It has been noticed, that after a list of contributions into lcr:
* 14e6fc8
* d8583d6
* 470fd5b
* aa8d3ed
the gateways matching when a search is being done based on
the IP address and the src_port (through the list of GWs with the same IP)
works, but not fully correct.
It is only related to the usage with the third parameter 'src_port',
when calling from_gw() and from_any_gw(), and, only when the search
is done through the gateways list, which have the same IP address
(but different ports). If gateways have different IPs, the issue
is not catchable.
The problem is in the algorithm used for matching based on
two objects (ip_addr and src_port) - the binary search.
It's not fully suitable for a search based on two (or more) objects.
The binary works completely fine, when only one object is used for searching,
but works not fully correct when the search is based on a comparison
of two (or more) objects. A division happening based on the value of
the first object, gives a chance that the second object will never
be found in the current half being looked up.
This commit concerns switching to a full cycling through the list of
gateways of the same lcr_id, and gives a proper work of the do_from_gw().
The slight drawback of the new method is that we have to do a cycling
through the whole list of GWs of the same lcr_id, but on the other hand
it is much more efficient than trying to build up a matching using binary
based on two objects (ip_addr and src_port) being used for comparison.
Change-Id: I8751adb5137a79e585c21b182319e7496412d81f
This commit deprecates:
* tm: previously added function t_append_branch_by_contact().
* tsilo: previously added functions ts_append_by_contact() / ts_append_by_contact_to()
TM. This commit introduces a possibility of TM's API to append a branch
based on a specific location (Contact's URI).
The function 't_append_branches()' has been extended, and is now able
to obtain a parameter Contact of str type and takes care of appending a branch only for it.
The way how the function acts now depends on if the contact parameter of str type
is given or not (empty or not). If the contact parameter is given, then only a desired location
is meant for appending. If not found in the location table, an append will not happen for this AOR.
Otherwise create branches for all existing location records
of this particular AOR. Search for locations is done in the location table.
TSILO. Add support of a lookup and branch creating by contact
Improvement of the TSILO module, which allows to run a lookup using the provided RURI,
but only to create a new branch for the Contact, which is set in the currently processed REGISTER.
Or optionally a Contact URI value given as a parameter to the function.
If the Contact URI for a lookup, is given as a parameter,
it is possible to provide it as a pseudo-variable.
Hence it is now possible to append new branch(es) for only UAC(s)
getting REGISTERED at the moment of calling script function 'ts_append_by_contact()'.
Previously it was only possible to create new branches
for all previously present location records (for this specific URI), which was in some circumstances undesired.
New script functions introduced: ts_append_by_contact(table, ruri [, contact])
New module functions introduced:
* w_ts_append_by_contact2() / ki_ts_append_by_contact()
* w_ts_append_by_contact3 / ki_ts_append_by_contact_uri()
Change-Id: Idd4775000f8e80999455680704c64a1208c67b3f
* skip_headers is now ignored when a branch's xavps are cloned,
this is needed to transparency that all headers that came with
the SIP message are avaialable via $x_hdr, wheres pv_headers
skips them when they are applied to the SIP message.
Change-Id: I742b373a3a0d38c755aa5ecfb8683e172a4d52f2
generate an uuid for PUBLISH id field instead of callid
https://github.com/kamailio/kamailio/issues/2906
Not exactly the same version from upstream since we
have changes in the module not yet merged
Change-Id: I50dd72f427b94da41c1d46236c05aa041b15f1cb
We need to rewrok TSILO module and partially TM's API,
so that it is able to run a lookup using the provided RURI,
but only create a new branch for the Contact,
which is set in the currently processed REGISTER.
Optionally add a possibility to provide a contact as a pseudo-variable.
Hence we will be able to append new branches for only UAC(s) comming
online now (being REGISTERED).
New script functions introduced, tsilo:
- ts_append_by_contact(table, ruri [, contact])
New functions introduced, tsilo:
- w_ts_append_by_contact2() / ki_ts_append_by_contact()
- w_ts_append_by_contact3 / ki_ts_append_by_contact_uri()
- ts_append_by_contact() / ts_append_by_contact_to()
New functions introduced, TM:
- t_append_branch_by_contact()
Change-Id: Id321a261b137c83bfa38a150c6ff67ca312ae61d
It has been noticed, that after a list of contributions into lcr.so of kamailio,
the from_gw() and from_any_gw() functions are not working fully correct.
However it is only related to the usage with the third parameter 'src_port'.
Last time (last commit into lcr.so), the back compatibility was provided not to break anything for the kamailio community,
this means, it's only related to usage with the 'src_port' parameter.
from_gw() and from_any_gw() functions are not working fine,
when we have tens of gateways with the same IP address in the 'lcr_gw', and you use src_port for searching as well.
It works fine, if you have 2-5 gateways, but not with tens of them.
This is all the matter of the implementation and some limitations of how
the iteration through the memory is done (memory allocated for a list of specific structures),
and binary search used for that.
This is our own Sipwise's patch to handle that case which is intended to be kept internally.
This patch introduces a more simple iteration, but it works more efficient,
when the search for two objects at a time for matching items is done.
Change-Id: I0e7fb2ea59edb7d22636e58f002ca9cb59d42315
Improve binary search in the lcr module and add a possibility
to do a matching not only based on an IP address of a GW, but also using a source port.
When a possibility to use 'src_port' parameter in from_gw() and from_any_gw() was introduced,
the bsearch() remained untouched, and hence the matching (iteration through existing GWs)
is now done only based on an IP address.
This leads to the issue, when there are more than one GW with the same IP address in gws table,
and from_gw() and from_any_gw() functions are used with the 'src_port' parameter,
it can happen that a wrong GW is picked out by bsearch() from gws table (lcr_gw) and
a check by from_gw() and from_any_gw() returns False.
Hence the matching based on IP address and source port is required for bsearch(),
when from_gw() and from_any_gw() functions are used with the 'src_port' parameter.
This means backwards compatibility is still present (when one uses functions without 'src_port').
Change-Id: I90327ef69d8132cc9e11522567cbc026e0808602
A simple change, which allows to check, if a coming request matches
not only by an IP address and a transport protocol,
but also using the source port of a request.
Improvements are done for the following functions of the lcr module:
- from_any_gw()
- from_gw()
Back compatibility provided, hence it does not affect already existing behavior.
Change-Id: Id0496fad75ce64f1abecaa4747b4a77ec6b87cbf
If we set Diversion in split_headers and we get a header like
> "RULTEST, normalaa" <sip:0046341355354340@whatever.net;user=phone>;reason=unconditional
There was false detection of two Diversion headers.
Skip split marker between double quotes to avoid this
Change-Id: Ib2189bc0029f244fd59f14df5b3d1df745271038
Since we use Redis SETs to simulate non-unique indexes, we can use these
SETs to perform a much faster table scan when an appropriate index
exists. For usrloc expiry query in a typical scenario, this brings the
list of key strings that need to be checked down from 170k to 19k.
Change-Id: I1c4841be5cd28ad42196b4da6b5bb3e6e1b6ed50
Instead of making the COUNT increase and falloff dependent on the number
of results produced, we make it dependent on the query run time. This is
to prevent Redis being blocked by extended consecutive SCAN queries.
Each individual query should not run longer than 100 ms, and we also add
a short pause before going to the next SCAN to give competing queries a
chance to run.
Change-Id: I6b88cacf795513cf3e0aeefc67fbedef8e1bb8c1
We have value-less master keys to keep track of entries. These keys are
useless when looking for fast-searching keys and are not better than a
full table scan, but are erroneously seen as a better key match than
range-matching keys. Therefore just ignore value-less master keys.
Change-Id: Ibc9ad87a42412b6cf21eac47fbe0ee8e39af4aff
(cherry picked from commit 79b4ac5bb75fe83914fddc104a8fc7c7ce2d0681)
* to control how stopper column is used.
0 => default behavior
1 => stopper will be checked after sort process
Change-Id: I2b59111e40a1aedbd6c5446a5906b53c687a8ad4
Use the avp value when creating P-Asserted-Identity instead of
values from Headers in order to allow to insert E164 format value.
If no value, skip creation of header.
Change-Id: I81282da27325d02987b3dc1079605de0d6b340a8
* if we use pv_get_spec_name() cfgt will throw errors trying to get values
like $x_hdr($xavp(whatever=>y)) when there's no $xavp(whatever=>y)
Change-Id: I88f91e30f969e309c566f6c15495796350b0ba5f
When AVP is used instead of value from header, we were missing
the call to parse_uri() so value at parsed_uri was empty
Change-Id: I37ce0626691eae4d313b39474d6d57cd8baf87dc
When a Redis key can be deleted directly, we still need to build the
list of type keys for that table in order to delete dependent entries,
which are the indexes used for faster DB access.
Change-Id: Ib8fdad428946263fc9eba466868347fee219dfbc
* __dialog_created() -> get_dialog_data() -> creates dlg callbacks for
__dialog_sendpublish()
__dialog_sendpublish() can be executed by two different process and
if we use refresh_pubruri_avps_flag shared memory can change in between
Change-Id: Id81b66209c27a900033f557cfa175ca3067aa34c
* remove pvh_remove_xavi()
* rework pvh_set_header():
use core pv_get_spec_*
use PV_IDX_NONE to detect no index case
remove values instead of adding NULL if PV_IDX_ALL
* changes needed from upstream:
https://github.com/kamailio/kamailio/pull/2341
Introducing PV_IDX_NONE so we can know when idx == 0
if $x_hdr(HEADER) or $x_hdr(HEADER[0])
So we can append or set the value
Change-Id: I496d65cc7e0e92167967edfb592fff6a6b0d6e3b
* for split_headers and for applying multi index
headers str_hash_case_get() is now used instead of str_hash_get()
to follow the case insensitive approach
Change-Id: I9c99f91e68836df593e0120596494829aab2363a
* header names are now case insenstitive when used to store
or retrieve values
* check for skip headers should be insensitive too
Change-Id: Ie0f5c45ce5ce24623c6233e342b646525c6081b9
* pv_headers same version from upstream
* move patches contributed upstream from sipwise to upstream
Change-Id: I7a484c589d1c28e869490e2737f94c9daf8bb9ff
Fails due to:
| Traceback (most recent call last):
| File "/usr/lib/x86_64-linux-gnu/kamailio/kamctl/dbtextdb/dbtextdb.py", line 1239, in <module>
| main(sys.argv)
| File "/usr/lib/x86_64-linux-gnu/kamailio/kamctl/dbtextdb/dbtextdb.py", line 1233, in main
| except (Error, e):
| NameError: name 'e' is not defined
Change-Id: I09ed6d0bd1b6fdad5f9b04cd54fef1ee43a44312
Victor Seva
Guillem Jover
Donat Zenichev
Kirill Solomko
Sipwise Jenkins Builder
Richard Fuchs
Michael Prokop