If pktinfo were supported then we could listen on :: for ipv6 and get
the information about the destination address from pktinfo structure passed
in recvmsg but this method is not portable and libuv doesn't support it - so
we need to listen on all interfaces.
We should verify that this doesn't impact performance (we already do it for
ipv4) and either remove all the ipv6pktinfo detection code or think of fixing
libuv.
- use UV_{TC,UD}P_IPV6ONLY for IPv6 sockets, keeping the pre-netmgr
behaviour.
- add a new listening_error bool flag which is set if the child
listener fails to start listening. This fixes a bug where named would
hang if, e.g., we failed to bind to a TCP socket.
For multithreaded TCP listening we need to pass a bound socket to all
listening threads. Instead of using uv_pipe handle passing method which
is quite complex (lots of callbacks, each of them with its own error
handling) we now use isc_uv_export() to export the socket, pass it as a
member of the isc__netievent_tcpchildlisten_t structure, and then
isc_uv_import() it in the child thread, simplifying the process
significantly.
These functions can be used to pass a uv handle between threads in a
safe manner. The other option is to use uv_pipe and pass the uv_handle
via IPC, which is way more complex. uv_export() and uv_import() functions
existed in libuv at some point but were removed later. This code is
based on the original removed code.
The Windows version of the code uses two functions internal to libuv;
a patch for libuv is attached for exporting these functions.
After the network manager rewrite, tcp-higwater stats was only being
updated when a valid DNS query was received over tcp.
It turns out tcp-quota is updated right after a tcp connection is
accepted, before any data is read, so in the event that some client
connect but don't send a valid query, it wouldn't be taken into
account to update tcp-highwater stats, that is wrong.
This commit fix tcp-highwater to update its stats whenever a tcp connection
is established, independent of what happens after (timeout/invalid
request, etc).
- make tcp listening IPC pipe name saner
- put the pipe in /tmp on unices
- add pid to the pipe name to avoid conflicts between processes
- fsync directory in which the pipe resides to make sure that the
child threads will see it and be able to open it
even when worker is paused (e.g. interface reconfiguration). This is
needed to prevent deadlocks when reconfiguring interfaces - as network
manager is paused then, but we still need to stop/start listening.
- Proper handling of TCP listen errors in netmgr - bind to the socket first,
then return the error code.
When listening for TCP connections we create a socket, bind it
and then pass it over IPC to all threads - which then listen on
in and accept connections. This sounds broken, but it's the
official way of dealing with multithreaded TCP listeners in libuv,
and works on all platforms supported by libuv.
For BIND 9.16+, TLS aware compiler is required, and using
ISC_THREAD_LOCAL is preferred way of using Thread Local Storage. The
isc_thread_key API is no longer used anywhere and hence was removed from
BIND 9.
The new ISC_THREAD_LOCAL macro unifies usage of platform dependent
Thread Local Storage definition thread_local vs __thread vs
__declspec(thread) to a single macro.
The commit also unifies the required level of support for TLS as for
some parts of the code it was mandatory and for some parts of the code
it wasn't.
FCTX_ATTR_SHUTTINGDOWN needs to be set and tested while holding the node
lock but the rest of the attributes don't as they are task locked. Making
fctx->attributes atomic allows both behaviours without races.
- restore support for tcp-initial-timeout, tcp-idle-timeout,
tcp-keepalive-timeout and tcp-advertised-timeout configuration
options, which were ineffective previously.
- add timeout support for TCP and TCPDNS connections to protect against
slowloris style attacks. currently, all timeouts are hard-coded.
- rework and simplify the TCPDNS state machine.
