Fixes an inconsequential mem leak (mediator would shut down after
logging the error anyway)
Change-Id: I6d3bdbabc70d2fbae063490479da7d9722cc08df
(cherry picked from commit 8388782643)
To allow mediator to continue to run after an SQL insert failure, retry
the inserts one by one after the bulk insert failed. This makes it
possible to 1) narrow down which row caused the error and make this
clearly visible in the log, and 2) allow for a small number of failed
inserts that can be ignored. The latter is needed to allow for broken
acc records that need to be discarded anyway.
Change-Id: Idf233cac3c05be108abbba061f4661255525c468
(cherry picked from commit 84f44fa09d)
Use a queue to keep track of the position of each CDR entry as it is
added into the SQL string. This allows us to inspect each entry
individually if the insert goes wrong.
Change-Id: I2f95e35030c3f55f8ccd3efb3ef20684f301af2e
(cherry picked from commit 8a3226759f)
All `medmysql_str` except CDRs have an associated queue, kept separately
in `medmysql_cdr_batch`. Move it directly into `medmysql_str` to make
things easier.
Change-Id: Idd6d31596e5e121a44c9b7922be66c24e7a43518
(cherry picked from commit f345a86efb)
This makes it possible to trace which CDR record was created from which
acc records and at which point.
Change-Id: I3645ccf244bf7d86b6f70181c57e47dfa204f7b9
(cherry picked from commit 64d50ee701)
We cannot overwrite the reallocated variable, otherwise on failure we
leak it.
Change-Id: I541537ade71086c9ceb851d8c5223364e3908b07
Warned-by: cppcheck (memleakOnRealloc)
While the code is not threaded, better be future-proof and use safer
APIs, which in addition has less side-effects as it does not set the
global TZ related variables.
Change-Id: I57fbe6683590cc2fbc0da508819553c040764739
Warned-by: lgtm
Whether a bit-field is signed or unsigned is implementation specific, so
we should be explicit about this.
Change-Id: I8bce847d6ece3d08fc93fa77227f6f970d1353de
Warned-by: lgtm
On Debian buster the latter contains transitive dependencies in Libs,
instead of in Libs.private, which leaks linking implementation details.
Change-Id: Ia131fe82444b8c2204976df75be047ab6d17c561
We need readonly access to /dev/urandom, otherwise the rand()
calls might fail, as seen in kamailio-config-tests.
Change-Id: Id132191994ae5fe74ec3ebb7d34a4a5d50769dbc
Thanks: Victor Seva for reporting
In carrier environments we have foreign DB hosts
configured in /etc/ngcp-mediator/ngcp-mediator.conf,
therefore we can't easily apply IP address filtering.
JFTR, new and current systemd hardening state for ngcp-mediator:
| $ sudo COLUMNS=142 systemd-analyze security ngcp-mediator | grep -v '✓'
| NAME DESCRIPTION EXPOSURE
| ✗ PrivateNetwork= Service has access to the host's … 0.5
| ✗ User=/DynamicUser= Service runs as root user 0.4
| ✗ RestrictAddressFamilies=~AF_(INET… Service may allocate Internet soc… 0.3
| ✗ RestrictAddressFamilies=~… Service may allocate exotic socke… 0.3
| ✗ DeviceAllow= Service has a device ACL with som… 0.1
| ✗ IPAddressDeny= Service does not define an IP add… 0.2
| ✗ RestrictAddressFamilies=~AF_PACKET Service may allocate packet socke… 0.2
| ✗ SystemCallFilter=~@privileged System call whitelist defined for… 0.2
| ✗ RestrictAddressFamilies=~AF_NETLI… Service may allocate netlink sock… 0.1
| ✗ RootDirectory=/RootImage= Service runs within the host's ro… 0.1
| SupplementaryGroups= Service runs as root, option does…
| RemoveIPC= Service runs as root, option does…
| ✗ RestrictAddressFamilies=~AF_UNIX Service may allocate local sockets 0.1
|
| → Overall exposure level for ngcp-mediator.service: 2.1 OK 🙂
Change-Id: I0e7c474eddd5d4d4c77b9bda157448294ed0a5c4
ngcp-mediator service state BEFORE this change:
| $ systemd-analyze security ngcp-mediator | tail -1
| → Overall exposure level for ngcp-mediator.service: 9.5 UNSAFE 😨
ngcp-mediator service state AFTER this change:
| $ sudo COLUMNS=142 systemd-analyze security ngcp-mediator | grep -v '✓'
| NAME DESCRIPTION EXPOSURE
| ✗ PrivateNetwork= Service has access to the host's network 0.5
| ✗ User=/DynamicUser= Service runs as root user 0.4
| ✗ RestrictAddressFamilies=~AF_(INET|INET6) Service may allocate Internet sockets 0.3
| ✗ RestrictAddressFamilies=~… Service may allocate exotic sockets 0.3
| ✗ DeviceAllow= Service has a device ACL with some special devices 0.1
| ✗ IPAddressDeny= Service defines IP address whitelits with only localhost entries 0.1
| ✗ RestrictAddressFamilies=~AF_PACKET Service may allocate packet sockets 0.2
| ✗ SystemCallFilter=~@privileged System call whitelist defined for service, and @privileged is included 0.2
| ✗ RestrictAddressFamilies=~AF_NETLINK Service may allocate netlink sockets 0.1
| ✗ RootDirectory=/RootImage= Service runs within the host's root directory 0.1
| SupplementaryGroups= Service runs as root, option does not matter
| RemoveIPC= Service runs as root, option does not apply
| ✗ RestrictAddressFamilies=~AF_UNIX Service may allocate local sockets 0.1
|
| → Overall exposure level for ngcp-mediator.service: 2.0 OK 🙂
Change-Id: If33c303b9df465393f15c0b685d1aab54a465df6
We only ever go into the error handling code before malloc has
succeeded, which means ‘entries’ will always be NULL.
Change-Id: Iaedd1ac668249b0e4516f782c24a2f420e8682a5
Warned-by: coverity
While MariaDB uses my_bool here, MySQL uses the standard bool type. Both
are compatible, so use the more standard one.
Change-Id: I648bc8eb55f7617f6b4eb796a33bc77ce15ca34b
We need to respect user-controlled variables and only append to them
whatever we require for the build.
Change-Id: Ie86415863f806e2c76b6a47929c65aafc0a2d177
With the increased length of src/dst_leg strings, we must also increase
the buffer size for the insert string.
Did I mention that we should really move to dynamic strings here...
Change-Id: I0e77155d36f434055e36124d37bf699179f887b1
This is relevant for records coming from Redis, as the insertion into
the trash and backup tables is done from the strings kept in memory.
Change-Id: I1398f7bfc0d4a0c2da8b8104dc4589ea9d983da7
(cherry picked from commit 6e957e69b7)
/var/lock was deprecated with Debian/wheezy (see
https://wiki.debian.org/ReleaseGoals/RunDirectory), so
/var/lock/mediator.lock is actually /run/lock/mediator.lock.
Reflect this in the definition of MEDIATOR_LOCK_FILE.
Change-Id: If50a5fdcfafb289d59ae828f750388e103b5b3ad
Due to the movement of call transfer handling from sems pbx to
standard sems, in case of blind call transfer the BYE ACC is
generated with call_id suffix '_pbx-1_xfer-1' instead of '_xfer-1'.
To be more precise, currently we creates 2 ACCs for INVITEs:
- one without any suffix
- one with '_xfer-1" suffix
and just one for the BYE:
- with '_xfer-1' suffix
Mediator is able to use the BYE to close both the INVITEs
With the new implementation we creats 2 ACCs for INVITEs:
- one without any suffix
- one with '_pbx-1_xfer-1" suffix
and just one for the BYE:
- with '_pbx-1_xfer-1' suffix
The sinlge BYE record is not used to close both the ACC INVITEs.
Change-Id: Id7a68fab8d84ffe987084e925fecc1e051cbaccd
Strings, in particular call IDs, that are coming from Redis are not
necessarily valid UTF-8 since Redis doesn't really do UTF-8. The MySQL
DB schema expects strings in valid UTF-8 however (even though call IDs
really should be raw binary strings), resulting in an error if an
UTF-8 invalid string is attempted to be inserted.
Solve this by verifying each string's UTF-8 correctness before inserting
it, and forcing it to be interpreted as the 100% permissive latin1
charset if it isn't.
Change-Id: I5a218083bc4e3d7a47d1f77911f7ef9a88ce9bd2
Sipwise Jenkins Builder
Richard Fuchs
Guillem Jover
Alessio Garzi
Michael Prokop
Marco Capetta