If IPv6 is disabled, then it is useless to try to resolve
IPv6 address because "struct sockaddr" does not have enough
space to store IPv6 address.
Fixes#1487
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This is useful to enable error/warning logging across the net
codebase (less useful for debug level logging, but that's true
for CONFIG_NET_LOG_GLOBAL already).
Implementation-wise, instead of keeping adding to long list of
"select"'s in CONFIG_NET_LOG_GLOBAL and thus introduce component
inter-dependencies, add "default y if NET_LOG_GLOBAL" to
individual components' logging options.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
This is the first part of a large refactoring of LwM2M library
message functions and will simplify observer handling later.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
All throughout the LwM2M library we use sockaddr values which are
basically the same as the net_app_ctx's remote addr. There's no
reason to keep these extra sockaddr values around. The net_app
framework client won't accept incoming requests on sockaddr other
than the one we're connected to.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
This is the final stage of moving the LwM2M library internals to
the net_app APIs. This means we can support DTLS and other
built-in features in the future. All of the logic for
establishing the network connection is removed from the sample
app.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
In preparation for the move to net_app APIs, we will need
to pass net_app_ctx structures around to the following
functions:
lwm2m_udp_sendto()
udp_request_handler()
Let's add the parameter as net_context for now so the
transition will be smoother later.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
This allows use to associate easily the replies / pending operations
with a specific network connection.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
The LwM2M library does not use net_app APIs internally. To help
this effort let's establish a user facing structure "lwm2m_ctx"
(similar to http_client_ctx and mqtt_ctx) and start it off by
wrappering the net_context structure.
Future patches will add user setup options to this structure and
eventually remove the net_context structure in favor of a net_app_ctx.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
This implements mDNS client from RFC 6762. What this means that
caller is able to resolve "hostname.local" names using multicast DNS.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Errors has been ignored when using TLV writer to create/write object
instance/resources. Modify to propagate the error back to the caller.
To reproduce the issue, try to create IPSO light control object
instances twice. Since the default instance count is 1, the second one
should be rejected and responded w/ error. But the current
implementation will respond w/ 2.04.
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
We did not check the requested object/object instance/resource exists or
not before we adding an observer. Correct it by checking the existence
first.
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
We should stop sending out notification to the peer when the
object/object instances requested to be observed is removed
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
We were using sys_slist_remove() to remove object, object instance and
observer w/o passing the previous sys_snode_t to it (NULL).
This will instruct the function to treat the node as the list head and
result in unexpected behavior after the removal.
Correct it by using sys_slist_find_and_remove() or passing the previous
node to the function.
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
When a request demands to create a new object instance, it will search
whether the request object instance exists or not. However, current
implementation does not reset the lwm2m_engine_obj_inst at the time it
is deleted. It only removes the object instance from the sys list.
Correct the behavior by resetting both object instance and resource
instances at the time it's deleted. Also, consolidate function
lwm2m_delete_handler() and lwm2m_delete_obj_inst().
To reproduce the issue, try to create light control object instance
(/3301), delete the created instance and create it again. You shall find
following error message dumped.
> [ipso_light_control] [ERR] light_control_create: Can not create
instance - already existing: 0
> [lib/lwm2m_engine] [ERR] lwm2m_create_obj_inst: unable to create obj -
3311 instance 0
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
The commit 725be227 ("net/mgmt/event: Commands must be > 0 so
masking them works") prevented IPv6 address setting when an
application was initialized. The check in subsys/net/lib/app/init.c
needs to be adjusted because of that change.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
When new socket context is created on accepting connection to a
listening socket, its recv_q FIFO should be initialized. Without
initialization, this worked by a chance when FIFO structure was
simple, but recent change to add dlist to it (which now needs
proper initialization) exposed this issue.
Jira: ZEP-2576
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
If the query name is already in numeric format, there is no
need to send the query string to DNS server as we can just
convert it ourselves.
Jira: ZEP-2562
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
This commit removes IP address parsing from DNS init and
replaces it by call to net_ipaddr_parse().
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
TLS and DTLS are not related to each other so allow DTLS to be
enabled even if TLS is disabled.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The IP header was stripped by _net_app_ssl_mux() when it received
IP packet. This is fine but if the application expects the get
the IP header, then there is a problem. Fix this by saving IP
header to ssl_context and then putting it back in front of the
packet when the data is passed to application.
Note that this IP header is not used by net_app when the packet
is sent because TLS/DTLS creates a tunnel for transferring packets
and user can only sent packets via this tunnel.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
k_delayed_work_cancel now only fail if it hasn't been submitted which
means it is not in use anyway so it safe to reset its data regardless
of its return.
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
The original commit 8ebaf29927 ("net: http: dont timeout
on HTTP requests w/o body") was intended to handle a case
where an HTTP response had been retrieved from the server but
the HTTP parser couldn't meet the criteria for calling
"on_message_complete". For example, a POST to a REST API
where the server doesn't return anything but an HTTP
status code.
It was a really bad idea to check a semaphore count. There
is a lot of kernel logic built into semaphores and how the
count is adjusted. The assumption that the value is 0
after the k_sem_give() is incorrect. It's STILL 0 if
something is pending with a k_sem_take(). By the time
k_sem_give() is done executing the other thread has now
been kicked and the count is back to 0.
This caused the original check to always pass and in turn
breakage was noticed in the http_client sample.
Let's do this the right way by setting a flag when
on_message_complete is called and if that flag is not set
by the time we reach recv_cb, let's give back the semaphore
to avoid a timeout.
Jira: ZEP-2561
Signed-off-by: Michael Scott <michael.scott@linaro.org>
The HTTP header field pointers are saved for each HTTP request.
But the counter that saves the pointers was never reset to initial
value when the connection was dropped. This meant that the header
field values were only proper for first HTTP request.
Jira: ZEP-2463
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
The server needs global enable/disable status instead of only being
able to enable or disable just the TLS server part.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
If we have specified AF_UNSPEC when initializing application
server local address, then we try to bind to both IPv4 and IPv6
addresses. The old code did not honor the port number in this
case but used some random value for port.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
If user has specified a specific local address, then honor that
and do not try to bind IPv4 context if only IPv6 is defined,
and vice versa for IPv6.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Setting just IPv4 address as was allowed before isn't enough for
real-world usage (e.g. accessing DNS and outside servers in general).
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Per LwM2M spec (7.3.2.4 Operation on Object):
"If the payload (New Value) conveys an Object Instance ID in conflict
with one already present in the LwM2M Client, the complete request
MUST be rejected and a "Bad Request" error code MUST be sent back."
Let's do that.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
Per LwM2M specification 7.3.2.4, "Optional Resources MAY be conveyed
in the "New Value" parameter as well; the LwM2M Client MAY ignore the
optional resources it doesn't support."
Update TLV/JSON writer to ignore error when object fields are not
found (treated as optional resource). This will allow the resources
supported being written.
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
[michael.scott@linaro.org: re-worked patch post addition of CREATE
operation.]
Signed-off-by: Michael Scott <michael.scott@linaro.org>
Prior to this patch, a CREATE operation was handled as a WRITE operation
after the object instance was created. This becomes problematic when
handling of optional resources differs between these 2 operations.
Let's introduce an actual CREATE operation and use it later to create
these differences.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
We are returning EINVAL from content format write ops when an object
field definition is not found (an optional field which is not
implemented). Instead, return ENOENT which lets the LwM2M engine
know to send ZOAP_RESPONSE_CODE_NOT_FOUND to the LwM2M server at the
end of handle_request().
NOTE: This behavior is not correct when we call the writer right after
a CREATE operation where the data is assigned to resources for the
first time. This case will be handled in a follow-up patch once we're
able to distinguish between a WRITE and a CREATE in the write op
handler.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
POSIX requires struct sockaddr's field to be named "sa_family"
(not just "family"):
http://pubs.opengroup.org/onlinepubs/009696699/basedefs/sys/socket.h.html
This change allows to port POSIX apps easier (including writing
portable apps using BSD Sockets compatible API).
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Per LwM2M specification 5.3.1 Register, report "ct=11543" when JSON is
supported. Also, report the resource type as rt="oma.lwm2m" when "ct="
presents.
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
Per LwM2M specification 5.3.1 Register. When object instances are
available, object ID can be ignored in registration message
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
According to LwM2M specification 5.2.7.3 Bootstrap DISOCVER,
security object is only reported to the bootstrap server.
Correct the behavior to (1) report server object to the server
(2) do not report security object at registration time
Signed-off-by: Robert Chou <robert.ch.chou@acer.com>
This is how it's called in the main docs, so use this same phrase in
Kconfig and samples too.
Also, added some articles to docs.
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
Make sure that a network packet is sent after calling
http_response_send_data(). Othwerwise the packets might be
piling up and not sent in timely manner.
Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com>
Currently, the HTTP_NETWORK_TIMEOUT setting is hard-coded as 20 seconds.
Not every application may want to wait that long, so let's change this
to a CONFIG option: CONFIG_HTTP_CLIENT_NETWORK_TIMEOUT
NOTE: This also removes HTTP_NETWORK_TIMEOUT from the public http.h
include file. It was not being used externally to HTTP client sources.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
A TCP FIN message is passed on to user apps as a tcp_received_callback
with a NULL pkt parameter. This means the connection is closing and
the app should do whatever cleanup it needs as there will be no further
callbacks for the current TCP connection.
Currently, if a HTTP client request doesn't receive a "body" which
the HTTP parser can use to trigger on_message_complete, then the request
will end up timing out and most apps will think an error has occurred.
Instead, let's handle the TCP FIN message and return the waiting
semaphore, leaving the app to deal with whatever has been set in the
current HTTP context response data (IE: http_status).
This fixes using HTTP client to send POST data to servers which
only respond with HTTP_OK status and no body.
Signed-off-by: Michael Scott <michael.scott@linaro.org>
