1872 /* first add those mentioned in the order list,
1873 * but only, if they were previously generated/added to the sdes_out */
>>> CID 1530346: Control flow issues (DEADCODE)
>>> Execution cannot reach the expression "NULL" inside this statement: "l = (cpq_order ? cpq_order-...".
1874 for (GList *l = cpq_order ? cpq_order->head : NULL; l; l = l->next)
Change-Id: I58f84ba82d215a9eb6cbd97548e8e9e8a954bdc6
This is a new option flag, which provides the ordered list,
in which to add crypto suites into the SDP body.
Right now they're always added in the order given in the source code.
Flag usage example:
`SDES-order:AES_256_CM_HMAC_SHA;AES_256_CM_HMAC_SHA1_32;AES_192_CM_HMAC_SHA1_80;`
This means — those listed SDES crypto suites will be added
into the generated SDP body at the top of crypto suites list, in the given order.
But, each of them is added, only if it is about to be added/generated.
In other words, the `SDES-order:` flag itself doesn't add crypto suites,
it just affects the order of those suites to be added.
And the rest of non-mentioned suites, which are also to be added,
will be appended after those given, in the free manner of ordering.
Important thing to remember - it doesn't change the crypto suite tag
for the recipient, even though changing the order of them.
Additionally.
This flag does not contradict with `SDES-nonew`, `SDES-only-` and `SDES-no-` flags.
It just orders the list of crypto suites already prepared to be sent out.
Change-Id: I0fec54f9e2f3cd4913e905e8afe825712f82d1ae
A new function dedicated to SDES crypto suites policy checks
has been introduced: 'crypto_params_sdes_check_limitations()'.
Use it to decrease an amount of repeating code blocks
related to SDES checks.
Change-Id: I0ac242a63107a9f3a41f95a57e3d3675645ac18d
Add a new flag to only accept these individual crypto suites
and none of the others.
For example, `SDES-only-NULL_HMAC_SHA1_32`
would only accept the crypto suite `NULL_HMAC_SHA1_32` for
the offer being generated.
This also takes precedence over the `SDES-no-` flag(s),
if used together, so the `SDES-no` will be not taken into account.
This has two effects:
- if a given crypto suite was present in a received offer,
it will be kept, so will be present in the outgoing offer; and
- if a given crypto suite was not present in the received offer,
it will be added to it. The rest, which is not mentioned,
will be dropped/not added.
Flag name: 'SDES-only-<crypto name>'
Additionally: add another new flag 'SDES-nonew'.
It will not add any new crypto suites into the offer.
It takes precedence over the `SDES-no` and `SDES-only` flags,
if used in combination.
Change-Id: Ic4fa03957ee3d4d24b0c4f3fd003eada05f49b0b
Unmark a monologue that has been scheduled for deletion when it's
associated with another one, which happens during offer/answer.
Fixes a regression from 53dbef7e1 which removed this logic from
__monologue_unkernelize with re-adding it elsewhere.
Change-Id: I037162e91fec42631680f7767f58b172fd6e04db
`call_get_monologue_new` is supposed to always return a full dialogue,
but an error in invocation (using a from-tag that doesn't belong to an
offer/answer) can lead to the second half being unset. Return an error
in this case.
Change-Id: I84b21ff5e5c0403fc07cae83fee206705ecff8b3
Use the new `associated_tags` table to determine which tags are
associated with which. Iterate the associations between tags in a
tree-like manner and do this at the moment the `delete` command is
received. Break up the `associated_tags` links at this time, and
determine which tags would be left dangling and mark all of these for
deletion. If no tags are left after this process, mark the entire call
for deletion.
The previous approach was cumbersome and prone to errors. Using tag
names and branch names to determine which tags are associated with which
is a pointless hurdle, and using a table of associations that is
explicitly kept for this purpose is a much cleaner approach. Also
postponing the decision about which tags to delete until the time the
deletion actually happens can lead to tags not being deleted, when they
really should be (e.g. A -> B, delete A, A -> C).
Change-Id: I03ae57d0a2117ecd721372c1a49468fc34dd630c
Keep track which tags (monologues) were created together as part of an
offer/answer exchange with a separate hash table, regardless of whether
these monologues actually have tagged names or are just nameless
branches.
Change-Id: I60aa114c8caf6ecdff4705e3399f60190d04dda6
During offer/answer, we remove existing subscriptions before
re-establishing the new set of subscriptions. But we must also remove
the subscriptions of previous subscribers in order to keep the lists of
subscriptions and subscribers correctly reciprocal.
Change-Id: I7d39fa59892159f41033ae6a40e7cb51d861b12e
Support multiple tone frequencies for DTMF-security=tone to enable
audibly distinguishing multiple consecutive DTMF events from one
another.
Change-Id: I6fa33a5768aae198220d0b0cc4c53308c5661a52
Based on the information gotten from Richard Fuchs
document the main objects in the code, to let the code be more
understandable for other code readers.
Mainly documented:
- call
- call_monologue
- call_subscription
- call_media
- packet_stream
- stream_fd
- sink_handler
- rtpe_callhash / rtpe_callhash_lock
Change-Id: I0cf122bea2d9c3f198b48da134a70301564ff1f9
Legacy protocols don't set sp->num_ports. Use a sensible default for
this case.
Probably a regression from 2d2d7665b
Change-Id: Idcbc477a68b6db70a91a5d082736ac642c50ab15
This makes it possible to have different media silencing options for
different outputs. Functionally this commit alone is a no-op.
Change-Id: I967c3e07ea4645bb49ccb76db12d51ded2d72f06
Create a dedicated struct to hold certain attributes shared by both sink
handlers and media subscriptions, as a preparation to simplify handling
these attributs.
Change-Id: I866159c33ed6d6a2873d2cf68c4906ea705d253e
If an offer is going to a side that is already known to support
rtcp-mux, don't try to offer a fallback RTCP port. Do this by pretending
that rtcp-mux=require is set, which leads to the correct behaviour. This
is needed because the RTCP fallback port might already be closed, so
trying to include it in the offer would require opening a new set of
ports, which is undesirable.
closes#1494
Change-Id: I550bec08379c799cb7dd090a70d090ae47462467
Keep a running lifetime total of all "gauge" type metrics. Also track
the square of the sums of all "gauge" type metrics in order to determine
the standard deviation.
Change-Id: I23f60774a6421636f1a913674c7d1b54a1c5f702
When doing the initial answer, the packet_stream endpoint port isn't
filled in yet. Use the stream_params port instead to test for rejected
streams.
closes#1499
Change-Id: I8f315d95521f874fb8c5e6222263d017800b5fc9
When ports are closed early (while the call is still running), we must
first update a slave rtpengine with this new information (that these
ports are now closed) before actually releasing the ports ourselves. Not
doing so leads to a race condition where the master instance re-uses a
port that was just closed before the slave instance knows about the port
being closed.
We implement this using a thread-local list to keep track of ports that
were released while processing a control message, and process this list
to actually close the ports only after Redis has been updated.
Additional calls to the function to close the ports are placed in
strategic locations to make sure this is triggered in every code path.
closes#1495
Change-Id: I803f4594f30ca315da0b84c6e76893f54ca3a7c9
This eliminates a spurious false warning log message for rejected
streams that use a dummy payload type
Change-Id: Id628cafb8d7c4ea576cd01ff35f5dd9cd2151280
Since we're creating a dummy sfd to hold the SRTCP context when we don't
have an actual RTCP port, we must make sure to remember and re-use this
dummy sfd during a re-invite. Otherwise we end up creating a duplicate
dummy sfd, which is detected as a different sfd and thus triggers an ICE
restart.
Change-Id: Iadc91e163bd15a3cd5f57656b52941724c920143
There's no need to open ports on non-primary interfaces if ICE is not in
use as these ports will not be used or seen by anyone.
This mostly obsoletes the `save-interface-ports` config option, with the
exception of ICE advertised by the offerer. We currently have no option
to reject ICE from the offerer during the offer phase, so ports would
always be opened on that side.
Relevant to #1164 and 001abe5
Change-Id: I43df70bc0ec49b81f63aec97c776e48617b2acfd
This enables the same behaviour towards the offerer when rtcp-mux=demux
or =accept is used, as we have towards the answerer when
rtcp-mux=require is used.
Change-Id: I56a1cea84efce0c2db1b58c500629d0e54d582f4
Special handling for codec lists that were received as part of an
answer: If the list includes a codec that was not offered, ignore that
codec. This prevents transcoders from being set up that were not
requested.
This brought to light some tests that were actually broken.
Change-Id: Iac71056ec5e10b5de5567917974f2c4e0261eb0c
If DTLS is rejected in an answer via `DTLS=off` we must forget that DTLS
was previously offered, as otherwise a re-invite would detect the
fingerprint as changed if the re-invite doesn't offer DTLS again. We
also make sure DTLS is shut down if during stream init DTLS is not
given, when it was present before.
Change-Id: I48ee6f0ec5ec02f558a6799951552ea2272d0e96
This is useful for functions which are used both from a timer and from
other callers. These functions would reset the logging context at their
end to free the reference held by the logging context, which would
wrongly reset the logging context when the same function was called from
a different code path. Using a stack with push/pop semantics makes it
safe to use these functions from any code path.
Additionally introduce an explicit reset function that clears the entire
stack regardless of context. This reset function is called at the end of
every work iteration in every worker thread, just in case not everything
was popped from the stack.
Change-Id: I0e2c142b95806b26473c65a882737e39d161d24d
commit a2e5cfb8e5
Author: Razvan Crainea <razvan@opensips.org>
Date: Thu Jan 13 16:16:19 2022 +0200
Add tests for subscribe requests on paused media
commit fa58596a9f
Author: Razvan Crainea <razvan@opensips.org>
Date: Wed Jan 12 22:01:27 2022 +0200
Swap media direction check for `subscribe request`
as @rfuchs mentioned in his review, the SEND/RECV media flags are set
according to rtpengine's perspective, not the media flow's one.
commit e1e9a157c0
Author: Razvan Crainea <razvan@opensips.org>
Date: Wed Jan 12 19:27:42 2022 +0200
Fix `subscribe request` SDP media direction
When building the SDP for a `subscribe request` command, take into
consideration the media direction of the source stream - if stream is
`recvonly`, then we do not have anything to send, thus the direction
should be advertised as `inactive`, rather than `sendonly`.
Change-Id: I2d78bbec8ad584774f3c90f0ce5cca42f57f7b0f
With multiple media subscriptions, codec handlers are called
consecutively, once for each forwarding chain, leading to DTMF events
reported multiple times. The DTMF trigger must therefore keep track of
the state in the upper media object, not in the codec handlers.
Change-Id: I9ceaf406e093f25b7c037a325a0f2a7a91954922
This makes it possible to add new streams without specifying the
direction/interface again.
Reported in #1366
Change-Id: I8f320ecbe72f123d755ba80370de9c40960eb0f0
While doing the A/B reassociation during an offer/answer exchange, we
don't (necessarily) want to remove all existing subscriptions. Instead
we cant to unsubscribe all subscribers so we don't do media forking, but
leaving existing subscriptions alone to make early media reception
possible. This mirros the old behaviour.
Change-Id: Ib9e6671ca2d23d1eb4509d7cf939015c816cc622
Also add a safeguard against filling the remote peer address with an
address from the wrong family
closes#1305
Change-Id: Iac18212b4d526a2f7d49a06ddcd724aa89b06060
The contents of the ->next element cannot be accessed completely lock
free as they're zeroed out during call removal. Instead grab a reference
to the linked next call before releasing the lock, and also lock the
next element before moving on. This requires a more granular locking as
not to interfere with call removal: One lock to protect the contained
call and the ->next, and another to protect the ->prev
Change-Id: I5474ea3f88e3276f93ba62a952b3be13c0c182e9
commit b0c722da69ad088a2eddced12b37c0546a514890
Author: Daniel Hauptmann <dhauptmann@sipwise.com>
Date: Mon Jul 26 15:35:51 2021 +0200
changed flag bit length
in call_interfaces.h changed bit length of reuse_codec from 0 to 1
commit 0313a747532d5987f25fa9edb202aa460bf98dd1
Author: Daniel Hauptmann <dhauptmann@sipwise.com>
Date: Mon Jul 26 15:29:20 2021 +0200
inversed reuse_codec logic
in test-transcode.c and call.c, reuse_codec = 0 (default) will now result in using codec_store_populate instead of codec_store_populate_reuse
commit b876bd686bd30df21a5962aca16fc1c85574f554
Author: Daniel Hauptmann <dhauptmann@sipwise.com>
Date: Mon Jul 26 15:18:19 2021 +0200
adding option to minimalize changes in the codec_store_population
added function codec_store_populate_reuse in codec.c which replaces codec_store_populate but makes fewer changes to the GLists with the old and new codecs
added flag to enable this feature (disabled by default)
commit 6fd0b701c9589b2fae00300801e02a9b5cc397ab
Author: Daniel Hauptmann <dhauptmann@sipwise.com>
Date: Mon Jul 26 14:44:42 2021 +0200
Added Option to minimize change in the codecs
In codec.c added function to populate codec store with the fewest changes between the old and new GList which contains the codecs.
Added new testroutine in test-transcode.c line 1500
Added flag to call_interfaces.h to optionally enable this feature
Change-Id: If58d9a07d114b05dfb75553a87eb4372ae949fbb
commit 3bf554a8fbae7e948343699f40d935693618b764
Author: Daniel Hauptmann <dhauptmann@sipwise.com>
Date: Fri Jul 23 13:58:02 2021 +0200
changing codec-exchange behaviour
in codec.c line 3288 function codec_store_populate now doesnt empty dst and copy new codec from src to it, instead codecs from src will be appended to dst and codec from dst, which are not
being contained by src are being removed
Change-Id: Id6b7ee65595f9cc5c71ef557c7bac5ee38f97cbe
Complete overhaul of the codec handling code:
*) obsolete flags `asymmetric codecs`, `symmetric codecs`, `reorder
codecs`
*) support proper codec offer/answer
*) split codec manipulation (strip/offer/accept/etc) into separate
functions for clarity and better code maintenance
*) fully update codec handlers in both directions after an answer
*) explicit allocation and handling of codecs and payload types in a
codec_store object
*) improve codec matchup logic during answer
*) more explicit handling of supplemental codecs (CN/DTMF)
*) remove now obsolete hacks for handling certain use cases
Change-Id: I996705ba8fe339524c2f70e6bb0fd854f9a1f4fb
This solves problems when the same SSRC is looped through the same call
multiple times in different mono/dialogues, with different parameters.
Change-Id: I1d033cb1f012574d82b5bcbfffe11eb5f983cfd8
The JSON context is already in an object, therefore skip the surrounding
{} enclosure.
Also don't run the global MQTT timer when MQTT is disabled.
closes#1290
Change-Id: I63c622bd339545e625ee18def33f21de2533f949
This makes the type in line with string(3) functions and eliminates some
compiler warnings.
Also update the related bencode data type.
Change-Id: I7ef4024f4b5a0f737b3dbe03bcd078032395bce6
If we receive an SDP with a DTLS fingerprint, by default we adopt the
hash function used for that fingerprint in subsequent communication with
that peer. However, if the SDP is an answer, and we previously used a
different hash function in the offer towards that peer, then a later
re-invite offer would be sent with a different fingerprint, causing an
unexpected DTLS restart. Instead, make sure we don't change fingerprints
if one was already sent.
Change-Id: I603bb86ce2d7121556c161749ed08128dd0b63b2
Avoid using an explicit test for the trickle ICE endpoint address and
instead use the TRICKLE_ICE flag.
Change-Id: I821ced7e77c798004c9d88ec072848b7f6e24a85
When set to `false`, no changes at all. (default)
When set to `true`, bind only one desired family local address.
Also add info in rtpengine.pod file.
Also add log for sfd with no call.
If a supplemental codec was removed by the codec tracker, we must reset
any codec handlers on the opposite side that were using this output
codec.
Change-Id: I7fb9d0ec7a061682fe3b28d9ead14ac263aff7c3
add explicit flag for fingerprint length instead of using the hash
function
restart DTLS after a shutdown due to a reset
clear fingerprint buffer when running the hash
suppress stray log message when no fingerprint was seen yet
add option to test script
closes#1095
Change-Id: Ic58ea6c5aa48a215743d6a3c1b371fc4a5ea77ce
An endpoint switching protocols is normally passed through to the peer,
but in an answer that is usually not desirable. Change the default
behaviour to stick to the original protocol that was offered even if the
answerer changes protocols.
Change-Id: Ib288549f4b1c9ab57a6333c6b7dd511537af96f7
Initially created branches don't have a tag associated with them, so
they don't appear in `other_tags`. We need to keep track of which
branches were created from a single from-tag through their via-branch
values so we know when all branches have been deleted.
closes#1037
Change-Id: I4c6ce602b99a9104de98098cd06ca399eb28e1b2
1) In an SRTP re-invite offer, fixes not a full set of crypto suites
being present
2) In a re-invite offer that switches from RTP to SRTP, fixes SRTP not
being initialised at all
Change-Id: I911442d2cba17ecf6af482cfe922d4e9db2eda8d
In case of an offer with a via-branch followed by a delete without a
via-branch (cancelled call), the call erroneously remains open after
deleting one half of the call. The reason is that un-answered branches
do not appear in the `other_tags` list and so are left out from being
checked.
Change-Id: Ib008f32ef5ee06a7ca997c900c9a3adc85b0f10d
fix cleanup being skipped on redis slaves
fixes an SDES related Redis mem leak
adds a hash for the ports free list to avoid duplicate entries
fixes#898
Change-Id: I34aad67290ff5ef8824142682aac03cb600d0ecb
Untagged branches (only known by via-branch) don't appear in the `tags`
hash and don't have any `other_tags`. The logic to determine whether the
call is completely finished therefore must take these into account
separately.
We take care to remove destroyed monologues from the `viabranches` hash
to keep the count up to date, and determine whether a call is terminated
when there is no more than 1 tagged monologue left and no untagged
monologues.
Closes#875
Change-Id: I6b9618d598f4a95970cd2a452f06574423932b09
If B accepts a crypto suite that was not listed as the first, in order
to support SRTP passthrough, we correctly answer to A with only that one
crypto suite. But we must also remove all other crypto suites from our
list of supported crypto suites internally, because we use the first one
to init our crypto contexts.
fixes#829
Change-Id: Id07343d7b24648208e3a4b4e0b246949dce0385e
Move the RTCP_MUX_OVERRIDE flag to the opposite side of the dialogue to
preserve options given during a branched call.
closes#793
Change-Id: I0bd7621ba22fbfe4f41d115ec2e5dab65283ae01
Every packet_stream gets a send_timer allocated, but the teardown
routine skips the refcount decrease for fallback RTCP streams (when
rtcp-mux is in use), resulting in mem leak
fixes#753
Change-Id: Ib3a4ef8a81135918f08e28e127e4bb557b8ea05d
Adds a new option ptime-reverse to complement setting of ptime towards
the offerer. This and ptime setting are now ignored in answers.
Change-Id: Icbc04f191cbc194b75b72a97832fcaba58feb10e
When we receive an incoming SDES parameter, we must match them against the
previously sent outgoing SDES parameters, choose the one that matches
what we just received and eliminate all others. This is a no-op if none
were sent previously (original offer).
Issue only appears in a re-invite when the first offered crypto suite is
accepted.
fixes#631
Change-Id: I4991d0aaf0b29c1ba66045ed0e5281fc18c8af2e
We should offer all crypto suites that we support. If passing through
SDES, we should amend the list of crypto suites with all additional ones
that we support that weren't included in the received offer.
closes#577
Change-Id: I9b6c16e8eadecf01cdbc8043bd8361e0f683e456
The redis onekey concepts is introduced to reduce traffic to redis
and redis notification traffic.
It modifies the current structure for one call in redis, which are
multiple keys with pre- and postfixes and the callid in between to
one key with the structure "json-<callid>". The value is a json
formatted string with the previous multi-key identifiers in it.
instead of outright rejecting local endpoints advertised by remote
clients (to allow for deliberately daisy chaining packet streams), we
flag them for checking against actual packet loops as we did before
Change-Id: I5652e86e12f3c1c5053ea70b01e8d128ebf47751
fixes#65
this also obsoletes the old loop detection
Change-Id: I850d81500c45828af2c4d50d80278ec2d599c2a0
(cherry picked from commit 3254278cfd55167fb881cc665328744183773728)
We had initialization code for recording scattered through
"call_interfaces.c", "call.c", and "recording.c". I moved more of the
actual code into functions within recording.c under the parent function
`detect_setup_recording`. We call this function from "call_interfaces.c".
I moved the disjointed bit of PCAP initialization to occur right below
where we toggle recording on or off.
We want to be able to associate call files with a call without the
presence of identifying metadata within the metadata file. To accomplish
this, we prepend the call-id to the start of the pcap recording files
and the call metadata files.
Even though call-id is supposed to be unique, because of paranoia we
keep some of the random affix hex string, but we reduced it down to an
8-byte random value.
Also, some minor argument ordering and name refactoring for random
string generation functions.
We used to sometimes free the generic metadata (passed in through
rtpengine commands) before writing it to disk. Then we were writing
blank metadata to our metadata files. We fixed the ordering of
our `free` operations.
This involved moving all code from fs.(c|h) to recording.(c|h).
We still spoof packets, so the UDP will look like all monologues are coming
over the same port and will probably look like they are all one stream if
you look at the PCAP file.
Command line option is "--recording-dir".
Renamed inner recording spool "recordings" to "pcaps".
This is to avoid name sharing conflicts with the "--recording-dir" command
line option, which specifies the recordings spool directory, and the
"$RECORDING_DIR/recordings" inner directory. Changing the inner directory
name to "pcaps" removes this name collision.
In the process, I changed the function names in fs.h to be consistent with
other functions. The names are structure like "$OBJECT_$VERB".
Fixes and changes:
- Only create the metadata file if the call is being recorded.
- Only write to the metadata file if we actually created it (NULL check).
- Make sure we have metadata before putting it on the call object
- Correctly overwrite recording metadata without leaking memory
- Set the no kernalization flag per call instead of for *every* packet.
- Logging cleanup.
File system code is now in fs.{h|c}. This includes:
- spool directory setup
- metadata file management
- pcap file creation and writing
Random hex string generation is now in str.h.
We create a metadata file for each call. The metadata files will all end up
in a spool directory for the rtpengine.
Each in-progress file has the format: "rtpengine-meta-$RANDHEX.tmp" and
goes in /tmp/. When a call finishes, it is moved to the spool directory
in sub-directory /var/spool/rtpengine/metadata/ and we change the file
extension to ".txt".
The metadata file contains references to all PCAP recording files associated
with a call, and it includes generic metadata at the tail of the file.
One absolute path for a PCAP file per line, followed by two blank lines,
followed by the metadata passed in to the rtpengine through an external
command.
RTP Engine checks for the spool directory "/var/spool/rtpengine" on startup.
If it's not there, it fails. If it's there, it sets up "metadata" and
"recordings" inner directories. This is where RTP Engine will write call
metadata files and PCAP files.
Creating a random filename with a prefix and a suffix is now done through a
generic function. We use this to create pcap recording files in the /tmp/
directory, and will soon use it to create metadata files.
RTP Engine creates PCAP files for recorded calls on offer answer instead
of initial offer.
We make up bogus values for the nonessential parts of the PCAP, UDP, and
IP headers. We might be able to pull these from other parts of the RTP
Engine, but that information was unnecessary for recording calls so they
can be recorded to audio files.
If you change the packet headers, be really careful about byte order and
datatype size!
Pass in "record-call" flag over `rtpengine_offer` or `rtpengine_answer`
message. RTP Engine tracks files used to record pcaps and send them back
in the response message.
Pipes call audio (unencrypted from both ends) to recording files.
Sets up file descriptors for local files to dump RTP recordings.
A file and a file descriptor per monologue in a call.
Recorded streams will be running in user daemon mode, not in kernel mode.
This removes first 12 octets from packet to record just the rtp.
- add --subscribe_keyspace list config parameter.
- don't delete foreign calls by timers
- fix synchronization of foreign calls (use a separate redis_notify database)
- fix statistics for control channel calls.
- fix deletion of foreign calls upon del notifications
- update rtpengine-ctl tool
Removes the explicit redis-read-db configuration and reduces the option
to one redis DB and one redis write DB. If only the redis DB is
configured, then it will be used for all operations. If both are
configured, then the redis DB will be used for reading and the write DB
will be used for writing (updates).
Change-Id: I8d5a32c53c9416b514c98d69c3afe7c547e530ad
The session limit is only for calls an rtpengine is responsible for.
Foreign calls (coming in via redis notification) are not counted as
long as the rtpengine is not responsible for those calls.
At least that means that the limit may exceed if the calls the rtpengine
is responsible for plus the former foreign calls are greater than the limit.
This will happen suddenly when the rtpengine becomes responsible for the
foreign calls.
supports aliasing a local interface multiple times with the same local
address for different advertised addresses
closes#216
Change-Id: I6f98d1a17290b0bb1831e48ad89fc61d8b2d7914
Thoughts on that topic so far:
There's one thing to keep also in mind. What do we do if the call
changes (streams) and the backup node is notified ? Currently we only
know (by subscribing to the 'notifier-' prefix that in fact it has
changed, but we don't know what has changed in detail. Subscribing to
everything would lead to the problem that we have to take care about
synchronising the the new streams with the old ones. Without having a
look at the code that might be a lot of effort and ... I guess that's
why richard likes to ... have clean states of the calls. Synchronizing
is always a mess. Easier to delete and setup new.
I thought about the following solution for makes things more abstract
and easier to understand:
1. Whenever that call on a backup node (foreign call) has seen a packet
(also timers are started then), we do not process any notification by a
redis notification for this call (caus' the RTP-IP has already been
switched). More precise and abstract that means: If a node has taken
over the responsibility for that call (by having seen a packet), it
assumes that notifications from the former node to be dropped. The call
will be deleted when either the timer fires or (for other companies) the
control channel address has also been switched and via that channel the
call is deleted.
2. If a call has not seen a packet (inactive call on the backup node,
seen as not responsible for that call but could become so), we accept
all commands for that call on the backup node in the same way as on the
original node. That means also deletions in-between and so on. I mean if
the original node does it, why not accepting to do the same way on the
backup node ?
Currently, every rtpengine will subscribe to redis-keyspace notification
so it will receive a notification when an call is inserted. If the call
is not already handeled by the rtpengine, the call will be restored.
The reason for this is to have in-place redundancy. Imagine you have
multiple rtpengines running, eachone will have all calls of the others.
When one rtpengine fails somehow, infrastructure guys use BGP in order to
'move' the IP address from one rtpengine to another. Thisone can handle
the new calls instantly since they're already recovered by
redis-notification feature.
Next step is internally identify those calls in order to prevent some
timers to delete the calls where no RTP flows. Second will be
something we call 'partitioning'. It means that the subscription
to a redis notify will only be for the keyspace a dedicated rtpengine
writes to. This leads to the point that you can make redundancy groups
(partitions) of the rtpengines.
This brings master up to date with branch `rfuchs/socket-rework` at
commit `b1bcc096b7`. The branches have diverged too much for a proper
merge, so this is a manual (squashed) merge.
The old master before this merge can be found in branch
`old-master-before-socket-rework` (commit `82199216b2`).
This is a complete rewrite of all socket handling routines. The most
important functional change is that sockets aren't indiscriminately
bound to INADDR_ANY (or rather in6addr_any), but instead are always
bound to their respective local interface address and with the correct
address family.
Side effects of this are that in multi-homed environments, multiple
sockets must be opened (one per interface address and family) which must
be taken into account when considering RLIMIT_NOFILE values. As a
benefit, this change allows rtpengine to utilize the full UDP port space
per interface address, instead of just one port space per machine.
The socket abstraction also makes it possible to support RTP over TCP in
the future.
Change-Id: If6cf4f42136229490186d2d2482fb4fc140c2b53
This enables switching over the endpoint of a dialogue to a new client
with potentially different capabilities
Change-Id: I973b3c5d004014cc73e6ebc3e87d03c742bc1951
(cherry picked from commit 69b7586aa4)
(cherry picked from commit 93332f0f6e)
Avoid buffer overflows.
Also make cdrbuffend -= 1; append spaces to the TRUNCATED message because
syslog trims the output to 8205 chars on a line for big log buffers.
Donat Zenichev
Richard Fuchs
John Burke
Stefan Mititelu
Daniel Hauptmann
Guillem Jover
Julien Chavanton
Julien Chavanton
balajeesv
Semen Darienko
Balajee SV
Claudiu Boriga
Michael Prokop
Frederic-Philippe Metz
Lucian Balaceanu
Pawel Kuzak
Dylan Mikus
Eric Green
Stefan Mititelu
Lucian Balaceanu