On Sat, Aug 18, 2018 at 09:40:05PM +0200, Simon Horman wrote:
> On Tue, Aug 07, 2018 at 05:07:58PM +0530, [email protected] wrote:
> > From: Numan Siddique <[email protected]>
> >
> > The python function ovs.socket_util.check_connection_completion() uses
> > select()
> > (provided by python) to monitor the socket file descriptor. The select()
> > returns 1 when the file descriptor becomes ready. For error cases like -
> > 111 (Connection refused) and 113 (No route to host) (POLLERR),
> > ovs.poller._SelectSelect.poll()
> > expects the exceptfds list to be set by select(). But that is not the case.
> > As per the select() man page, writefds list will be set for POLLERR.
> > Please see "Correspondence between select() and poll() notifications"
> > section of select(2)
> > man page.
> >
> > Because of this behavior, ovs.socket_util.check_connection_completion()
> > returns success
> > even if the remote is unreachable or not listening on the port.
> >
> > This patch fixes this issue by using poll() to check the connection status
> > similar to
> > the C implementation of check_connection_completion().
> >
> > A new function 'get_system_poll() is added in ovs/poller.py which returns
> > the
> > select.poll() object. If select.poll is monkey patched by eventlet/gevent,
> > it
> > gets the original select.poll() and returns it.
> >
> > The test cases added in this patch fails without the fix.
> >
> > Suggested-by: Ben Pfaff <[email protected]>
> > Signed-off-by: Numan Siddique <[email protected]>
>
> I believe that this patch is the cause of the testsuite failure currently
> exhibited in the master branch when tested using travis-ci.
>
> https://travis-ci.org/openvswitch/ovs/jobs/417506303
>
> ...
>
> > diff --git a/tests/ovsdb-idl.at b/tests/ovsdb-idl.at
> > index 014382850..e8a26e971 100644
> > --- a/tests/ovsdb-idl.at
> > +++ b/tests/ovsdb-idl.at
> > @@ -1720,3 +1720,19 @@ OVSDB_CHECK_IDL_COMPOUND_INDEX_WITH_REF([set,
> > simple3 idl-compound-index-with-re
> > 007: check simple4: empty
> > 008: End test
> > ]])
> > +
> > +m4_define([CHECK_STREAM_OPEN_BLOCK_PY],
> > + [AT_SETUP([$1])
> > + AT_SKIP_IF([test $2 = no])
> > + AT_KEYWORDS([Check PY Stream open block - $3])
> > + AT_CHECK([ovsdb_start_idltest "ptcp:0:127.0.0.1"])
> > + PARSE_LISTENING_PORT([ovsdb-server.log], [TCP_PORT])
> > + WRONG_PORT=$(($TCP_PORT+1))
> > + AT_CHECK([$3 $srcdir/test-stream.py tcp:127.0.0.1:$TCP_PORT], [0],
> > [ignore])
> > + AT_CHECK([$3 $srcdir/test-stream.py tcp:127.0.0.1:$WRONG_PORT], [1],
> > [ignore])
> > + OVSDB_SERVER_SHUTDOWN
> > + AT_CHECK([$3 $srcdir/test-stream.py tcp:127.0.0.1:$TCP_PORT], [1],
> > [ignore])
> > + AT_CLEANUP])
> > +
> > +CHECK_STREAM_OPEN_BLOCK_PY([Check PY2 Stream open block], [$HAVE_PYTHON],
> > [$PYTHON])
> > +CHECK_STREAM_OPEN_BLOCK_PY([Check PY3 Stream open block], [$HAVE_PYTHON],
> > [$PYTHON3])
>
> I wonder if the problem exhibited by travis-ci would be resolved
> by changing $HAVE_PYTHON to $HAVE_PYTHON3 on the "PY3 Stream" line.
>
> Also, I wonder if s/PYTHON/PYTHON2/ is appropriate on the "PY2 Stream" line.
You're right, thanks. I sent a fix:
https://patchwork.ozlabs.org/patch/959273/
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
