NAME
redbug - restricted debugging utility
SYNOPSIS
start in non-blocking mode
redbug:start(Trc,[Opts]) -> {integer(NoOfProcs),integer(NoOfFuncs)} |
{atom(ErrorType),term(ErrorReason)}
start blocking mode
redbug:start(Trc,[Opts]) -> {atom(StopReason),list(TraceMessages)} |
{atom(ErrorType),term(ErrorReason)}
stop
redbug:stop() -> stopped | not_started
help
redbug:help() -> ok
DESCRIPTION
redbug is a tool to interact with the Erlang trace facility. It will instruct the Erlang VM to generate so-called 'trace messages' when certain events (such as a particular function being called) occur. It uses a safe subset of the tracing functionality, and exits if it feels overloaded, e.g. if it gets flooded by trace messages. It runs in the background, collecting trace messages, until it reaches one of its termination criteria (number of messages/file size or elapsed time). The trace messages are either printed (i.e. human readable) to a file or to the screen; or written to a trc file. Using a trc file puts less stress on the system, but there is no way to count the messages (so the 'msgs' opt is ignored), and the files can only be read by special tools (such as 'bread'). Printing and trc files cannot be combined. By default (i.e. if the 'file' opt is not given), messages are printed.
Trc: trace type
list('send'|'receive'|string(RTP))
RTP: restricted trace pattern. the RTP has the form:
"MFA when GUARDS -> ACTIONS"
where MFA can be;
"mod", "mod:fun", "mod:fun/3" or "mod:fun('_',atom,X)"
GUARD is something like;
"X==1" or "is_atom(A)"
and ACTION is;
"return" and/or "stack" (separated by ",")
So, an RTP looks something like this;
"ets:lookup(T,hostname) when is_integer(T) -> stack"
Note that bindings (like we're binding 'T' above) works as expected. So this RTP;
"maps:to_list(#{a:=T,c:=#{d:=T}})"
will not trigger on this call;
maps:to_list(#{a=>b,c=>#{d=>e,f=>g}}).
but will trigger on this;
maps:to_list(#{a=>b,c=>#{d=>b,f=>g}})."
Opts: list({Opt,Val})
general opts:
time (15000) stop trace after this many milliseconds
msgs (10) stop trace after this many messages
target (node()) node to trace on
cookie ('') target node cookie
blocking (false) block start/2, return a list of messages
procs (all) atom(all|new|running) | list(Proc)
where Proc is pid() | atom(RegName) | {pid,I2,I3}
max_queue (5000) exit if redbug-internal queue gets this long
max_msg_size (50000) exit if seeing a message bigger than this
debug (false) bigger error messages
print-related opts
arity (false) print arity instead of argument list
buffered (no) buffer messages till end of trace
print_pids (false) print pid instead of registered name
print_calls (true) print calls
print_file (standard_io) print to this file
print_msec (false) print milliseconds on time stamps
print_depth (999999) formatting depth for "~P"
print_re ("") print only strings that match this regexp
print_return (true) print return value
print_fun ('') custom print fun. gets called once for each trace
message. It can be a fun/1, (called as F(Msg),
return value is ignored), or a fun/2 (called as
F(Msg,Acc), return is next Acc)
trc file related opts
file (none) use a trc file based on this name
file_size (1) size of each trc file
file_count (8) number of trc files
EXAMPLES
%% Basic call trace
1> redbug:start("erlang:demonitor").
{30,2}
15:39:00 <{erlang,apply,2}> {erlang,demonitor,[#Ref<0.0.0.21493>]}
15:39:00 <{erlang,apply,2}> {erlang,demonitor,[#Ref<0.0.0.21499>]}
15:39:00 <{erlang,apply,2}> {erlang,demonitor,[#Ref<0.0.0.21500>]}
redbug done, timeout - 3
%% As above, print pids
2> redbug:start("erlang:demonitor",[{msgs,1},print_pids]).
{30,2}
15:42:04 <0.31.0> {erlang,demonitor,[#Ref<0.0.0.21616>]}
redbug done, msg_count - 1
%% As above, print return value. The return value is a separate message.
3> redbug:start("erlang:demonitor->return",[{msgs,2},print_pids]).
{30,2}
15:43:22 <0.31.0> {erlang,demonitor,[#Ref<0.0.0.21677>]}
15:43:22 <0.31.0> erlang:demonitor/1 -> true
redbug done, msg_count - 2
%% As above, also print the call stack. Note that not all functions in the
%% call chain are on the stack, only functions we will return to (this is a
%% consequence of tail call optimization.)
4> redbug:start("erlang:demonitor->return,stack",[{msgs,2},print_pids]).
{30,2}
15:44:35 <0.31.0> {erlang,demonitor,[#Ref<0.0.0.21726>]}
shell:eval_loop/3
shell:eval_exprs/7
shell:exprs/7
erl_eval:do_apply/6
redbug:start/2
15:44:35 <0.31.0> erlang:demonitor/1 -> true
redbug done, msg_count - 2
%% Trace on messages that the shell process receives.
5> redbug:start('receive',[{procs,[self()]}]).
{1,0}
15:15:47 <{erlang,apply,2}> <<< {running,1,0}
15:17:49 <{erlang,apply,2}> <<< timeout
redbug done, timeout - 2
%% As above, but also trace on sends from the shell process. note that in
%% this case the 'print_pids' opt would hide that there is a send to the
%% group server.
7> redbug:start([send,'receive'],[{procs,[self()]}]).
{1,0}
15:36:36 <{erlang,apply,2}> <<< {running,1,0}
15:36:36 <{erlang,apply,2}> <{group,server,3}> <<< {io_request,<0.31.0>,
<0.24.0>,
{get_geometry,columns}}
redbug done, timeout - 2
%% Call trace with a function head match. Note that the first call to
%% ets:tab2list/1 does not trigger the tracer.
8> redbug:start("ets:tab2list(inet_db)",[{msgs,2},print_pids]).
{30,1}
9> ets:tab2list(ac_tab),ok.
ok
10> ets:tab2list(inet_db),ok.
ok
15:47:15 <0.31.0> {ets,tab2list,[inet_db]}
redbug done, timeout - 1
%% As above, but use the 'blocking' opt. redbug:start/2 blocks until end of
%% trace, and returns the stop reason and a list of trace messages.
10> spawn(fun()->receive after 2000->ets:tab2list(inet_db) end end).
<0.540.0>
11> redbug:start("ets:tab2list(inet_db)",[blocking]).
{timeout,[{call,{{ets,tab2list,[inet_db]},<<>>},
{<0.540.0>,{erlang,apply,2}},
{15,50,43,776041}}]}