| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
|
|
|
|
|
|
| |
This reporter passes the measurement values to the logging subsystem
as DSTATS (which is currently DLGLOBAL) level INFO messages.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently case statements are used to select the right reporter
functions. This makes it difficult to add new reporter types,
especially if they are not going to reside in the same file.
This commit introduces per reporter function pointer for
open, close, send_count, and send_item. They are checked for
non-NULL before being called or skipped.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
| |
To support reporters without network configuration, this commit
introduces the have_net_config flag to provide corresponding error
messages.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If the MTU is given, combine several messages into a single UDP
packet until the limit is reached. Flush all reporters after the
values have been scanned.
New vty commands (node config-stats):
mtu <100-65535> Enable multi-metric packets and set the maximum
packet size (in byte)
no mtu Disable multi-metric packets
Note that single messages that are longer than the given MTU (minus
28 octets protocol overhead) will be dropped.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
| |
The osmo_counter_difference returns the counter value difference
since the last call of this function with the given counter object.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
| |
This calls stats_flush in regular intervals which polls the
statistical values and calls the active reporters when values have
changed.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
| |
This commit provides the stats reporting framework that can manage
several types of measurement reporters. Initially support for
rate_ctr and the statsd protocol is included.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
| |
Fix type and add a note about the semantics of the handler's return
code.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
| |
This commit adds the following functions:
stat_item_for_each_group Call a handler for each group
stat_item_for_each_item Call a handler for each item of a
group
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For global value reporting, some additional helper functions are
needed. The statsd protocol expects differential counter values,
which are currently not provided by rate_ctr (except for s/m/h/d
intervals).
This commit adds several helper functions to rate_ctr:
- rate_ctr_difference returns the counter delta since the last
call to this function for a given counter
- rate_ctr_for_each_counter
iterates through each counter of a group
- rate_ctr_for_each_group
iterates through all globally registered
counter groups
Note that the rate_ctr_difference function can only be used by a
single backend, since it modifies the 'previous' field in the
rate_ctr obj.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently each stat item has a separate index value which basically
counts each single value added to the item and which can be used by
a reporter to get all new values that have not been reported yet.
The drawback is, that such an index must be stored for each stat
item.
This commit introduces a global index which is incremented for each
new stat item value. This index is then stored together with the item
value. So a single stored index per reporter is sufficient to make
sure that only new values are reported.
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This commit adds instrumentation function to gather measurement
and statistical values similar to counter groups.
Multiple values can be stored per item, which can be retrieved in
FIFO order. Getting values from the item does not modify its state to
allow for multiple independant backends (e.g. VTY and statd).
When a new value is set, the oldest value gets silently overwritten.
Lost values are skipped when getting values from the item.
Sponsored-by: On-Waves ehf
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
The cast is not present in the original macro and was added
in OpenBSC with commit (be68f6fc6cde1367a4481d2e774a64e2cd657267)
to help using the header with C++. Due the recent issue in
the bssgp flow control code, let's remove the cast to see
wrong pointer usage. In case C++ code doesn't compile anymore
we can see how to add a cast back to it.
Change was proposed by Michael McTernan
|
|
|
|
|
|
|
|
| |
In OSX the int declares linkage and for libosmo-abis we have
two static asserts with the same name in two different compilation
units. When adding the "unused" attribute I removed the typedef.
I verified with a gcc 4.9.2 that no new warnings will be shown
when compiling libosmo-abis.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
On GNU __BYTE_ORDER and __LITTLE_ENDIAN is defined and
the include is "endian.h" on FreeBSD it is "sys/endian.h"
and LITTLE_ENDIAN/_LITTLE_ENDIAN and BYTE_ORDER/_BYTE_ORDER
is defined.
Create a header file that defines OSMO_IS_LITTLE_ENDIAN
and OSMO_IS_BIG_ENDIAN and can be used as
#if OSMO_IS_LITTLE_ENDIAN
do_little_endian
#elif OSMO_IS_BIG_ENDIAN
do_big_endian
#else
#error "Unknown endian"
#endif
|
|
|
|
|
| |
Submitted-by: altaf <altaf329@gmail.com>
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
|
| |
We can curse about CSN.1 and 3GPP but I think we want to call
it cursor here.
|
|
|
|
|
|
|
|
|
|
|
|
| |
We want to see from which category/subsystem a certain log message
is coming from and use a different timestamp format as well. Add
two new bitfields. This doesn't change the size of the structure
and on 32bit we still have 27bits left.
The extended timestamp will take preference over the current and
default timestamp format.
Fixes: SYS#602
|
|
|
|
|
| |
We want to use libosmocore/libosmovty in the GGSN sourcecode
and reserve a global region here.
|
|
|
|
|
| |
according to Holger, using AGPLv3+ at the time was a mistake and the
license should always have indicated GPLv2+.
|
| |
|
|
|
|
|
|
| |
This enables the persistent configuration of let's say
'log filter imsi 012345678' for a given log file in the config
file.
|
| |
|
|
|
|
| |
... and make libctrl code use it
|
|
|
|
|
| |
... and rather introduce a general osmo_talloc_replace_string() to
libosmocore.
|
|
|
|
|
| |
This used to be private in osmo-bts/common/abis.c, where it really
didn't belong.
|
| |
|
|
|
|
|
|
| |
ISO C forbids 'return' with expression, in function returning void
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
|
|
| |
Turns out we use this header in C++ code ...
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
|
| |
Submitted-by: Max <max.suraev@fairwaves.co>
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
|
| |
Submitted-by: Max <max.suraev@fairwaves.co>
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
| |
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
|
| |
Submitted-by: Max <max.suraev@fairwaves.co>
Signed-off-by: Sylvain Munaut <tnt@246tNt.com>
|
|
|
|
|
|
| |
Added some function for adding the unix domain socket support.
Signed-off-by: Alvaro Neira Ayuso <anayuso@sysmocom.de>
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently it takes 3s to establish a SAPI 3 SACCH connection with
osmo-bts. This is due to the fact, that a broken SABME request is
sent first and and is ignored by the MS. Then, after a T200 timeout
(2s) the SABME command is sent again (this time correctly) and
answered by the MS.
The first SABME message is broken (it has a length field of 3 and
ends with 3 bytes from the tail of the original RSL message),
because of it is expected throughout lapdm.c that msg buffers
containing RSL have msg->l2h == msg->data. Some abis input drivers
fulfill this but IPA doesn't, thus the 3 bytes of the IPA header
are still part of the msg and confuse length computation.
Since internal fields of the msg are modified directly, this is
difficult to see.
This patch adds a new function msgb_pull_to_l3() that explicitely
skips over all headers prepending L3 and therefore resets l1h and
l2h. This function is then used instead of msgb_pull_l2h() which
only worked correctly when msg->l2h == msg->data. In addition,
code manipulating msg->tail and msg->len directly has been replaced
by calls to msgb_trim().
Note that this patch does not fix all issues of this case in the LADP
related code.
Ticket: SYS#192
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
|
|
|
| |
This function works like osmo_hexdump() and returns a static buffer
containing hex bytes along with markers for the layers.
Note that it uses osmo_hexdump() internally, thus a call to
msgb_hexdump() invalidates the buffer that has been returned by an
earlier call to osmo_hexdump(). In short: don't mix them in a single
call printf().
Sponsored-by: On-Waves ehf
|
|
|
|
|
|
|
|
|
| |
The testcase didn't work on Ubuntu 12.04 because vty_create will
directly call vty_event (e.g. not through the plt). This means
that the approach to override vty_event in the testcase failed.
Use the signal interface of libosmocore and make the testcase
use it. The signals can be generally useful as well.
|
|
|
|
|
|
|
|
| |
This reverts commit bf991bc0f3006eddf45d494ee7e476a75dbaf36e and
fixes the bug instead. It's just sufficient to remove the defined()
test, since identifiers that aren't macros are just considered as
0 when used with #if. (see cpp info page, section 4.2.2 'If'
in cpp-4.6.info)
|
|
|
|
|
|
|
| |
The code started to use #if defined(...) but the value was always
for GCC/Clang. The only difference was that for older compilers
the value of the definition was 0. Conditionally define these
macros.
|
|
|
|
|
| |
Separate feature checking from implementation. Supported are clang and
gcc.
|
|
|
|
|
|
|
|
|
|
| |
Add a new core/defs.h file for definitions that can be used from
within header files without including prototypes and extern
declarations. It's primarily meant for macro definitions and basic
typedefs.
Move the definition of OSMO_DEPRECATED there and use compiler version
info to avoid compiler errors.
|
|
|
|
|
| |
this causes compiler warnings in user code ever since commit
55cf02221f0654a3f48888f3b13766b671a120f0
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The modified macro is still working on a gcc 4.7 and gcc 4.8 after the
removal of the typedef and addition of the unused attribute.
include/osmocom/core/msgb.h: In function ‘msgb_alloc_headroom’:
include/osmocom/core/utils.h:40:51: warning: typedef ‘dummyheadroom_bigger’ locally defined but not used [-Wunused-local-typedefs]
#define osmo_static_assert(exp, name) typedef int dummy##name [(exp) ? 1 : -1];
^
include/osmocom/core/msgb.h:386:2: note: in expansion of macro ‘osmo_static_assert’
osmo_static_assert(size > headroom, headroom_bigger);
^
|
|
|
|
|
|
|
|
|
| |
This is changing the semantic of the assert. The regression tests
now either need to check the stderr result, the exit status or print
a message when all tests are completed.
This is not that bad as the osmo_generate_backtrace is printing to
the stdout right now.
|
|
|
|
|
| |
This function will re-open all existing log files in the application,
praticularly useful for SIGHUP handlers in case of logrotate
|
|
|
|
|
| |
The tests should unconditionally assert, regardless of debug settings.
This uses the OSMO_ prefix as it's in the global namespace.
|
|
|
|
|
|
|
| |
The log target can be used via log alarms and show alarms.
Why? This feature was proposed/requested at
http://openbsc.osmocom.org/trac/wiki/Tasks/ErrorLogTarget
All messages use the same amount of space, prioritizing simplicity.
|
|
|
|
|
|
|
| |
I noticed some more issues and it is the easiest to revert and include
the fixed version.
This reverts commit 73377229bb33ab79682ce4b126a63602d13304ad.
|
|
|
|
|
|
|
| |
The log target can be used via log alarms and show alarms.
Why? This feature was proposed/requested at
http://openbsc.osmocom.org/trac/wiki/Tasks/ErrorLogTarget
All messages use the same amount of space, prioritizing simplicity.
|