Commit Graph

8 Commits

Author SHA1 Message Date
Willy Tarreau
cfc4f24d80 REORG: vars: move the "proc" scope variables out of the global struct
The presence of this field causes a long dependency chain because almost
everyone includes global-t.h, and vars include sample_data which include
some system includes as well as HTTP parts.

There is absolutely no reason for having the process-wide variables in
the global struct, let's just move them into vars.c and vars.h. This
reduces from ~190k to ~170k the preprocessed output of version.c.
2021-05-08 12:11:29 +02:00
Willy Tarreau
2b71810cb3 CLEANUP: lists/tree-wide: rename some list operations to avoid some confusion
The current "ADD" vs "ADDQ" is confusing because when thinking in terms
of appending at the end of a list, "ADD" naturally comes to mind, but
here it does the opposite, it inserts. Several times already it's been
incorrectly used where ADDQ was expected, the latest of which was a
fortunate accident explained in 6fa922562 ("CLEANUP: stream: explain
why we queue the stream at the head of the server list").

Let's use more explicit (but slightly longer) names now:

   LIST_ADD        ->       LIST_INSERT
   LIST_ADDQ       ->       LIST_APPEND
   LIST_ADDED      ->       LIST_INLIST
   LIST_DEL        ->       LIST_DELETE

The same is true for MT_LISTs, including their "TRY" variant.
LIST_DEL_INIT keeps its short name to encourage to use it instead of the
lazier LIST_DELETE which is often less safe.

The change is large (~674 non-comment entries) but is mechanical enough
to remain safe. No permutation was performed, so any out-of-tree code
can easily map older names to new ones.

The list doc was updated.
2021-04-21 09:20:17 +02:00
Miroslav Zagorac
a8bdf2b655 MINOR: opentracing: transfer of context names without prefix
In order to enable the assignment of a context name, and yet exclude the
use of that name (prefix in this case) when extracting the context from
the HTTP header, a special character '-' has been added, which can be
specified at the beginning of the prefix.

So let's say if we look at examples of the fe-be configuration, we can
transfer the context via an HTTP header without a prefix like this:

  fe/ot.cfg:
        ..
        span "HAProxy session"
            inject "" use-headers
        event on-backend-http-request

Such a context can be read in another process using a name that has a
special '-' sign at the beginning:

  be/ot.cfg:
    ot-scope frontend_http_request
        extract "-ot-ctx" use-headers
        span "HAProxy session" child-of "-ot-ctx" root
        ..

This means that the context name will be '-ot-ctx' but it will not be
used when extracting data from HTTP headers.

Of course, if the context does not have a prefix set, all HTTP headers
will be inserted into the OpenTracing library as context.  All of the
above will only work correctly if that library can figure out what is
relevant to the context and what is not.
2021-04-15 08:40:08 +02:00
Miroslav Zagorac
4b3eb0a940 MINOR: opentracing: correct calculation of the number of arguments in the args[]
It is possible that some arguments within the configuration line are not
specified; that is, they are set to a blank string.

For example:
  keyword '' arg_2

In that case the content of the args field will be like this:
  args[0]:                  'keyword'
  args[1]:                  NULL pointer
  args[2]:                  'arg_2'
  args[3 .. MAX_LINE_ARGS): NULL pointers

The previous way of calculating the number of arguments (as soon as a
null pointer is encountered) could not place an argument on an empty
string.

All of the above is essential for passing the OpenTracing context via
the HTTP headers (keyword 'inject'), where one of the arguments is the
context name prefix.  This way we can set an empty prefix, which is very
useful if we get context from some other process that can't add a prefix
to that data; or we want to pass the context to some process that cannot
handle the prefix of that data.
2021-04-15 08:40:05 +02:00
Willy Tarreau
ff88270ef9 MINOR: pool: move pool declarations to read_mostly
All pool heads are accessed via a pointer and should not be shared with
highly written variables. Move them to the read_mostly section.
2021-04-10 19:27:41 +02:00
Miroslav Zagorac
98272253d8 MINOR: opentracing: register config file and line number on log servers
In commit 9533a7038 new parameters have been added to the declaration
of function parse_logsrv().

This patch should be backported to all branches where the OpenTracing
filter is located.
2021-04-08 11:10:27 +02:00
Miroslav Zagorac
f2263435d7 BUG/MINOR: opentracing: initialization after establishing daemon mode
This patch solves the problem reported in github issue #1204, where the
OpenTracing filter cannot communicate with the selected tracer if HAProxy
is run in daemon mode.  The author of the reported issue uses Zipkin
tracer, while in this example Jaeger tracer is used (see gdb output below).

The problem is that the OpenTracing library is initialized before HAProxy
initialize the daemon mode.  Establishing this mode kills the OpenTracing
thread, after which the correct operation of the OpenTracing filter is no
longer possible.  Also, HAProxy crashes on deinitialization of the
OpenTracing library.

The initialization of the OpenTracing library has been moved from the
flt_ot_init() function (which is started before switching the HAProxy to
daemon mode) to the flt_ot_init_per_thread() function (which is run after
switching the HAProxy to daemon mode).

Gdb output of crashed HAProxy process:

  [Thread debugging using libthread_db enabled]
  Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
  Core was generated by `../../../haproxy -f sa/haproxy.cfg'.
  Program terminated with signal SIGSEGV, Segmentation fault.
  #0  0x00007f8131fd5629 in pthread_join (threadid=140192831239936, thread_return=0x0) at pthread_join.c:45
  45      pthread_join.c: No such file or directory.
  (gdb) where
  #0  0x00007f8131fd5629 in pthread_join (threadid=140192831239936, thread_return=0x0) at pthread_join.c:45
  #1  0x00007f812f15abc7 in std:🧵:join() ()
     from /tmp/haproxy-os-master/contrib/opentracing/test/libjaeger_opentracing_plugin-0.5.0.so
  #2  0x00007f812f0fb6f7 in jaegertracing::reporters::RemoteReporter::close() ()
        from /tmp/haproxy-os-master/contrib/opentracing/test/libjaeger_opentracing_plugin-0.5.0.so
  #3  0x00007f812f0b7055 in jaegertracing::reporters::CompositeReporter::close() ()
           from /tmp/haproxy-os-master/contrib/opentracing/test/libjaeger_opentracing_plugin-0.5.0.so
  #4  0x00007f812f0b9136 in jaegertracing::Tracer::Close() ()
              from /tmp/haproxy-os-master/contrib/opentracing/test/libjaeger_opentracing_plugin-0.5.0.so
  #5  0x00007f81309def32 in ot_tracer_close (tracer=0x55fb48057390) at ../../src/tracer.cpp:91
  #6  0x000055fb41785705 in ot_close (tracer=0x55fb48061168) at contrib/opentracing/src/opentracing.c:208
  #7  0x000055fb4177fc64 in flt_ot_deinit (p=<optimized out>, fconf=<optimized out>) at contrib/opentracing/src/filter.c:215
  #8  0x000055fb418bc038 in flt_deinit (proxy=proxy@entry=0x55fb4805ce50) at src/filters.c:360
  #9  0x000055fb41893ed1 in free_proxy (p=0x55fb4805ce50) at src/proxy.c:315
  #10 0x000055fb41888809 in deinit () at src/haproxy.c:2217
  #11 0x000055fb41889078 in deinit_and_exit (status=0) at src/haproxy.c:2343
  #12 0x000055fb4173d809 in main (argc=<optimized out>, argv=<optimized out>) at src/haproxy.c:3230

This patch should be backported to all branches where the OpenTracing
filter is located.
2021-04-02 18:21:19 +02:00
Willy Tarreau
a63d1a0863 CONTRIB: move contrib/opentracing to addons/ot
This one is the last optional module to build with haproxy, so let's move
it to addons/. It was renamed to "ot" as it was the only one whose USE_*
option did not match the directory name, now this is consistent.

Few changes were required, only the Makefile, and doc were adjusted, as
the directory was already self-contained and relocatable.
2021-04-02 17:48:42 +02:00