem that enables a researcher to do
this is the tracing subsystem.
@menu
* Tracing Motivation::
* Overview::
* Using the Tracing API::
* Using the Tracing Helpers::
* Tracing implementation details::
@end menu
@node Tracing Motivation
@section Motivation
There are many ways to get information out of a program. The most straightforward
way is to just directly print the information to the standard output, as in,
@verbatim
#include <iostream>
...
int main ()
{
...
std::cout << ``The value of x is `` << x << std::endl;
...
}
@end verbatim
This is workable in small environments, but as your simulations get more and more
complicated, you end up with more and more prints and the task of parsing and
performing computations on the output begins to get harder and harder.
Another thing to consider is that every time a new tidbit is needed, the software
core must be edited and another print introduced. There is no standardized way
to control all of this output, so the amount of output tends to grow without
bounds. Eventually, the bandwidth required for simply outputting this information
begins to limit the running time of the simulation. The output files grow to
enormous sizes and parsing them becomes a problem.
@command{ns-3} provides a simple mechanism for logging and providing some control
over output via @emph{Log Components}, but the level of control is not very fine
grained at all. The logging module is a relatively blunt instrument.
It is desirable to have a facility that allows one to reach into the core system
and only get the information required without having to change and recompile the
core system. Even better would be a system that notified the user when an item
of interest changed or an interesting event happened.
The @command{ns-3} tracing system is designed to work along those lines and is
well-integrated with the Attribute and Config substems allowing for relatively
simple use scenarios.
@node Overview
@section Overview
The tracing subsystem relies heavily on the @code{ns-3} Callback and Attribute
mechanisms. You should read and understand the corresponding sections of the
manual before attempting to understand the tracing system.
The ns-3 tracing system is built on the concepts of independent tracing sources
and tracing sinks; along with a uniform mechanism for connecting sources to sinks.
Trace sources are entities that can signal events that happen in a simulation and
provide access to interesting underlying data. For example, a trace source could
indicate when a packet is received by a net device and provide access to the
packet contents for interested trace sinks. A trace source might also indicate
when an interesting state change happens in a model. For example, the congestion
window of a TCP model is a prime candidate for a trace source.
Trace sources are not useful by themselves; they must be connected to other pieces
of code that actually do something useful with the information provided by the source.
The entities that consume trace information are called trace sinks. Trace sources
are generators of events and trace sinks are consumers.
This explicit division allows for large numbers of trace sources to be scattered
around the system in places which model authors believe might be useful. Unless
a user connects a trace sink to one of these sources, nothing is output. This
arrangement allows relatively unsophisticated users to attach new types of sinks
to existing tracing sources, without requiring editing and recompiling the core
or models of the simulator.
There can be zero or more consumers of trace events generated by a trace source.
One can think of a trace source as a kind of point-to-multipoint information link.
The ``transport protocol'' for this conceptual point-to-multipoint link is an
@code{ns-3} @code{Callback}.
Recall from the Callback Section that callback facility is a way to allow two
modules in the system to communicate via function calls while at the same time
decoupling the calling function from the called class completely. This is the
same requirement as outlined above for the tracing system.
Basically, a trace source @emph{is} a callback to which multiple
functions may be registered. When a trace sink expresses
interest in receiving trace events, it adds a callback to a list of callbacks
held by the trace source. When an interesting event happens, the trace source
invokes its @code{operator()} providing zero or more parameters. This tells
the source to go through its list of callbacks invoking each one in turn. In
this way, the parameter(s) are communicated to the trace sinks, which are just
functions.
@subsection The Simplest Example
It will be useful to go walk a quick example just to reinforce what we've said.
@smallformat
@example
#include ``ns3/object.h''
#include ``ns3/uinteger.h''
#include ``ns3/traced-value.h''
#include ``ns3/trace-source-accessor.h''
#include <iostream>
using namespace ns3;
@end example
@end smallformat
The first thing to do is include the required files. As mentioned above, the
trace system makes heavy use of the Object and Attribute systems. The first
two includes bring in the declarations for those systems. The file,
@code{traced-value.h} brings in the required declarations for tracing data
that obeys value semantics.
In general, value semantics just means that you can pass the object around,
not an address. In order to use value semantics at all you have to have an
object with an associated copy constructor and assignment operator
available. We extend the requirements to talk about the set of operators
that are pre-defined for plain-old-data (POD) types. Operator=, operator++,
operator--, operator+, operator==, etc.
What this all means is that you will be able to trace changes to an object
made using those operators.
@smallformat
@example
class MyObject : public Object
{
public:
static TypeId GetTypeId (void)
{
static TypeId tid = TypeId ("MyObject")
.SetParent (Object::GetTypeId ())
.AddConstructor<MyObject> ()
.AddTraceSource ("MyInteger",
"An integer value to trace.",
MakeTraceSourceAccessor (&MyObject::m_myInt))
;
return tid;
}
MyObject () {}
TracedValue<uint32_t> m_myInt;
};
@end example
@end smallformat
Since the tracing system is integrated with Attributes, and Attributes work
with Objects, there must be an @command{ns-3} @code{Object} for the trace source
to live in. The two important lines of code are the @code{.AddTraceSource} and
the @code{TracedValue} declaration.
The @code{.AddTraceSource} provides the ``hooks'' used for connecting the trace
source to the outside world. The @code{TracedValue} declaration provides the
infrastructure that overloads the operators mentioned above and drives the callback
process.
@smallformat
@example
void
IntTrace (Int oldValue, Int newValue)
{
std::cout << ``Traced `` << oldValue << `` to `` << newValue << std::endl;
}
@end example
@end smallformat
This is the definition of the trace sink. It corresponds directly to a callback
function. This function will be called whenever one of the operators of the
@code{TracedValue} is executed.
@smallformat
@example
int
main (int argc, char *argv[])
{
Ptr<MyObject> myObject = CreateObject<MyObject> ();
myObject->TraceConnectWithoutContext ("MyInteger", MakeCallback(&IntTrace));
myObject->m_myInt = 1234;
}
@end example
@end smallformat
In this snippet, the first thing that needs to be done is to create the object
in which the trace source lives.
The next step, the @code{TraceConnectWithoutContext}, forms the connection
between the trace source and the trace sink. Notice the @code{MakeCallback}
template function. Recall from the Callback section that this creates the
specialized functor responsible for providing the overloaded @code{operator()}
used to ``fire'' the callback. The overloaded operators (++, --, etc.) will
use this @code{operator()} to actually invoke the callback. The
@code{TraceConnectWithoutContext}, takes a string parameter that provides
the name of the Attribute assigned to the trace source. Let's ignore the bit
about context for now since it is not important yet.
Finally, the line,
@verbatim
myObject->m_myInt = 1234;
@end verbatim
should be interpreted as an invocation of @code{operator=} on the member
variable @code{m_myInt} with the integer @code{1234} passed as a parameter.
It turns out that this operator is defined (by @code{TracedValue}) to execute
a callback that returns void and takes two integer values as parameters --
an old value and a new value for the integer in question. That is exactly
the function signature for the callback function we provided -- @code{IntTrace}.
To summarize, a trace source is, in essence, a variable that holds a list of
callbacks. A trace sink is a function used as the target of a callback. The
Attribute and object type information systems are used to provide a way to
connect trace sources to trace sinks. The act of ``hitting'' a trace source
is executing an operator on the trace source which fires callbacks. This
results in the trace sink callbacks registering interest in the source being
called with the parameters provided by the source.
@subsection Using the Config Subsystem to Connect to Trace Sources
The @code{TraceConnectWithoutContext} call shown above in the simple example is
actually very rarely used in the system. More typically, the @code{Config}
subsystem is used to allow selecting a trace source in the system using what is
called a @emph{config path}.
For example, one might find something that looks like the following in the system
(taken from @code{examples/tcp-large-transfer.cc})
@smallformat
@example
void CwndTracer (uint32_t oldval, uint32_t newval) {}
...
Config::ConnectWithoutContext (
"/NodeList/0/$ns3::TcpL4Protocol/SocketList/0/CongestionWindow",
MakeCallback (&CwndTracer));
@end example
@end smallformat
This should look very familiar. It is the same thing as the previous example,
except that a static member function of class @code{Config} is being called instead
of a method on @code{Object}; and instead of an @code{Attribute} name, a path is
being provided.
The first thing to do is to read the path backward. The last segment of the path
must be an @code{Attribute} of an @code{Object}. In fact, if you had a pointer to
the @code{Object} that has the ``CongestionWindow'' @code{Attribute} handy (call it
@code{theObject}), you could write this just like the previous example:
@smallformat
@example
void CwndTracer (uint32_t oldval, uint32_t newval) {}
...
theObject->TraceConnectWithoutContext ("CongestionWindow", MakeCallback (&CwndTracer));
@end example
@end smallformat
It turns out that the code for @code{Config::ConnectWithoutContext} does exactly that.
This function takes a path that represents a chain of @code{Object} pointers and follows
them until it gets to the end of the path and interprets the last segment as an
@code{Attribute} on the last object. Let's walk through what happens.
The leading ``/'' character in the path refers to a so-called namespace. One of the
predefined namespaces in the config system is ``NodeList'' which is a list of all of
the nodes in the simulation. Items in the list are referred to by indices into the
list, so ``/NodeList/0'' refers to the zeroth node in the list of nodes created by
the simulation. This node is actually a @code{Ptr<Node>} and so is a subclass of
an @code{ns3::Object}.
As described in the Object Model section, @code{ns-3} supports an object aggregation
model. The next path segment begins with the ``$'' character which indicates a
@code{GetObject} call should be made looking for the type that follows. When a
node is initialized by an @code{InternetStackHelper} a number of interfaces are
aggregated to the node. One of these is the TCP level four protocol. The runtime
type of this protocol object is ``ns3::TcpL4Protocol''. When the @code{GetObject}
is executed, it returns a pointer to the object of this type.
The @code{TcpL4Protocol} class defines an Attribute called ``SocketList'' which is
a list of sockets. Each socket is actually an @code{ns3::Object} with its own
@code{Attributes}. The items in the list of sockets are referred to by index just
as in the NodeList, so ``SocketList/0'' refers to the zeroth socket in the list
of sockets on the zeroth node in the NodeList -- the first node constructed in the
simulation.
This socket, the type of which turns out to be an @code{ns3::TcpSocketImpl} defines
an attribute called ``CongestionWindow'' which is a @code{TracedValue<uint32_t>}.
The @code{Config::ConnectWithoutContext} now does a,
@smallformat
@example
object->TraceConnectWithoutContext ("CongestionWindow", MakeCallback (&CwndTracer));
@end example
@end smallformat
using the object pointer from ``SocketList/0'' which makes the connection between
the trace source defined in the socket to the callback -- @code{CwndTracer}.
Now, whenever a change is made to the @code{TracedValue<uint32_t>} representing the
congestion window in the TCP socket, the registered callback will be executed and
the function @code{CwndTracer} will be called printing out the old and new values
of the TCP congestion window.
@node Using the Tracing API
@section Using the Tracing API
There
are three levels of interaction with the tracing system:
@itemize @bullet
@item Beginning user can easily control which objects are participating in tracing;
@item Intermediate users can extend the tracing system to modify the output format
generated or use existing trace sources in different ways, without modifying the
core of the simulator;
@item Advanced users can modify the simulator core to add new tracing sources and
sinks.
@end itemize
@node Using the Tracing Helpers
@section Using the Tracing Helpers
The @code{ns-3} trace helpers provide a rich environment for configuring and
selecting different trace events and writing them to files. In previous
sections, primarily ``Building Topologies,'' we have seen several varieties
of the trace helper methods designed for use inside other (device) helpers.
Perhaps you will recall seeing some of these variations:
@verbatim
pointToPoint.EnablePcapAll ("second");
pointToPoint.EnablePcap ("second", p2pNodes.Get (0)->GetId (), 0);
csma.EnablePcap ("third", csmaDevices.Get (0), true);
pointToPoint.EnableAsciiAll (ascii.CreateFileStream ("myfirst.tr"));
@end verbatim
What may not be obvious, though, is that there is a consistent model for all of
the trace-related methods found in the system. We will now take a little time
and take a look at the ``big picture''.
There are currently two primary use cases of the tracing helpers in @code{ns-3}:
Device helpers and protocol helpers. Device helpers look at the problem
of specifying which traces should be enabled through a node, device pair. For
example, you may want to specify that pcap tracing should be enabled on a
particular device on a specific node. This follows from the @code{ns-3} device
conceptual model, and also the conceptual models of the various device helpers.
Following naturallyu from this, the files created follow a
<prefix>-<node>-<device> naming convention.
Protocol helpers look at the problem of specifying which traces should be
enabled through a protocol and interface pair. This follows from the @code{ns-3}
protocol stack conceptual model, and also the conceptual models of internet
stack helpers. Naturally, the trace files should follow a
<prefix>-<protocol>-<interface> naming convention.
The trace helpers therefore fall naturally into a two-dimensional taxonomy.
There are subtleties that prevent all four classes from behaving identically,
but we do strive to make them all work as similarly as possible; and whenever
possible there are analogs for all methods in all classes.
@verbatim
| pcap | ascii |
-----------------+------+-------|
Device Helper | | |
-----------------+------+-------|
Protocol Helper | | |
-----------------+------+-------|
@end verbatim
Let say at this point that @code{mixins} would probably be the best way to
approach the kind of class hierarchy found in @code{src/helper/trace-helper.h}
but our Python bindings generator does not support multiple inheritance and so
this choice is excluded for now. We were forced to linearize the inheritance
tree and use single inheritance for the trace functionality. This results in a
somewhat odd arrangement of classes, an unusual naming convention and implies
a hierarchy of functionality avialable in those classes. This isn't really a
concern for users of this functionality, but it does make our discussion here
a bit more difficult in some cases.
Let's take a quick look at all four of these cases.
@subsection Pcap Tracing Device Helpers
The goal of these helpers is to make it easy to add a consistent pcap trace
facility to an @code{ns-3} device. We want all of the various flavors of
pcap tracing to work the same across all devices, so the methods of these
helpers are inherited by device helpers. Take a look at
@code{src/helper/trace-helper.h} if you want to follow the discussion while
looking at real code.
The class @code{PcapHelperForDevice} provides the high level functionality
for using pcap tracing in an @code{ns-3} device. Every device must implement a
single virtual method inherited from this class.
@verbatim
virtual void EnablePcapInternal (std::string prefix, Ptr<NetDevice> nd, bool promiscuous) = 0;
@end verbatim
The signature of this method reflects the device-centric view of the situation
at this level. All of the public methods inherited from class
2@code{PcapUserHelperForDevice} reduce to calling this single device-dependent
implementation method. For example, the lowest level pcap method,
@verbatim
void EnablePcap (std::string prefix, Ptr<NetDevice> nd, bool promiscuous = false);
@verbatim
will call the device implementation of @code{EnablePcapInternal} directly. All
other public pcap tracing methods build on this implementation to provide
additional user-level functionality. What this means to the user is that all
device helpers in the system will have all of the pcap trace methods available;
and these methods will all work in the same way across devices if the device
implements @code{EnablePcapInternal} correctly.
@subsubsection Pcap Tracing Device Helper Methods
@verbatim
void EnablePcap (std::string prefix, Ptr<NetDevice> nd, bool promiscuous = false);
void EnablePcap (std::string prefix, std::string ndName, bool promiscuous = false);
void EnablePcap (std::string prefix, NetDeviceContainer d, bool promiscuous = false);
void EnablePcap (std::string prefix, NodeContainer n, bool promiscuous = false);
void EnablePcap (std::string prefix, uint32_t nodeid, uint32_t deviceid, bool promiscuous = false);
void EnablePcapAll (std::string prefix, bool promiscuous = false);
@end verbatim
You are encouraged to peruse the Doxygen for class @code{PcapHelperForDevice}
to find the details of these methods; but to summarize ...
You can enable pcap tracing on a particular node/net-device pair by providing a
@code{Ptr<NetDevice>} to an @code{EnablePcap} method. The @code{Ptr<Node>} is
implicit since the net device must belong to exactly one @code{Node}.
For example,
@verbatim
Ptr<NetDevice> nd;
...
helper.EnablePcap ("prefix", nd);
@end verbatim
You can enable pcap tracing on a particular node/net-device pair by providing a
@code{std::string} representing an object name service string to an
@code{EnablePcap} method. The @code{Ptr<NetDevice>} is looked up from the name
string. Again, the @code<Node> is implicit since the named net device must
belong to exactly one @code{Node}. For example,
@verbatim
Names::Add ("server" ...);
Names::Add ("server/eth0" ...);
...
helper.EnablePcap ("prefix", "server/ath0");
@end verbatim
You can enable pcap tracing on a collection of node/net-device pairs by
providing a @code{NetDeviceContainer}. For each @code{NetDevice} in the container
the type is checked. For each device of the proper type (the same type as is
managed by the device helper), tracing is enabled. Again, the @code<Node> is
implicit since the found net device must belong to exactly one @code{Node}.
For example,
@verbatim
NetDeviceContainer d = ...;
...
helper.EnablePcap ("prefix", d);
@end verbatim
You can enable pcap tracing on a collection of node/net-device pairs by
providing a @code{NodeContainer}. For each @code{Node} in the @code{NodeContainer}
its attached @code{NetDevices} are iterated. For each @code{NetDevice} attached
to each node in the container, the type of that device is checked. For each
device of the proper type (the same type as is managed by the device helper),
tracing is enabled.
@verbatim
NodeContainer n;
...
helper.EnablePcap ("prefix", n);
@end verbatim
You can enable pcap tracing on the basis of node ID and device ID as well as
with explicit @code{Ptr}. Each @code{Node} in the system has an integer node ID
and each device connected to a node has an integer device ID.
@verbatim
helper.EnablePcap ("prefix", 21, 1);
@end verbatim
Finally, you can enable pcap tracing for all devices in the system, with the
same type as that managed by the device helper.
@verbatim
helper.EnablePcapAll ("prefix");
@end verbatim
In each of these cases, there is an additional parameter that defaults to false.
This parameter indicates that the trace should not be gathered in promiscuous
mode. If you do want your traces to include all traffic seen by the device
(and the device supports promiscuous mode) simply add a true parameter to any
of the calls above. For example,
@verbatim
Ptr<NetDevice> nd;
...
helper.EnablePcap ("prefix", nd, true);
@end verbatim
will enable promiscuous mode captures on the @code{NetDevice} specified by @code{nd}.
@subsubsection Pcap Tracing Device Helper Filename Selection
Implicit in all of the method descriptions above is the construction of the
complete filenames by the implementation method. By convention, pcap traces
in the @code{ns-3} system are of the form ``<prefix>-<node id>-<device id>.pcap''
As previously mentioned, every node in the system will have a system-assigned
node id; and every device will have an interface index (also called a device id)
relative to its node. By default, then, a pcap trace file created as a result
of enabling tracing on the first device of node 21 using the prefix ``prefix''
would be ``prefix-21-1.pcap''.
You can always use the @code{ns-3} object name service to make this more clear.
For example, if you use the object name service to assign the name ``server''
to node 21, the resulting pcap trace file name will automatically become,
``prefix-server-1.pcap'' and if you also assign the name ``eth0'' to the
device, your pcap file name will automatically pick this up and be called
``prefix-server-eth0.pcap''.
@subsection Ascii Tracing Device Helpers
The behavior of the ascii trace helpers is substantially similar to the pcap
case. Take a look at @code{src/helper/trace-helper.h} if you want to
follow the discussion while looking at real code.
The class @code{TraceHelperForDevice} adds the high level functionality for
using ascii tracing to class @code{PcapHelperForDevice}. If a device helper
inherits from class @code{TraceHelperForDevice} it gets both pcap and ascii
tracing powers. Along with the internal pcap function, every device must
also implement a single virtual method inherited from the ascii trace additions
to this class.
@verbatim
virtual void EnableAsciiInternal (Ptr<OutputStreamObject> stream, std::string prefix, Ptr<NetDevice> nd) = 0;
@end verbatim
The signature of this method reflects the device-centric view of the situation
at this level; and also the fact that the helper may be writing to a shared
output stream. All of the public ascii-trace-related methods inherited from
class @code{TraceHelperForDevice} reduce to calling this single device-
dependent implementation method. For example, the lowest level ascii trace
methods,
@verbatim
void EnableAscii (std::string prefix, Ptr<NetDevice> nd);
void EnableAscii (Ptr<OutputStreamObject> stream, Ptr<NetDevice> nd);
@verbatim
will call the device implementation of @code{EnableAsciiInternal} directly,
providing either a valid prefix or stream. All other public ascii tracing
methods will build on these low-level functions to provide additional user-level
functionality. What this means to the user is that all device helpers in the
system will have all of the ascii trace methods available; and these methods
will all work in the same way across devices if the devices implement
@code{EnablAsciiInternal} correctly.
@subsubsection Ascii Tracing Device Helper Methods
@verbatim
void EnableAscii (std::string prefix, Ptr<NetDevice> nd);
void EnableAscii (Ptr<OutputStreamObject> stream, Ptr<NetDevice> nd);
void EnableAscii (std::string prefix, std::string ndName);
void EnableAscii (Ptr<OutputStreamObject> stream, std::string ndName);
void EnableAscii (std::string prefix, NetDeviceContainer d);
void EnableAscii (Ptr<OutputStreamObject> stream, NetDeviceContainer d);
void EnableAscii (std::string prefix, NodeContainer n);
void EnableAscii (Ptr<OutputStreamObject> stream, NodeContainer n);
void EnableAscii (std::string prefix, uint32_t nodeid, uint32_t deviceid);
void EnableAscii (Ptr<OutputStreamObject> stream, uint32_t nodeid, uint32_t deviceid);
void EnableAsciiAll (std::string prefix);
void EnableAsciiAll (Ptr<OutputStreamObject> stream);
@end verbatim
You are encouraged to peruse the Doxygen for class @code{TraceHelperForDevice}
to find the details of these methods; but to summarize ...
There are twice as many methods available for ascii tracing as there were for
pcap tracing. This is because, in addition to the pcap-style model where traces
from each unique node/device pair are written to a unique file, we support a model
in which trace information for many node/device pairs is written to a common file.
This means that the <prefix>-<node>-<device> file name generation mechanism is
replaced by a mechanism to refer to a common file; and the number of API methods
is doubled to allow all combinations.
Just as in pcap tracing, you can enable ascii tracing on a particular
node/net-device pair by providing a @code{Ptr<NetDevice>} to an @code{EnableAscii}
method. The @code{Ptr<Node>} is implicit since the net device must belong to
exactly one @code{Node}. For example,
@verbatim
Ptr<NetDevice> nd;
...
helper.EnableAscii ("prefix", nd);
@end verbatim
In this case, no trace contexts are written to the ascii trace file since they
would be redundant. The system will pick the file name to be created using
the same rules as described in the pcap section, except that the file will
have the suffix ``.tr'' instead of ``.pcap''.
If you want to enable ascii tracing on more than one net device and have all
traces sent to a single file, you can do that as well by using an object to
refer to a single file. We have already seen this in the ``cwnd'' example
above:
@verbatim
Ptr<NetDevice> nd1;
Ptr<NetDevice> nd2;
...
Ptr<OutputStreamObject> stream = asciiTraceHelper.CreateFileStream ("trace-file-name.tr");
...
helper.EnableAscii (stream, nd1);
helper.EnableAscii (stream, nd2);
@verbatim
In this case, trace contexts are written to the ascii trace file since they
are required to disambiguate traces from the two devices. Note that since the
user is completely specifying the file name, the string should include the ``,tr''
for consistency.
You can enable ascii tracing on a particular node/net-device pair by providing a
@code{std::string} representing an object name service string to an
@code{EnablePcap} method. The @code{Ptr<NetDevice>} is looked up from the name
string. Again, the @code<Node> is implicit since the named net device must
belong to exactly one @code{Node}. For example,
@verbatim
Names::Add ("client" ...);
Names::Add ("client/eth0" ...);
Names::Add ("server" ...);
Names::Add ("server/eth0" ...);
...
helper.EnableAscii ("prefix", "client/eth0");
helper.EnableAscii ("prefix", "server/eth0");
@end verbatim
This would result in two files named ``prefix-client-eth0.tr'' and
``prefix-server-eth0.tr'' with traces for each device in the respective trace
file. Since all of the EnableAscii functions are overloaded to take a stream object,
you can use that form as well:
@verbatim
Names::Add ("client" ...);
Names::Add ("client/eth0" ...);
Names::Add ("server" ...);
Names::Add ("server/eth0" ...);
...
Ptr<OutputStreamObject> stream = asciiTraceHelper.CreateFileStream ("trace-file-name.tr");
...
helper.EnableAscii (stream, "client/eth0");
helper.EnableAscii (stream, "server/eth0");
@end verbatim
This would result in a single trace file called ``trace-file-name.tr'' that
contains all of the trace events for both devices. The events would be
disambiguated by trace context strings.
You can enable ascii tracing on a collection of node/net-device pairs by
providing a @code{NetDeviceContainer}. For each @code{NetDevice} in the container
the type is checked. For each device of the proper type (the same type as is
managed by the device helper), tracing is enabled. Again, the @code<Node> is
implicit since the found net device must belong to exactly one @code{Node}.
For example,
@verbatim
NetDeviceContainer d = ...;
...
helper.EnableAscii ("prefix", d);
@end verbatim
This would result in a number of ascii trace files being created, each of which
follows the <prefix>-<node id>-<device id>.tr convention. Combining all of the
traces into a single file is accomplished similarly to the examples above:
@verbatim
NetDeviceContainer d = ...;
...
Ptr<OutputStreamObject> stream = asciiTraceHelper.CreateFileStream ("trace-file-name.tr");
...
helper.EnableAscii (stream, d);
@end verbatim
You can enable ascii tracing on a collection of node/net-device pairs by
providing a @code{NodeContainer}. For each @code{Node} in the @code{NodeContainer}
its attached @code{NetDevices} are iterated. For each @code{NetDevice} attached
to each node in the container, the type of that device is checked. For each
device of the proper type (the same type as is managed by the device helper),
tracing is enabled.
@verbatim
NodeContainer n;
...
helper.EnableAscii ("prefix", n);
@end verbatim
This would result in a number of ascii trace files being created, each of which
follows the <prefix>-<node id>-<device id>.tr convention. Combining all of the
traces into a single file is accomplished similarly to the examples above:
You can enable pcap tracing on the basis of node ID and device ID as well as
with explicit @code{Ptr}. Each @code{Node} in the system has an integer node ID
and each device connected to a node has an integer device ID.
@verbatim
helper.EnableAscii ("prefix", 21, 1);
@end verbatim
Of course, the traces can be combined into a single file as shown above.
Finally, you can enable pcap tracing for all devices in the system, with the
same type as that managed by the device helper.
@verbatim
helper.EnableAsciiAll ("prefix");
@end verbatim
This would result in a number of ascii trace files being created, one for
every device in the system of the type managed by the helper. All of these
files will follow the <prefix>-<node id>-<device id>.tr convention. Combining
all of the traces into a single file is accomplished similarly to the examples
above.
@subsubsection Ascii Tracing Device Helper Filename Selection
Implicit in the prefix-style method descriptions above is the construction of the
complete filenames by the implementation method. By convention, ascii traces
in the @code{ns-3} system are of the form ``<prefix>-<node id>-<device id>.tr''
As previously mentioned, every node in the system will have a system-assigned
node id; and every device will have an interface index (also called a device id)
relative to its node. By default, then, an ascii trace file created as a result
of enabling tracing on the first device of node 21, using the prefix ``prefix'',
would be ``prefix-21-1.tr''.
You can always use the @code{ns-3} object name service to make this more clear.
For example, if you use the object name service to assign the name ``server''
to node 21, the resulting ascii trace file name will automatically become,
``prefix-server-1.tr'' and if you also assign the name ``eth0'' to the
device, your ascii trace file name will automatically pick this up and be called
``prefix-server-eth0.tr''.
@subsection Pcap Tracing Protocol Helpers
The goal of these helpers is to make it easy to add a consistent pcap trace
facility to protocols. We want all of the various flavors of pcap tracing to
work the same across all protocols, so the methods of these helpers are
inherited by stack helpers. Take a look at @code{src/helper/trace-helper.h}
if you want to follow the discussion while looking at real code.
In this section we will be illustrating the methods as applied to the protocol
@code{Ipv4}. To specify traces in similar protocols, just substitute the
appropriate type. For example, use a @code{Ptr<Ipv6>} instead of a
@code{Ptr<Ipv4>} and call @code{EnablePcapIpv6} instead of @code{EnablePcapIpv4}.
The class @code{PcapHelperForIpv4} provides the high level functionality for
using pcap tracing in the @code{Ipv4} protocol. Every protocol enabling these
methods must implement a single virtual method inherited from this class. There
will be a separate implementation for @code{Ipv6}, for example, but the only
difference will be in the method names and signatures. Different method names
are required to disambiguate class @code{Ipv4} from @coe{Ipv6} which are both
derived from class @code{Object}, and methods that share the same signature.
@verbatim
virtual void EnablePcapIpv4Internal (std::string prefix, Ptr<Ipv4> ipv4, uint32_t interface) = 0;
@end verbatim
The signature of this method reflects the protocol and interface-centric view
of the situation at this level. All of the public methods inherited from class
@code{PcapHelperForIpv4} reduce to calling this single device-dependent
implementation method. For example, the lowest level pcap method,
@verbatim
void EnablePcapIpv4 (std::string prefix, Ptr<Ipv4> ipv4, uint32_t interface);
@verbatim
will call the device implementation of @code{EnablePcapIpv4Internal} directly.
All other public pcap tracing methods build on this implementation to provide
additional user-level functionality. What this means to the user is that all
protocol helpers in the system will have all of the pcap trace methods
available; and these methods will all work in the same way across
protocols if the helper implements @code{EnablePcapIpv4Internal} correctly.
@subsubsection Pcap Tracing Protocol Helper Methods
These methods are designed to be in one-to-one correspondence with the @code{Node}-
and @code{NetDevice}- centric versions of the device versions. Instead of
@code{Node} and @code{NetDevice} pair constraints, we use protocol and interface
constraints.
Note that just like in the device version, there are six methods:
@verbatim
void EnablePcapIpv4 (std::string prefix, Ptr<Ipv4> ipv4, uint32_t interface);
void EnablePcapIpv4 (std::string prefix, std::string ipv4Name, uint32_t interface);
void EnablePcapIpv4 (std::string prefix, Ipv4InterfaceContainer c);
void EnablePcapIpv4 (std::string prefix, NodeContainer n);
void EnablePcapIpv4 (std::string prefix, uint32_t nodeid, uint32_t interface);
void EnablePcapIpv4All (std::string prefix);
@end verbatim
You are encouraged to peruse the Doxygen for class @code{PcapHelperForIpv4}
to find the details of these methods; but to summarize ...
You can enable pcap tracing on a particular protocol/interface pair by providing a
@code{Ptr<Ipv4>} and @code{interface} to an @code{EnablePcap} method. For example,
@verbatim
Ptr<Ipv4> ipv4 = node->GetObject<Ipv4> ();
...
helper.EnablePcapIpv4 ("prefix", ipv4, 0);
@end verbatim
You can enable pcap tracing on a particular node/net-device pair by providing a
@code{std::string} representing an object name service string to an
@code{EnablePcap} method. The @code{Ptr<Ipv4>} is looked up from the name
string. For example,
@verbatim
Names::Add ("serverIPv4" ...);
...
helper.EnablePcapIpv4 ("prefix", "serverIpv4", 1);
@end verbatim
You can enable pcap tracing on a collection of protocol/interface pairs by
providing an @code{Ipv4InterfaceContainer}. For each @code{Ipv4} / interface
pair in the container the protocol type is checked. For each protocol of the
proper type (the same type as is managed by the device helper), tracing is
enabled for the corresponding interface. For example,
@verbatim
NodeContainer nodes;
...
NetDeviceContainer devices = deviceHelper.Install (nodes);
...
Ipv4AddressHelper ipv4;
ipv4.SetBase ("10.1.1.0", "255.255.255.0");
Ipv4InterfaceContainer interfaces = ipv4.Assign (devices);
...
helper.EnablePcapIpv4 ("prefix", interfaces);
@end verbatim
You can enable pcap tracing on a collection of protocol/interface pairs by
providing a @code{NodeContainer}. For each @code{Node} in the @code{NodeContainer}
the appropriate protocol is found. For each protocol, its interfaces are
enumerated and tracing is enabled on the resulting pairs. For example,
@verbatim
NodeContainer n;
...
helper.EnablePcapIpv4 ("prefix", n);
@end verbatim
You can enable pcap tracing on the basis of node ID and interface as well. In
this case, the node-id is translated to a @code{Ptr{Node} and the appropriate
protocol is looked up in the node. The resulting protocol and interface are
used to specify the resulting trace source.
@verbatim
helper.EnablePcapIpv4 ("prefix", 21, 1);
@end verbatim
Finally, you can enable pcap tracing for all interfaces in the system, with
associated protocol being the same type as that managed by the device helper.
@verbatim
helper.EnablePcapIpv4All ("prefix");
@end verbatim
@subsubsection Pcap Tracing Protocol Helper Filename Selection
Implicit in all of the method descriptions above is the construction of the
complete filenames by the implementation method. By convention, pcap traces
taken for devices in the @code{ns-3} system are of the form
``<prefix>-<node id>-<device id>.pcap''. In the case of protocol traces,
there is a one-to-one correspondence between protocols and @code{Nodes}.
This is because protocol @code{Objects} are aggregated to @code{Node Objects}.
Since there is no global protocol id in the system, we use the corresponding
node id in file naming. Threfore there is a possibility for file name
collisions in automatically chosen trace file names. For this reason, the
file name convention is changed for protocol traces.
As previously mentioned, every node in the system will have a system-assigned
node id. Since there is a one-to-one correspondence between protocol instances
and node instances we use the node id. Each interface has an interface id
relative to its protocol. We use the convention
"<prefix>-n<node id>-i<interface id>.pcap" for trace file naming in protocol
helpers.
Therefore, by default, a pcap trace file created as a result of enabling tracing
on interface 1 of the Ipv4 protocol of node 21 using the prefix ``prefix''
would be ``prefix-n21-i1.pcap''.
You can always use the @code{ns-3} object name service to make this more clear.
For example, if you use the object name service to assign the name ``serverIpv4''
to the Ptr<Ipv4> on node 21, the resulting pcap trace file name will
automatically become, ``prefix-nserverIpv4-i1.pcap''.
@subsection Ascii Tracing Protocol Helpers
The behavior of the ascii trace helpers is substantially similar to the pcap
case. Take a look at @code{src/helper/trace-helper.h} if you want to
follow the discussion while looking at real code.
In this section we will be illustrating the methods as applied to the protocol
@code{Ipv4}. To specify traces in similar protocols, just substitute the
appropriate type. For example, use a @code{Ptr<Ipv6>} instead of a
@code{Ptr<Ipv4>} and call @code{EnableAsciiIpv6} instead of @code{EnableAsciiIpv4}.
The class @code{PcapAndAsciiHelperForIpv4} adds the high level functionality
for using ascii tracing to the PcapHelperForIpv4. Every protocol must
additionally implement a single virtual method inherited from this class.
@verbatim
virtual void EnableAsciiIpv4Internal (Ptr<OutputStreamObject> stream, std::string prefix,
Ptr<Ipv4> ipv4, uint32_t interface) = 0;
@end verbatim
The signature of this method reflects the protocol- and interface-centric view
of the situation at this level; and also the fact that the helper may be writing
to a shared output stream. All of the public methods inherited from class
@code{PcapAndAsciiTraceHelperForIpv4} reduce to calling this single device-
dependent implementation method. For example, the lowest level ascii trace
methods,
@verbatim
void EnableAsciiIpv4 (std::string prefix, Ptr<Ipv4> ipv4, uint32_t interface);
void EnableAsciiIpv4 (Ptr<OutputStreamObject> stream, Ptr<Ipv4> ipv4, uint32_t interface);
@verbatim
will call the device implementation of @code{EnableAsciiIpv4Internal} directly,
providing either the prefix or the stream. All other public ascii tracing
methods will build on these low-level functions to provide additional user-level
functionality. What this means to the user is that all device helpers in the
system will have all of the ascii trace methods available; and these methods
will all work in the same way across protocols if the protocols implement
@code{EnablAsciiIpv4Internal} correctly.
@subsubsection Ascii Tracing Device Helper Methods
@verbatim
void EnableAsciiIpv4 (std::string prefix, Ptr<Ipv4> ipv4, uint32_t interface);
void EnableAsciiIpv4 (Ptr<OutputStreamObject> stream, Ptr<Ipv4> ipv4, uint32_t interface);
void EnableAsciiIpv4 (std::string prefix, std::string ipv4Name, uint32_t interface);
void EnableAsciiIpv4 (Ptr<OutputStreamObject> stream, std::string ipv4Name, uint32_t interface);
void EnableAsciiIpv4 (std::string prefix, Ipv4InterfaceContainer c);
void EnableAsciiIpv4 (Ptr<OutputStreamObject> stream, Ipv4InterfaceContainer c);
void EnableAsciiIpv4 (std::string prefix, NodeContainer n);
void EnableAsciiIpv4 (Ptr<OutputStreamObject> stream, NodeContainer n);
void EnableAsciiIpv4 (std::string prefix, uint32_t nodeid, uint32_t deviceid);
void EnableAsciiIpv4 (Ptr<OutputStreamObject> stream, uint32_t nodeid, uint32_t interface);
void EnableAsciiIpv4All (std::string prefix);
void EnableAsciiIpv4All (Ptr<OutputStreamObject> stream);
@end verbatim
You are encouraged to peruse the Doxygen for class @code{PcapAndAsciiHelperForIpv4}
to find the details of these methods; but to summarize ...
There are twice as many methods available for ascii tracing as there were for
pcap tracing. This is because, in addition to the pcap-style model where traces
from each unique protocol/interface pair are written to a unique file, we
support a model in which trace information for many protocol/interface pairs is
written to a common file. This means that the <prefix>-n<node id>-<interface>
file name generation mechanism is replaced by a mechanism to refer to a common
file; and the number of API methods is doubled to allow all combinations.
Just as in pcap tracing, you can enable ascii tracing on a particular
protocol/interface pair by providing a @code{Ptr<Ipv4>} and an @code{interface}
to an @code{EnableAscii} method.
For example,
@verbatim
Ptr<Ipv4> ipv4;
...
helper.EnableAsciiIpv4 ("prefix", ipv4, 1);
@end verbatim
In this case, no trace contexts are written to the ascii trace file since they
would be redundant. The system will pick the file name to be created using
the same rules as described in the pcap section, except that the file will
have the suffix ``.tr'' instead of ``.pcap''.
If you want to enable ascii tracing on more than one interface and have all
traces sent to a single file, you can do that as well by using an object to
refer to a single file. We have already something similar to this in the
``cwnd'' example above:
@verbatim
Ptr<Ipv4> protocol1 = node1->GetObject<Ipv4> ();
Ptr<Ipv4> protocol2 = node2->GetObject<Ipv4> ();
...
Ptr<OutputStreamObject> stream = asciiTraceHelper.CreateFileStream ("trace-file-name.tr");
...
helper.EnableAsciiIpv4 (stream, protocol1, 1);
helper.EnableAsciiIpv4 (stream, protocol2, 1);
@verbatim
In this case, trace contexts are written to the ascii trace file since they
are required to disambiguate traces from the two interfaces. Note that since
the user is completely specifying the file name, the string should include the
``,tr'' for consistency.
You can enable ascii tracing on a particular protocol by providing a
@code{std::string} representing an object name service string to an
@code{EnablePcap} method. The @code{Ptr<Ipv4>} is looked up from the name
string. The @code<Node> in the resulting filenames is implicit since there
is a one-to-one correspondence between protocol instances and nodes,
For example,
@verbatim
Names::Add ("node1Ipv4" ...);
Names::Add ("node2Ipv4" ...);
...
helper.EnableAsciiIpv4 ("prefix", "node1Ipv4", 1);
helper.EnableAsciiIpv4 ("prefix", "node2Ipv4", 1);
@end verbatim
This would result in two files named ``prefix-nnode1Ipv4-i1.tr'' and
``prefix-nnode2Ipv4-i1.tr'' with traces for each interface in the respective
trace file. Since all of the EnableAscii functions are overloaded to take a
stream object, you can use that form as well:
@verbatim
Names::Add ("node1Ipv4" ...);
Names::Add ("node2Ipv4" ...);
...
Ptr<OutputStreamObject> stream = asciiTraceHelper.CreateFileStream ("trace-file-name.tr");
...
helper.EnableAsciiIpv4 (stream, "node1Ipv4", 1);
helper.EnableAsciiIpv4 (stream, "node2Ipv4", 1);
@end verbatim
This would result in a single trace file called ``trace-file-name.tr'' that
contains all of the trace events for both interfaces. The events would be
disambiguated by trace context strings.
You can enable ascii tracing on a collection of protocol/interface pairs by
providing an @code{Ipv4InterfaceContainer}. For each protocol of the proper
type (the same type as is managed by the device helper), tracing is enabled
for the corresponding interface. Again, the @code<Node> is implicit since
there is a one-to-one correspondence between each protocol and its node.
For example,
@verbatim
NodeContainer nodes;
...
NetDeviceContainer devices = deviceHelper.Install (nodes);
...
Ipv4AddressHelper ipv4;
ipv4.SetBase ("10.1.1.0", "255.255.255.0");
Ipv4InterfaceContainer interfaces = ipv4.Assign (devices);
...
...
helper.EnableAsciiIpv4 ("prefix", interfaces);
@end verbatim
This would result in a number of ascii trace files being created, each of which
follows the <prefix>-n<node id>-i<interface>.tr convention. Combining all of the
traces into a single file is accomplished similarly to the examples above:
@verbatim
NodeContainer nodes;
...
NetDeviceContainer devices = deviceHelper.Install (nodes);
...
Ipv4AddressHelper ipv4;
ipv4.SetBase ("10.1.1.0", "255.255.255.0");
Ipv4InterfaceContainer interfaces = ipv4.Assign (devices);
...
Ptr<OutputStreamObject> stream = asciiTraceHelper.CreateFileStream ("trace-file-name.tr");
...
helper.EnableAsciiIpv4 (stream, interfaces);
@end verbatim
You can enable ascii tracing on a collection of protocol/interface pairs by
providing a @code{NodeContainer}. For each @code{Node} in the @code{NodeContainer}
the appropriate protocol is found. For each protocol, its interfaces are
enumerated and tracing is enabled on the resulting pairs. For example,
@verbatim
NodeContainer n;
...
helper.EnableAsciiIpv4 ("prefix", n);
@end verbatim
This would result in a number of ascii trace files being created, each of which
follows the <prefix>-<node id>-<device id>.tr convention. Combining all of the
traces into a single file is accomplished similarly to the examples above:
You can enable pcap tracing on the basis of node ID and device ID as well. In
this case, the node-id is translated to a @code{Ptr{Node} and the appropriate
protocol is looked up in the node. The resulting protocol and interface are
used to specify the resulting trace source.
@verbatim
helper.EnableAsciiIpv4 ("prefix", 21, 1);
@end verbatim
Of course, the traces can be combined into a single file as shown above.
Finally, you can enable ascii tracing for all interfaces in the system, with
associated protocol being the same type as that managed by the device helper.
@verbatim
helper.EnableAsciiIpv4All ("prefix");
@end verbatim
This would result in a number of ascii trace files being created, one for
every interface in the system related to a protocol of the type managed by the
helper. All of these files will follow the <prefix>-n<node id>-i<interface.tr
convention. Combining all of the traces into a single file is accomplished
similarly to the examples above.
@subsubsection Ascii Tracing Device Helper Filename Selection
Implicit in the prefix-style method descriptions above is the construction of the
complete filenames by the implementation method. By convention, ascii traces
in the @code{ns-3} system are of the form ``<prefix>-<node id>-<device id>.tr''
As previously mentioned, every node in the system will have a system-assigned
node id. Since there is a one-to-one correspondence between protocols and nodes
we use to node-id to identify the protocol identity. Every interface on a
given rotocol will have an interface index (also called simply an interface)
relative to its protocol. By default, then, an ascii trace file created as a result
of enabling tracing on the first device of node 21, using the prefix ``prefix'',
would be ``prefix-n21-i1.tr''. Use the prefix to disambiguate multiple protocols
per node.
You can always use the @code{ns-3} object name service to make this more clear.
For example, if you use the object name service to assign the name ``serverIpv4''
to the protocol on node 21, and also specify interface one, the resulting ascii
trace file name will automatically become, ``prefix-nserverIpv4-1.tr''.
@node Tracing implementation details
@section Implementation details