BaseSelectorEventLoop.sock_connect()
There is a race condition in create_connection() used with wait_for() to have a
timeout. sock_connect() registers the file descriptor of the socket to be
notified of write event (if connect() raises BlockingIOError). When
create_connection() is cancelled with a TimeoutError, sock_connect() coroutine
gets the exception, but it doesn't unregister the file descriptor for write
event. create_connection() gets the TimeoutError and closes the socket.
If you call again create_connection(), the new socket will likely gets the same
file descriptor, which is still registered in the selector. When sock_connect()
calls add_writer(), it tries to modify the entry instead of creating a new one.
This issue was originally reported in the Trollius project, but the bug comes
from Tulip in fact (Trollius is based on Tulip):
https://bitbucket.org/enovance/trollius/issue/15/after-timeouterror-on-wait_for
This change fixes the race condition. It also makes sock_connect() more
reliable (and portable) is sock.connect() raises an InterruptedError.
There is a race condition in create_connection() used with wait_for() to have a
timeout. sock_connect() registers the file descriptor of the socket to be
notified of write event (if connect() raises BlockingIOError). When
create_connection() is cancelled with a TimeoutError, sock_connect() coroutine
gets the exception, but it doesn't unregister the file descriptor for write
event. create_connection() gets the TimeoutError and closes the socket.
If you call again create_connection(), the new socket will likely gets the same
file descriptor, which is still registered in the selector. When sock_connect()
calls add_writer(), it tries to modify the entry instead of creating a new one.
This issue was originally reported in the Trollius project, but the bug comes
from Tulip in fact (Trollius is based on Tulip):
https://bitbucket.org/enovance/trollius/issue/15/after-timeouterror-on-wait_for
This change fixes the race condition. It also makes sock_connect() more
reliable (and portable) is sock.connect() raises an InterruptedError.
Without this reset, starttls would fail if a connect/starttls was done after a
quit, because smtplib assumed the existing value of emspt_features was
accurate, but it gets reset when starttls completes (and the new value does
not contain the starttls capability, since tls is already started at that
point). (There may be additional places where this lack of reset was an
issue as well.)
Patch by Milan Oberkirch.
QueryPerformanceFrequency() cannot fail on Windows XP and later according to
its documentation: raise an exception on error and drop the fallback to the
system clock.
Other changes:
* The whole _PyTime API is private (not defined if Py_LIMITED_API is set)
* _PyTime_gettimeofday_info() also returns -1 on error
* Simplify PyTime_gettimeofday(): only use clock_gettime(CLOCK_REALTIME) or
gettimeofday() on UNIX. Don't fallback to ftime() or time() anymore.
clock_gettime(CLOCK_REALTIME) if available. As a side effect, Python now
depends on the librt library on Solaris and on Linux (only with glibc older
than 2.17).
Don't raise a TimeoutError if we reached the timeout and the future completed
in the same iteration of the event loop. A side effect of the bug is that
Queue.get() looses items.
Don't raise a TimeoutError if we reached the timeout and the future completed
in the same iteration of the event loop. A side effect of the bug is that
Queue.get() looses items.
Benjamin Peterson
Berker Peksag
Ned Deily
Jason R. Coombs
Victor Stinner
R David Murray
Antoine Pitrou
Gregory P. Smith
Terry Jan Reedy
Stefan Krah