Hi All,
We encountered an error upon running the "osmo-trx” using Ubuntu 16.04 OS. We have an existing "osmo-trx” in one of our server and running just fine. Upon checking for the version, the working “osmo-trx” is running at “0.1.9.20170530”, while the new one with an error is running at “0.1.9.20170621”.
Kindly see below error for your reference:
# osmo-trx
osmo-trx: symbol lookup error: osmo-trx: undefined symbol: _ZN3uhd4usrp10multi_usrp9ALL_GAINSE
Best Regard,
Ron Menez
ron.menez(a)entropysolution.com<mailto:ron.menez@entropysolution.com>
Hi.
I've got add few more parameters to OsmoBTS and OsmoPCU gerrit jobs in jenkins. So
far this has been done by manually adding parameter axis and writing combination
filter for axis parameters. It's error prone and rather ugly. Much better way would
be to use JobDSL and declare entire job programmatically but until that happens
there's nice intermediate solution:
https://plugins.jenkins.io/matrix-combinations-parameter
I'd like to install this plugin onto our jenkins and convert above mentioned jobs to
use it. Opinions, other ideas?
On a related note: who's responsible for jenkins config/update? Shall I just contact
that person directly if I need some plugin in jenkins, or shall I ask in ML?
--
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
Today we faced consistent osmo-bts-trx failures and I submitted a revert of
commit "RSL: receive and send multiple SI2q messages".
See https://osmocom.org/issues/2338.
Why did we not catch it earlier? There was a bug in the osmo-gsm-tester build
where the build could get stuck on a local branch instead of building newly
pulled changes. That was fixed about a day ago, so we saw the breakage today.
~N
Hi,
Some of you may have noticed that I have been lately fixing compilation
warnings + enabling -Wall and -Werror on jenkins build for several
osmocom libraries.
While doing so, I noticed that the following lines trying to enable
AddressSanitizer support are actually failing when building with the
FreeBSD jenkins slave, probably due to it using some other shell without
bash support:
./configure: CFLAGS+= -fsanitize=address -fsanitize=undefined: not found
./configure: CPPFLAGS+= -fsanitize=address -fsanitize=undefined: not found
The source lines are:
CFLAGS+=" -fsanitize=address -fsanitize=undefined"
CPPFLAGS+=" -fsanitize=address -fsanitize=undefined"
After changing them to support sh and submitting the change in gerrit
[1], AddressSanitizer was enabled in FreeBSD for first time, but I
started getting lots of link issues with AdressSanitizer (asan) related
symbols.
I read in several places that there seems to be a known issue with
FreeBSD 10.3 and clang-3.4.1 (what we are using) regarding use of
AddressSanitizer [2, 3].
I then tried configuring CC in the jenkins slave to use clang37 instead
of cc (clang-3.4.1), and then the linking issue seems to be solved, but
again a new compiler bug related to usage of __builtin_cpu_supports
shows up, preventing compilation of libosmocore [4].
I think the best solution is to disable explicitly AddressSanitizer in
contrib/jenkins.sh when running from FreeBSD, so the behaviour is
actually the same as before, in which it was implicitly not enabled due
to a syntax error. Once we move to FreeBSD11 and clang-3.8 we can try
enabling it again.
If nobody comes with a better solution or is against this, I will
proceed to do so during next hours/days.
[1] https://gerrit.osmocom.org/#/c/3024/2/configure.ac
[2]
https://stackoverflow.com/questions/30085036/linking-clang-address-sanitize…
[3]
https://webcache.googleusercontent.com/search?q=cache:CZmrK35zigEJ:ea5faa5p…
[4] https://bugs.llvm.org//show_bug.cgi?id=25510
--
- Pau Espin Pedrol <pespin(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
Trying to get the HNBGW to communicate with the MSC. I receive Iuh from the
femto cell as usual, send to the MSC, and end up in an infinite loop where
osmo-stp and osmo-hnbgw send the same message back and forth to each other.
osmo-hnbgw's local IuCS PC is 23, the MSC is PC 1.
It seems when osmo-hnbgw receives a message to PC 23, i.e. to itself, it thinks
that its own PC isn't local, finds its own as-clnt-CS in the table of routes,
and feeds it to m3ua_tx_xua_as(). osmo-stp then sends it back. repeat.
It seems overkill that a "client" like osmo-hnbgw should be capable of routing
in the first place. It could switch off routing entirely and receive & handle
every message it receives itself. But assuming that this were nonsense:
IIUC it should rather decide that its own PC 23 is local, in:
bool osmo_ss7_pc_is_local(struct osmo_ss7_instance *inst, uint32_t pc)
{
OSMO_ASSERT(ss7_initialized);
if (pc == inst->cfg.primary_pc)
return true;
/* FIXME: Secondary and Capability Point Codes */
return false;
}
Apparently this global primary_pc is set in osmo_sccp_simple_client(). Since
in the osmo-hnbgw I'm (so far) creating two simple clients with distinct PCs
(IuCS is 23, IuPS is 24), the primary_pc is overwritten with 24. After that we
always fail to notice 23 as a local PC and bounce messages for self right back
down the SCCP stack, because apparently we find a route for 23 instead.
A "matching" route is recognised by osmo_ss7_route_find_dpc():
if ((dpc & rt->cfg.mask) == rt->cfg.pc)
My own local client gets entered as route, with rt->cfg.mask == 0 and
rt->cfg.pc == 0, so an arbitrary dpc & 0 == 0, matches always. It seems
awfully easy to create a fatal infinite loop flooding the network and CPU --
simply send a message to a DPC neither side sees as local. Anyway...
Which way to go to serve more than one PC in a program? It appears that either
an osmo_sccp_simple_client() call's local PC or an osmo_sccp_user_bind_pc()'s
PC should be stored&found as a local PC. ... or both.
Is the "right" way to have two PCs: having two osmo_sccp_simple_client()s with
an sccp_user each; or rather one osmo_sccp_simple_client with two users?
Or, should we for the OsmoHNBGW just have one osmo_sccp_simple_client() and one
user, i.e. only one PC to receive both IuCS and IuPS messages? All we do is
feed both of them out to Iuh anyway... (This seems an easy way out of the multi
PC problem, but our MSC wants to serve one PC for BSSAP=2G and one PC for
RANAP=3G, so we will need to solve the multi-PC problem anyway.)
There are many directions I could head to, is any one better than the others?
Thanks!
~N
In january, osmo_sock_get_name() was introduced and prints the connection info
for a given fd, which looks like this:
127.0.0.1:2905<->127.0.0.1:60661
However, when I first saw this in a log, I was confused: I expect the local
address and port to be on the left, while in this string it is on the right.
Is it only me and I should get used to it, or can we reverse that so the local
part is on the left? I could argue that in most GSM graphics and ladder
diagrams, the client side is on the left, and that we would typically use this
string on the client side to indicate where we connected to. Is this even true?
Should there be an arg to pick the preferred order? :P
I'll submit a patch if anyone agrees.
~N
Hi.
I've problem figuring out how to use osmo-ci in jenkins: the build steps for
osmo-bts-gerrit and osmo-pcu-gerrit look pretty-much the same but osmo-pcu fails with:
osmo-layer1-headers.sh: command not found
while osmo-bts (which uses other scripts from osmo-ci) works fine.
Any ideas where this difference might come from?
--
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
Hi all,
I have some uncertainty about how to use the CTRL interface to set TRAPs.
In the OsmoBSC user manual (for example), the example provided doesn't
specify an option to set for the 'var' I might be particularly
interested in monitoring:
`$ ./bsc_control.py -d localhost -m`
Is this the default to monitor all? Can I go about narrowing this?
Also, I don't feel like I quite understand the TRAP portion of "Table
2: Variables available over control interface" in the OsmoBSC user
manual:
For example, the 'name' column seems to generally indicate the 'var'
that one would pass in a GET/SET command, so I'm assuming the 'name'
associated with the TRAP-enabled rows has the same significance. But
these 'names' have no real meaning to me in terms of my experience
with VTY, for example 'notification'.
I have tried to look toward the osmo-nitb unit tests for inspiration,
as there is fantastic coverage for GET/SET, out haven't seen similar
for TRAP.
Can someone please point me in the right direction?
Thanks,
Emily
