Ever since I upgraded to debian 9, osmo-hnbgw failed to link in my build:
/usr/bin/ld: hnbgw.o: undefined reference to symbol 'sctp_send@@VERS_1'
I noticed that adding -lsctp to the build fixed that, yet noticed that no build
slaves nor anyone else seemed to need that.
Some events coincided my re-install of deb9:
- There has always been an -lsctp in the Makefile.am for osmo-hnbgw, but it was
recently dropped in
http://git.osmocom.org/osmo-iuh/commit/?id=7235ea02d78299471d58f4202d8c8d68…
- I adopted the jenkins.sh -Werror flags in some builds to get fatal warnings.
Passed by './configure CFLAGS=-Werror'. Someone else added this in jenkins.sh
and I assumed that was a good way to do things.
- I noticed that I couldn't see debug symbols in gdb, so added 'CLFAGS=-g' in
my builds. read on...
Comparing with a docker build kindly provided by laforge, I noticed that a
plain 'clone; ./configure; make' does actually succeed on my system!
It is a convolution of CFLAGS causing the error.
By passing a 'CFLAGS=foo' to the osmo-iuh configure script, not only do I add
'foo', but I override CFLAGS defaults, and actually *remove* other CFLAGS.
Turns out, whether I pass CFLAGS=-g or not, -g is always part of the build
command lines: '-g -O2' is the default. Funnily enough, with CFLAGS=-g
passed, I end up removing the -O2 option.
A conclusion here is that in jenkins.sh, we should *not* pass CFLAGS to
configure, so that we don't override any default CFLAGS. A configure option
like --with-strict-warnings can add to CFLAGS in configure.ac instead
So now I guess because I added CFLAGS=-Werror before, I ended up removing -g.
Hence I saw a need to add -g again, goof. Did so in all builds while I was at
it, hence -g ended up in osmo-iuh as well.
The -g stuck around, and building osmo-iuh with CFLAGS=-g does, as I already
said, effectively remove the -O2 option for all things built in osmo-iuh.
Now I'm confused. Why does a lack of -O2 break a linkage.
It's not the osmo-hnbgw linkage per se that seems to be the cause. I can link
that last step with or without -O2 successfully. It must be a combination of
building a depending library without -O2 and then linking that to osmo-iuh.
I diffed the entire build log between a succeeding and a failing build of
osmo-iuh, and really the only difference everywhere is that the succeeding
build uses -O2 everywhere.
The only differences in config.log are the same: most conftest programs have
-O2 removed, few have a -g added. The conftest conclusions are all identical.
My big question now is: sctp_send() is actually called from hnbgw.c directly.
So IIUC, we should in fact be required to pass -lsctp to the linker for
osmo-hnbgw. If that is correct, how am I allowed to skip that by passing -O2?
That's all I figured out so far. I still would like to find out why this can
happen, any ideas would be appreciated.
Anyone should be able to reproduce this by building osmo-iuh with
./configure CFLAGS=-g
~N
Hi dexter,
I just noticed this patch that is merged, with the lines
+ LOGPFSML(reset->fsm, LOGL_NOTICE, "fsm-state (msc-reset): %s, fsm-event: %s\n",
get_value_string(fsm_state_names, ST_CONN), get_value_string(fsm_evt_names, event));
If you take a look at the log output this produces, you should see that LOGPFSM
already contains information that it is an FSM, which state it is in, and that
just before that an FSM event has fired. This log statement seems obsolete?
Here's an example where you did everything right, just add a message to the FSM
logging:
+ LOGPFSML(reset->fsm, LOGL_NOTICE, "SIGTRAN connection down, reconnecting...\n");
I marked in the (merged) patch those logging lines that seem to merely
duplicate what we already see in the logs: https://gerrit.osmocom.org/4754
~N
Hi all,
we recently had some (verbal) discussions on how to handle endpoint
allocations in a scenario where a single osmo-mgw shall be used by both
osmo-bsc and osmo-msc (in a "NITB style" setup).
There was some thinking about whether we should somehow divide fixed
ranges of the rtp-bridge endpoints between the BSC and the MSC and how
to avoid having to manually configure those.
I think I now found the proper solution for this: The call agent must
not even fully qualify the endpoint, but it can ask the MGW to allocate
one! To quote from the MGCP RFC:
EndpointId is the identifier for the connection endpoint in the
gateway where CreateConnection executes. The EndpointId can be
fully-specified by assigning a value to the parameter EndpointId in
the function call or it may be under-specified by using the "any of"
wildcard convention. If the endpoint is underspecified, the endpoint
identifier SHALL be assigned by the gateway and its complete value
returned in the SpecificEndPointId parameter of the response. When
the "any of" wildcard is used, the endpoint assigned MUST be in-
service and MUST NOT already have any connections on it. If no such
endpoint is available, error code 410 (no endpoint available) SHOULD
be returned. The "all of" wildcard MUST NOT be used.
So in our concrete example, the BSC (or MSC) could simply ask for
"rtpbridge/*@mgw" as end-point name in CRCX, and the MGW would then
dynamically allocate a suitable free rtpbridge endpoint type and return
it in the call agent in the BSC (or MSC). No shared knowledge about
endpoint number allocations / ranges or the like required.
Now we only need to implement this in osmo-mgw ;)
Related: https://osmocom.org/issues/2631
Regards,
Harald
--
- Harald Welte <laforge(a)gnumonks.org> http://laforge.gnumonks.org/
============================================================================
"Privacy in residential applications is a desirable marketing option."
(ETSI EN 300 175-7 Ch. A6)
Hi.
I've stumbled upon awesome hack which seems to be relevant:
https://ha.cking.ch/s8_data_line_locator/
It uses Simtrace to check how hw implant communicates with its sim. The guy also
tried to sniff GPRS traffic but failed (because he've used OpenBTS, he-he :) - I
wonder if it would work better with OsmoBTS.
Anyone else played with similar device already?
--
Max Suraev <msuraev(a)sysmocom.de> http://www.sysmocom.de/
=======================================================================
* sysmocom - systems for mobile communications GmbH
* Alt-Moabit 93
* 10559 Berlin, Germany
* Sitz / Registered office: Berlin, HRB 134158 B
* Geschaeftsfuehrer / Managing Director: Harald Welte
I recently found sanitizer build failures, for example in libosmocore, which
our gerrit build job doesn't catch. At first I thought something was amiss and
maybe our gerrit sanitizer build were broken.
Instead, it turns out my debian9 gcc -fsanitize=address -fsanitize=undefined
finds many more issues that the debian8 one does not.
https://jenkins.osmocom.org/jenkins/job/gerrit-libosmocore/36/a2=default,a3…
I already have a long series for mem leak fixes for libosmocore, mostly in the
regression testing code, but I count three real mem leak fixes in libosmocore
production code.
My plan is to merge the fixes and then actually move the libosmocore-gerrit to
debian 9 (or at least add debian 9) to make sure we don't add insanities that
the debian 9 build would find.
See https://gerrit.osmocom.org/4861 to https://gerrit.osmocom.org/4875.
I'm also taking a look at the other CellNet builds. The root motivation was to
find out why osmo-hnbgw doesn't work for me anymore currently.
~N
I'm finding heap use after free related to osmo_fsm:
- When dispatching a timer_cb event, we want to set fi->T back to 0. (see fsm_tmr_cb())
- So that we know which timer fired, we leave fi->T at its current value for
the timer_cb to be able to make decisions based on that. This is used for
example in fsm_timeout_cb() in osmo-bsc/osmo_bsc_mgcp.c.
- In other code, the timer_cb event may decide to fire events or call functions
which effect an FSM instance teardown and deallocation, meaning that in
fsm_tmr_cb(), we are *not* allowed to generally assume the fi still exists.
If we set fi->T = 0 after the timer_cb() was called, we may write to freed
mem.
- Also, if the timer_cb effects events with their own timeout, we will also
overwrite the fi->T value with zero after another timer has set its value in
T, causing failures to evaluate that timer later on.
Solutions so far:
- set fi->T = 0 first, and pass T to the timer_cb as arg instead. I think this
is the best and cleanest, but also the hardest in terms of backwards compat.
I have a patch that adds timer_cb2() that features a T arg; we need to submit
that, move all code that wants to use T to that timer_cb2(), and then we can
move the fi->T = 0 to before the timer_cb(). Actually, this is not being
backwards compatible at all, because all code trying to use fi->T as usual
will need adjustment, we break API either way and might as well add T to the
timer_cb directly instead... :/
- Actually not set fi->T = 0 at all. We will have T holding the value of the
timer that fired last, all old code will still find this T in fi->T, and
after the cb, we just leave it as it was. Does it really need to be cleared?
This has no compatibility issues and solves the use after free as well as
setting a new T from timer_cb() issues. Going for this one so far.
Thoughts?
~N
I am receiving a lot of build failure mails every day for quite a while now ...
not finding any time to look at them, I hope you guys are onto those?
~N
