Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 2.2.2015 23:09, Stuart Marks wrote:
On 2/2/15 2:33 AM, Jaroslav Bachorik wrote:
I've applied your comments and the code is a tad simpler now. I also
ironed out
a few more corner cases. I ran the test 500x in a tight loop and no
failure, yay!
Update: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.04
Hi Jaroslav,
Looks quite a bit more straightforward now. I'm pretty much OK with this
if you're OK with it; I think I've reviewed it enough times already. :-)
I have a couple comments on test_09 that you might want to address
before you push or sometime in the future.
Line 714 typo hugging = hogging
D'oh. Thanks.
But I'm not convinced this retry logic in the while-loop from lines
716-723 is necessary. If you've already opened a server socket on a
port, I've never seen a case where opening the same port *again* will
succeed, so why bother?
I saw the test failing without this retry logic. The purpose is to make
sure that the port is really taken before the test continues.
I'd suggest simply opening ServerSocket(0) and then getting the port via
getLocalPort(). I've never seen this fail. This case should work since
you actually want to open up some random port, instead of generating a
random port number for somebody else (the subprocess) to open. You can
then allocate one fewer random port. You might want to have a little
loop to check that the random port number isn't a duplicate of the
actual port that you just opened. I think this lets the code boil down a
bit further:
ServerSocket ss = new ServerSocket(0);
try {
int localPort = ss.getLocalPort();
int[] ports;
do {
ports = PortAllocator.allocatePorts(1);
} while (localPort != ports[0]);
AtomicBoolean checks = new AtomicBoolean(false);
...
} finally {
ss.close();
s.stop();
}
I'll give it a try and see how this behaves.
Thanks,
-JB-
s'marks
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Jaroslav,
Looks good for me!
I've changed the PortAllocator to allocate an array of unique random
ports instead of letting ServerSocket to take care of it.
I ran the test 200x in a tight loop without a failure.
I hope this will resolve this test's intermittent failures due to port
conflicts once and for all.
Update: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.03
-Dmitry
On 2015-01-09 15:17, Jaroslav Bachorik wrote:
Thank you all for the valuable input!
On 9.1.2015 03:50, Stuart Marks wrote:
Hi Jaroslav,
I'm distant enough from this code that I don't think I'm in a position
to say no you can't check this in, and I'm mindful of the fact that
this bug is a high priority and you want to get a fix in. But having
said that I think there's a surprising amount of complexity here for
what it does.
Implementing a retry-on-BindException policy seems to be fairly
sensible, since I assume it would be fairly invasive to modify the code
in the JVMs being forked to use an ephemeral port and send that
information back to the test.
My conjecture is however that the open/close/reopen logic is the primary
cause of the BindExceptions that occur. If you're going to retry on
BindException, you might as well choose a random port number instead of
doing the open/close to get a port number from the system.
The range that Dmitry suggests is reasonable, though I note that the
actual ephemeral port range used by the kernel will differ from OS to OS
and even from system to system. I don't know if that's really
significant though. If you end up choosing a port outside the ephemeral
range for some system, does it really matter?
If you do decide to have PortAllocator open and close a ServerSocket (in
order to find a previously unused port) I'd suggest removing the logic
that leaves the socket open until the first call to get(). That logic
reduces the possibility that some other process will open the socket
after the close but before the reopen. In my experience that's not
what's causing the BindExceptions. It could still happen, though, but
you're protected by the retry logic anyway. Leaving the socket open
longer actually hurts, I think, because it increases the chance that the
kernel won't have cleaned up the port by the time the test wants to
reopen it.
If you change PortAllocator to close the socket immediately, you can get
rid of the need to call release() in a finally-block of the caller. You
could also change the type of the functional interface to be
int[] - void
since the PortAllocator doesn't hold onto any resources that need to be
cleaned up. It just calls the execute() method and passes an array of n
port numbers.
It's probably necessary to have the socket close() call in a retry loop.
The socket is always closed immediately from the user process point of
view, so isClosed() will always return true immediately after the
close() call returns. You can verify this easily by looking in the
ServerSocket.java source code. I believe the state that prevents the
port from being reused immediately is private to the kernel and cannot
be observed from a user process, at least not without attempting to
reopen the socket.
Side note: one of the jcmd() overloads says that parameter 'c' (a
Consumer) may be null. It doesn't look like this is handled. If you
really want to support this, I'd assign () - { } to it if it's null so
that it can be called unconditionally. (Or just disallow null.)
I've changed the PortAllocator to allocate an array of unique random
ports instead of letting ServerSocket to take care of it.
I ran the test 200x in a tight loop without a failure.
I hope this will resolve this test's intermittent failures due to port
conflicts once and for all.
Update: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.03
Thanks,
-JB-
s'marks
On 1/6/15 2:00 PM, Dmitry Samersoff wrote:
Jaroslav,
It might be better to just choose a random digit between 49152–65535
and attempt to use it.
-Dmitry
On 2014-12-18 17:29, Jaroslav Bachorik wrote:
On 12/11/2014 03:43 PM, Dmitry Samersoff wrote:
Jaroslav,
You can set SO_LINGER to zero, in this case socket will be closed
immediately without waiting in TIME_WAIT
But there are no reliable way to predict whether you can take this
port
or not after you close it.
So the only valid solution is to try to connect to a random port
and if
this attempt fails try another random port. Everything else will cause
more or less frequent intermittent failures.
Thanks for all the suggestions!
http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02
I've enhanced the original patch with the retry logic using different
random port if starting the JMX agent on the provided port fails with
BindException.
I'm keeping there the changes for properly closing the ports opened for
the test purposes and also setting the SO_REUSEADDR - anyway, it does
not make sense to reuse
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 2/2/15 2:33 AM, Jaroslav Bachorik wrote:
I've applied your comments and the code is a tad simpler now. I also ironed out
a few more corner cases. I ran the test 500x in a tight loop and no failure,
yay!
Update: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.04
Hi Jaroslav,
Looks quite a bit more straightforward now. I'm pretty much OK with this if
you're OK with it; I think I've reviewed it enough times already. :-)
I have a couple comments on test_09 that you might want to address before you
push or sometime in the future.
Line 714 typo hugging = hogging
But I'm not convinced this retry logic in the while-loop from lines 716-723 is
necessary. If you've already opened a server socket on a port, I've never seen a
case where opening the same port *again* will succeed, so why bother?
I'd suggest simply opening ServerSocket(0) and then getting the port via
getLocalPort(). I've never seen this fail. This case should work since you
actually want to open up some random port, instead of generating a random port
number for somebody else (the subprocess) to open. You can then allocate one
fewer random port. You might want to have a little loop to check that the random
port number isn't a duplicate of the actual port that you just opened. I think
this lets the code boil down a bit further:
ServerSocket ss = new ServerSocket(0);
try {
int localPort = ss.getLocalPort();
int[] ports;
do {
ports = PortAllocator.allocatePorts(1);
} while (localPort != ports[0]);
AtomicBoolean checks = new AtomicBoolean(false);
...
} finally {
ss.close();
s.stop();
}
s'marks
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Stuart,
The range that Dmitry suggests is reasonable, though I note that the
actual ephemeral port range used by the kernel will differ from OS to
OS and even from system to system. I don't know if that's really
significant though. If you end up choosing a port outside the
ephemeral range for some system, does it really matter?
This range is assigned by IANA so it's a standard.
-Dmitry
On 2015-01-09 05:50, Stuart Marks wrote:
Hi Jaroslav,
I'm distant enough from this code that I don't think I'm in a position
to say no you can't check this in, and I'm mindful of the fact that
this bug is a high priority and you want to get a fix in. But having
said that I think there's a surprising amount of complexity here for
what it does.
Implementing a retry-on-BindException policy seems to be fairly
sensible, since I assume it would be fairly invasive to modify the code
in the JVMs being forked to use an ephemeral port and send that
information back to the test.
My conjecture is however that the open/close/reopen logic is the primary
cause of the BindExceptions that occur. If you're going to retry on
BindException, you might as well choose a random port number instead of
doing the open/close to get a port number from the system.
The range that Dmitry suggests is reasonable, though I note that the
actual ephemeral port range used by the kernel will differ from OS to OS
and even from system to system. I don't know if that's really
significant though. If you end up choosing a port outside the ephemeral
range for some system, does it really matter?
If you do decide to have PortAllocator open and close a ServerSocket (in
order to find a previously unused port) I'd suggest removing the logic
that leaves the socket open until the first call to get(). That logic
reduces the possibility that some other process will open the socket
after the close but before the reopen. In my experience that's not
what's causing the BindExceptions. It could still happen, though, but
you're protected by the retry logic anyway. Leaving the socket open
longer actually hurts, I think, because it increases the chance that the
kernel won't have cleaned up the port by the time the test wants to
reopen it.
If you change PortAllocator to close the socket immediately, you can get
rid of the need to call release() in a finally-block of the caller. You
could also change the type of the functional interface to be
int[] - void
since the PortAllocator doesn't hold onto any resources that need to be
cleaned up. It just calls the execute() method and passes an array of n
port numbers.
It's probably necessary to have the socket close() call in a retry loop.
The socket is always closed immediately from the user process point of
view, so isClosed() will always return true immediately after the
close() call returns. You can verify this easily by looking in the
ServerSocket.java source code. I believe the state that prevents the
port from being reused immediately is private to the kernel and cannot
be observed from a user process, at least not without attempting to
reopen the socket.
Side note: one of the jcmd() overloads says that parameter 'c' (a
Consumer) may be null. It doesn't look like this is handled. If you
really want to support this, I'd assign () - { } to it if it's null so
that it can be called unconditionally. (Or just disallow null.)
s'marks
On 1/6/15 2:00 PM, Dmitry Samersoff wrote:
Jaroslav,
It might be better to just choose a random digit between 49152–65535
and attempt to use it.
-Dmitry
On 2014-12-18 17:29, Jaroslav Bachorik wrote:
On 12/11/2014 03:43 PM, Dmitry Samersoff wrote:
Jaroslav,
You can set SO_LINGER to zero, in this case socket will be closed
immediately without waiting in TIME_WAIT
But there are no reliable way to predict whether you can take this port
or not after you close it.
So the only valid solution is to try to connect to a random port and if
this attempt fails try another random port. Everything else will cause
more or less frequent intermittent failures.
Thanks for all the suggestions!
http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02
I've enhanced the original patch with the retry logic using different
random port if starting the JMX agent on the provided port fails with
BindException.
I'm keeping there the changes for properly closing the ports opened for
the test purposes and also setting the SO_REUSEADDR - anyway, it does
not make sense to reuse the ephemeral test ports.
I've split the original test_06 test case in order to keep it readable
even with the new retry logic - and also to make each test case to test
just one scenario.
Cheers,
-JB-
-Dmitry
On 2014-12-11 17:06, Jaroslav Bachorik wrote:
On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote:
On 12/09/2014 01:39 AM, Stuart Marks wrote:
On 12/8/14 12:35 PM, Jaroslav Bachorik wrote:
Please, review the following test change
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Thank you all for the valuable input!
On 9.1.2015 03:50, Stuart Marks wrote:
Hi Jaroslav,
I'm distant enough from this code that I don't think I'm in a position
to say no you can't check this in, and I'm mindful of the fact that
this bug is a high priority and you want to get a fix in. But having
said that I think there's a surprising amount of complexity here for
what it does.
Implementing a retry-on-BindException policy seems to be fairly
sensible, since I assume it would be fairly invasive to modify the code
in the JVMs being forked to use an ephemeral port and send that
information back to the test.
My conjecture is however that the open/close/reopen logic is the primary
cause of the BindExceptions that occur. If you're going to retry on
BindException, you might as well choose a random port number instead of
doing the open/close to get a port number from the system.
The range that Dmitry suggests is reasonable, though I note that the
actual ephemeral port range used by the kernel will differ from OS to OS
and even from system to system. I don't know if that's really
significant though. If you end up choosing a port outside the ephemeral
range for some system, does it really matter?
If you do decide to have PortAllocator open and close a ServerSocket (in
order to find a previously unused port) I'd suggest removing the logic
that leaves the socket open until the first call to get(). That logic
reduces the possibility that some other process will open the socket
after the close but before the reopen. In my experience that's not
what's causing the BindExceptions. It could still happen, though, but
you're protected by the retry logic anyway. Leaving the socket open
longer actually hurts, I think, because it increases the chance that the
kernel won't have cleaned up the port by the time the test wants to
reopen it.
If you change PortAllocator to close the socket immediately, you can get
rid of the need to call release() in a finally-block of the caller. You
could also change the type of the functional interface to be
int[] - void
since the PortAllocator doesn't hold onto any resources that need to be
cleaned up. It just calls the execute() method and passes an array of n
port numbers.
It's probably necessary to have the socket close() call in a retry loop.
The socket is always closed immediately from the user process point of
view, so isClosed() will always return true immediately after the
close() call returns. You can verify this easily by looking in the
ServerSocket.java source code. I believe the state that prevents the
port from being reused immediately is private to the kernel and cannot
be observed from a user process, at least not without attempting to
reopen the socket.
Side note: one of the jcmd() overloads says that parameter 'c' (a
Consumer) may be null. It doesn't look like this is handled. If you
really want to support this, I'd assign () - { } to it if it's null so
that it can be called unconditionally. (Or just disallow null.)
I've changed the PortAllocator to allocate an array of unique random
ports instead of letting ServerSocket to take care of it.
I ran the test 200x in a tight loop without a failure.
I hope this will resolve this test's intermittent failures due to port
conflicts once and for all.
Update: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.03
Thanks,
-JB-
s'marks
On 1/6/15 2:00 PM, Dmitry Samersoff wrote:
Jaroslav,
It might be better to just choose a random digit between 49152–65535
and attempt to use it.
-Dmitry
On 2014-12-18 17:29, Jaroslav Bachorik wrote:
On 12/11/2014 03:43 PM, Dmitry Samersoff wrote:
Jaroslav,
You can set SO_LINGER to zero, in this case socket will be closed
immediately without waiting in TIME_WAIT
But there are no reliable way to predict whether you can take this port
or not after you close it.
So the only valid solution is to try to connect to a random port and if
this attempt fails try another random port. Everything else will cause
more or less frequent intermittent failures.
Thanks for all the suggestions!
http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02
I've enhanced the original patch with the retry logic using different
random port if starting the JMX agent on the provided port fails with
BindException.
I'm keeping there the changes for properly closing the ports opened for
the test purposes and also setting the SO_REUSEADDR - anyway, it does
not make sense to reuse the ephemeral test ports.
I've split the original test_06 test case in order to keep it readable
even with the new retry logic - and also to make each test case to test
just one scenario.
Cheers,
-JB-
-Dmitry
On 2014-12-11 17:06, Jaroslav Bachorik wrote:
On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote:
On 12/09/2014 01:39 AM, Stuart Marks wrote:
On 12/8/14 12:35 PM, Jaroslav Bachorik wrote:
Please, review the following test change
Issue : https://bugs.openjdk.java.net/browse/JDK-8066708
Webrev:
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Hi Jaroslav,
I'm distant enough from this code that I don't think I'm in a position to say
no you can't check this in, and I'm mindful of the fact that this bug is a
high priority and you want to get a fix in. But having said that I think there's
a surprising amount of complexity here for what it does.
Implementing a retry-on-BindException policy seems to be fairly sensible, since
I assume it would be fairly invasive to modify the code in the JVMs being forked
to use an ephemeral port and send that information back to the test.
My conjecture is however that the open/close/reopen logic is the primary cause
of the BindExceptions that occur. If you're going to retry on BindException, you
might as well choose a random port number instead of doing the open/close to get
a port number from the system.
The range that Dmitry suggests is reasonable, though I note that the actual
ephemeral port range used by the kernel will differ from OS to OS and even from
system to system. I don't know if that's really significant though. If you end
up choosing a port outside the ephemeral range for some system, does it really
matter?
If you do decide to have PortAllocator open and close a ServerSocket (in order
to find a previously unused port) I'd suggest removing the logic that leaves the
socket open until the first call to get(). That logic reduces the possibility
that some other process will open the socket after the close but before the
reopen. In my experience that's not what's causing the BindExceptions. It could
still happen, though, but you're protected by the retry logic anyway. Leaving
the socket open longer actually hurts, I think, because it increases the chance
that the kernel won't have cleaned up the port by the time the test wants to
reopen it.
If you change PortAllocator to close the socket immediately, you can get rid of
the need to call release() in a finally-block of the caller. You could also
change the type of the functional interface to be
int[] - void
since the PortAllocator doesn't hold onto any resources that need to be cleaned
up. It just calls the execute() method and passes an array of n port numbers.
It's probably necessary to have the socket close() call in a retry loop. The
socket is always closed immediately from the user process point of view, so
isClosed() will always return true immediately after the close() call returns.
You can verify this easily by looking in the ServerSocket.java source code. I
believe the state that prevents the port from being reused immediately is
private to the kernel and cannot be observed from a user process, at least not
without attempting to reopen the socket.
Side note: one of the jcmd() overloads says that parameter 'c' (a Consumer) may
be null. It doesn't look like this is handled. If you really want to support
this, I'd assign () - { } to it if it's null so that it can be called
unconditionally. (Or just disallow null.)
s'marks
On 1/6/15 2:00 PM, Dmitry Samersoff wrote:
Jaroslav,
It might be better to just choose a random digit between 49152–65535
and attempt to use it.
-Dmitry
On 2014-12-18 17:29, Jaroslav Bachorik wrote:
On 12/11/2014 03:43 PM, Dmitry Samersoff wrote:
Jaroslav,
You can set SO_LINGER to zero, in this case socket will be closed
immediately without waiting in TIME_WAIT
But there are no reliable way to predict whether you can take this port
or not after you close it.
So the only valid solution is to try to connect to a random port and if
this attempt fails try another random port. Everything else will cause
more or less frequent intermittent failures.
Thanks for all the suggestions!
http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02
I've enhanced the original patch with the retry logic using different
random port if starting the JMX agent on the provided port fails with
BindException.
I'm keeping there the changes for properly closing the ports opened for
the test purposes and also setting the SO_REUSEADDR - anyway, it does
not make sense to reuse the ephemeral test ports.
I've split the original test_06 test case in order to keep it readable
even with the new retry logic - and also to make each test case to test
just one scenario.
Cheers,
-JB-
-Dmitry
On 2014-12-11 17:06, Jaroslav Bachorik wrote:
On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote:
On 12/09/2014 01:39 AM, Stuart Marks wrote:
On 12/8/14 12:35 PM, Jaroslav Bachorik wrote:
Please, review the following test change
Issue : https://bugs.openjdk.java.net/browse/JDK-8066708
Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00
The test fails very intermittently when RMI registry is trying to
bind
to a port
previously used in the test (via ServerSocket).
This seems to be caused by the sockets created via `new
ServerSocket(0)` and
being in reusable mode. The fix attempts to prevent this by
explicitly
forbidding the reusable mode.
Hi Jaroslav,
I happened to see this fly by, and there
[ping] Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 18.12.2014 15:29, Jaroslav Bachorik wrote: On 12/11/2014 03:43 PM, Dmitry Samersoff wrote: Jaroslav, You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT But there are no reliable way to predict whether you can take this port or not after you close it. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. Thanks for all the suggestions! http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02 I've enhanced the original patch with the retry logic using different random port if starting the JMX agent on the provided port fails with BindException. I'm keeping there the changes for properly closing the ports opened for the test purposes and also setting the SO_REUSEADDR - anyway, it does not make sense to reuse the ephemeral test ports. I've split the original test_06 test case in order to keep it readable even with the new retry logic - and also to make each test case to test just one scenario. Cheers, -JB- -Dmitry On 2014-12-11 17:06, Jaroslav Bachorik wrote: On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote: On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. Further investigation shows that the problem was rather the client connecting to a socket being shut down. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout state involving the socket address or port. It also turns out that the test does not close the server sockets properly so there might be several sockets being opened or timed out dangling around. I've
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Jaroslav, It might be better to just choose a random digit between 49152–65535 and attempt to use it. -Dmitry On 2014-12-18 17:29, Jaroslav Bachorik wrote: On 12/11/2014 03:43 PM, Dmitry Samersoff wrote: Jaroslav, You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT But there are no reliable way to predict whether you can take this port or not after you close it. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. Thanks for all the suggestions! http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02 I've enhanced the original patch with the retry logic using different random port if starting the JMX agent on the provided port fails with BindException. I'm keeping there the changes for properly closing the ports opened for the test purposes and also setting the SO_REUSEADDR - anyway, it does not make sense to reuse the ephemeral test ports. I've split the original test_06 test case in order to keep it readable even with the new retry logic - and also to make each test case to test just one scenario. Cheers, -JB- -Dmitry On 2014-12-11 17:06, Jaroslav Bachorik wrote: On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote: On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. Further investigation shows that the problem was rather the client connecting to a socket being shut down. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 12/17/14 10:40 AM, Dmitry Samersoff wrote: 1. Ever if you set SO_LINGER to zero, socket will not be closed immediately. see TCP shutdown sequence. Which TCP shutdown sequence? The one that involves FIN_WAIT_1, FIN_WAIT_2, and TIME_WAIT? That's the state machine for a connected socket; the sockets in question here have never been connected. There is still apparently a state the socket goes through before the port is actually freed, but it might not have anything to do with TCP. If you have references for this behavior I'd appreciate them. What I've learned about this issue is only through empirical observation. By the way, more empiricism: I can reproduce the EADDRINUSE on Solaris a large fraction of the time, by running multiple programs (or threads) that simply close and reopen distinct sockets. I've also observed that setting SO_REUSEPORT (introduced in Solaris 11) seems to avoid this problem entirely. Unfortunately SO_REUSEPORT isn't available from Java, it doesn't exist on all systems, and I don't know if it would have this same behavior on other systems on which it does exist. 2. In a native world it's quite easy to find the port your rmi server uses - it could be achieved by parsing /proc/pid/net/tcp on Linux or using special API on windows and solaris. I think your definition of easy differs from mine. :-) The context is the jdk regression tests, does support native code. I'd prefer not to have to explore this new area, especially in addition to writing a bunch of system-specific native code. 3. For rmid and port provided from outside I think the only reliable way to get what you need is: [...] I think it's pretty clear at this point that the open-close-reopen approach can't be made reliable in any platform-independent way. For rmid I think I might create a new mode that opens an ephemeral port and sends that to the test driver somehow. Looks like Jaroslav is proceeding with a retry strategy. 4. For rmid you can also emulate inetd behaviour - i.e. driver open a server port, communicate it to client than redirect everything that come to this port to stdin of rmid. Thanks, but unfortunately this is actually one of the modes that needs to be tested in rmid. It has a mode where it opens and listens on its own socket, and another mode where it inherits one from its parent process, so that it can be invoked from inetd. s'marks -Dmitry On 2014-12-17 01:55, Stuart Marks wrote: Hi Dmitry, Strictly speaking you are correct. As soon as you close a socket, there is a possibility -- perhaps vanishingly small but nonzero -- that you might not be able to open it again. The first scenario, where the user of the socket itself opens the socket using an ephemeral port (e.g. new ServerSocket(0)) is of course preferred. This avoids race conditions entirely. It's the second case that I'm still wrestling with, and maybe Jaroslav too. It's fairly difficult to get such black box systems to open an ephemeral port and report it back, as opposed to opening up their service on some port number handed in from the outside. (For RMI, rmid is the culprit here. I don't know about JMX.) What makes this difficult is that the rmid service is running in a separate VM, so getting reliable information back from it can be difficult. It's also fairly difficult to establish the retry logic in such cases. If the service fails with a BindException, maybe -- maybe -- it was because there was a conflict over the port, and a retry is warranted. But this needs to be distinguished from other failure modes that might occur, that should be reported as failures instead of causing a retry. In principle, this is possible to do, of course, it's just that it involves more restructuring of the tests, and possibly adding debug/test code to rmid. (It may yet come to that.) I'm still pondering the reasons that, in the open/close/reopen scenario, why the reopen might fail. The obvious reason is that some other process on the system has opened that port between the close and the reopen. I admit that this is a possibility. However, with the open/close/reopen scenario in place, we see tests that fail up to 15% of the time with BindExceptions. This is an extraordinarily high failure rate to be caused by some random other process happening to open the same port in the few microseconds between the close and reopen. It's simply not believable to me. My thinking is still that the port isn't ready for reuse until a small amount of time after it's closed. I have some test programs that exercise sockets in a particular way (e.g., from multiple threads, or opening and closing batches of sockets) that can reproduce the problem on some systems, and these test programs seem to behave better if a time delay is added between the close and the reopen. The exact circumstances under which the problem occurs is difficult to pin down and seems OS specific, and so choosing the right delay time is very difficult. But it
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 12/11/2014 03:43 PM, Dmitry Samersoff wrote: Jaroslav, You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT But there are no reliable way to predict whether you can take this port or not after you close it. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. Thanks for all the suggestions! http://cr.openjdk.java.net/~jbachorik/8066708/webrev.02 I've enhanced the original patch with the retry logic using different random port if starting the JMX agent on the provided port fails with BindException. I'm keeping there the changes for properly closing the ports opened for the test purposes and also setting the SO_REUSEADDR - anyway, it does not make sense to reuse the ephemeral test ports. I've split the original test_06 test case in order to keep it readable even with the new retry logic - and also to make each test case to test just one scenario. Cheers, -JB- -Dmitry On 2014-12-11 17:06, Jaroslav Bachorik wrote: On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote: On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. Further investigation shows that the problem was rather the client connecting to a socket being shut down. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout state involving the socket address or port. It also turns out that the test does not close the server sockets properly so there might be several sockets being opened or timed out dangling around. I've updated the test so it is setting
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Stuart, 1. Ever if you set SO_LINGER to zero, socket will not be closed immediately. see TCP shutdown sequence. 2. In a native world it's quite easy to find the port your rmi server uses - it could be achieved by parsing /proc/pid/net/tcp on Linux or using special API on windows and solaris. 3. For rmid and port provided from outside I think the only reliable way to get what you need is: Write a driver that: 1. starts rmid -port some_random_value 2. connect to this port to make sure rmid is actually started (no rmi, just plain tcp connect) 3. if (2) fail return to (1) 4. run client with this port number. 4. For rmid you can also emulate inetd behaviour - i.e. driver open a server port, communicate it to client than redirect everything that come to this port to stdin of rmid. -Dmitry On 2014-12-17 01:55, Stuart Marks wrote: Hi Dmitry, Strictly speaking you are correct. As soon as you close a socket, there is a possibility -- perhaps vanishingly small but nonzero -- that you might not be able to open it again. The first scenario, where the user of the socket itself opens the socket using an ephemeral port (e.g. new ServerSocket(0)) is of course preferred. This avoids race conditions entirely. It's the second case that I'm still wrestling with, and maybe Jaroslav too. It's fairly difficult to get such black box systems to open an ephemeral port and report it back, as opposed to opening up their service on some port number handed in from the outside. (For RMI, rmid is the culprit here. I don't know about JMX.) What makes this difficult is that the rmid service is running in a separate VM, so getting reliable information back from it can be difficult. It's also fairly difficult to establish the retry logic in such cases. If the service fails with a BindException, maybe -- maybe -- it was because there was a conflict over the port, and a retry is warranted. But this needs to be distinguished from other failure modes that might occur, that should be reported as failures instead of causing a retry. In principle, this is possible to do, of course, it's just that it involves more restructuring of the tests, and possibly adding debug/test code to rmid. (It may yet come to that.) I'm still pondering the reasons that, in the open/close/reopen scenario, why the reopen might fail. The obvious reason is that some other process on the system has opened that port between the close and the reopen. I admit that this is a possibility. However, with the open/close/reopen scenario in place, we see tests that fail up to 15% of the time with BindExceptions. This is an extraordinarily high failure rate to be caused by some random other process happening to open the same port in the few microseconds between the close and reopen. It's simply not believable to me. My thinking is still that the port isn't ready for reuse until a small amount of time after it's closed. I have some test programs that exercise sockets in a particular way (e.g., from multiple threads, or opening and closing batches of sockets) that can reproduce the problem on some systems, and these test programs seem to behave better if a time delay is added between the close and the reopen. The exact circumstances under which the problem occurs is difficult to pin down and seems OS specific, and so choosing the right delay time is very difficult. But it does strengthen this conjecture in my mind. Naturally it would be better if there were a way to determine when a port is available for reuse without actually opening it. I'm not aware of any such way, but I'm holding onto a little hope that one can be found. s'marks On 12/11/14 10:18 AM, Dmitry Samersoff wrote: Stuart, As soon as you close socket, you open a door for the race. So you need another communication channel to pass a port number (or bind result) between a client and a server without closing a socket on the server side. Typical scenario used by network related code is: 1. Server opens the socket 2. Server binds to port(0) 3. Server gets port number assigned by OS 4. Server informs client (e.g. write the port down to known file, broadcast it etc) 5. Client establishes connection. If the server is a blackbox and have to get a port number from outside, scenario looks like: WHILE(!success and !timeout) 1. Driver chooses random port number 2. Driver runs a server with this number 3. Driver checks that server is actually listening on this port (e.g. try to connect by it self) WEND 4. Driver runs a client with this port number or bails out with descriptive error message. -Dmitry On 2014-12-11 20:53, Stuart Marks wrote: On 12/11/14 7:09 AM, olivier.lagn...@oracle.com wrote: On 11/12/2014 15:43, Dmitry Samersoff wrote: You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT SO-LINGER did not help either in my case (see my previous mail to
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Hi Dmitry, Strictly speaking you are correct. As soon as you close a socket, there is a possibility -- perhaps vanishingly small but nonzero -- that you might not be able to open it again. The first scenario, where the user of the socket itself opens the socket using an ephemeral port (e.g. new ServerSocket(0)) is of course preferred. This avoids race conditions entirely. It's the second case that I'm still wrestling with, and maybe Jaroslav too. It's fairly difficult to get such black box systems to open an ephemeral port and report it back, as opposed to opening up their service on some port number handed in from the outside. (For RMI, rmid is the culprit here. I don't know about JMX.) What makes this difficult is that the rmid service is running in a separate VM, so getting reliable information back from it can be difficult. It's also fairly difficult to establish the retry logic in such cases. If the service fails with a BindException, maybe -- maybe -- it was because there was a conflict over the port, and a retry is warranted. But this needs to be distinguished from other failure modes that might occur, that should be reported as failures instead of causing a retry. In principle, this is possible to do, of course, it's just that it involves more restructuring of the tests, and possibly adding debug/test code to rmid. (It may yet come to that.) I'm still pondering the reasons that, in the open/close/reopen scenario, why the reopen might fail. The obvious reason is that some other process on the system has opened that port between the close and the reopen. I admit that this is a possibility. However, with the open/close/reopen scenario in place, we see tests that fail up to 15% of the time with BindExceptions. This is an extraordinarily high failure rate to be caused by some random other process happening to open the same port in the few microseconds between the close and reopen. It's simply not believable to me. My thinking is still that the port isn't ready for reuse until a small amount of time after it's closed. I have some test programs that exercise sockets in a particular way (e.g., from multiple threads, or opening and closing batches of sockets) that can reproduce the problem on some systems, and these test programs seem to behave better if a time delay is added between the close and the reopen. The exact circumstances under which the problem occurs is difficult to pin down and seems OS specific, and so choosing the right delay time is very difficult. But it does strengthen this conjecture in my mind. Naturally it would be better if there were a way to determine when a port is available for reuse without actually opening it. I'm not aware of any such way, but I'm holding onto a little hope that one can be found. s'marks On 12/11/14 10:18 AM, Dmitry Samersoff wrote: Stuart, As soon as you close socket, you open a door for the race. So you need another communication channel to pass a port number (or bind result) between a client and a server without closing a socket on the server side. Typical scenario used by network related code is: 1. Server opens the socket 2. Server binds to port(0) 3. Server gets port number assigned by OS 4. Server informs client (e.g. write the port down to known file, broadcast it etc) 5. Client establishes connection. If the server is a blackbox and have to get a port number from outside, scenario looks like: WHILE(!success and !timeout) 1. Driver chooses random port number 2. Driver runs a server with this number 3. Driver checks that server is actually listening on this port (e.g. try to connect by it self) WEND 4. Driver runs a client with this port number or bails out with descriptive error message. -Dmitry On 2014-12-11 20:53, Stuart Marks wrote: On 12/11/14 7:09 AM, olivier.lagn...@oracle.com wrote: On 11/12/2014 15:43, Dmitry Samersoff wrote: You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT SO-LINGER did not help either in my case (see my previous mail to Jaroslav). That ended-up in using another hard-coded (supposedly free) port. Note that was before RMI tests used randomly allocated ports. But there are no reliable way to predict whether you can take this port or not after you close it. This is what I observed in my case. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. IIRC think this is what is currently done in RMI tests. The RMI tests are still suffering from this problem, unfortunately. The RMI test library gets a random port with new ServerSocket(0), gets the port number, closes the socket, then returns the port to the caller. The caller then assumes that it can use that port as it wishes. That's when the BindException can occur. There are about 10 RMI test bugs in the database
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote: On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. Further investigation shows that the problem was rather the client connecting to a socket being shut down. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout state involving the socket address or port. It also turns out that the test does not close the server sockets properly so there might be several sockets being opened or timed out dangling around. I've updated the test so it is setting SO_REUSEADDR for all the new ServerSockets instances + introduced the mechanism to run the test code while properly cleaning up any allocated ports. http://cr.openjdk.java.net/~jbachorik/8066708/webrev.01/ -JB- -JB- s'marks
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Jaroslav, You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT But there are no reliable way to predict whether you can take this port or not after you close it. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. -Dmitry On 2014-12-11 17:06, Jaroslav Bachorik wrote: On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote: On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. Further investigation shows that the problem was rather the client connecting to a socket being shut down. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout state involving the socket address or port. It also turns out that the test does not close the server sockets properly so there might be several sockets being opened or timed out dangling around. I've updated the test so it is setting SO_REUSEADDR for all the new ServerSockets instances + introduced the mechanism to run the test code while properly cleaning up any allocated ports. http://cr.openjdk.java.net/~jbachorik/8066708/webrev.01/ -JB- -JB- s'marks -- Dmitry Samersoff Oracle Java development team, Saint Petersburg, Russia * I would love to change the world, but they won't give me the sources.
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Hi Jaroslav, On 11/12/2014 15:06, Jaroslav Bachorik wrote: Further investigation shows that the problem was rather the client connecting to a socket being shut down. I remember I met this situation for an RMI fix a while ago and IIRC no flag setting could help (SO_REUSEADDR as well), the port kept being unavailable. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout state involving the socket address or port. It also turns out that the test does not close the server sockets properly so there might be several sockets being opened or timed out dangling around. I think this is the main reason why we see these intermittent failures. I've updated the test so it is setting SO_REUSEADDR for all the new ServerSockets instances + introduced the mechanism to run the test code while properly cleaning up any allocated ports. Olivier. On 11/12/2014 15:06, Jaroslav Bachorik wrote: On 12/09/2014 01:25 PM, Jaroslav Bachorik wrote: On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. Further investigation shows that the problem was rather the client connecting to a socket being shut down. It sounds like setting SO_REUSEADDR to false should prevent this failure. From the ServerSocket javadoc: When a TCP connection is closed the connection may remain in a timeout state for a period of time after the connection is closed (typically known as the TIME_WAIT state or 2MSL wait state). For applications using a well known socket address or port it may not be possible to bind a socket to the required SocketAddress if there is a connection in the timeout state involving the socket address or port. It also turns
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 12/11/14 7:09 AM, olivier.lagn...@oracle.com wrote: On 11/12/2014 15:43, Dmitry Samersoff wrote: You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT SO-LINGER did not help either in my case (see my previous mail to Jaroslav). That ended-up in using another hard-coded (supposedly free) port. Note that was before RMI tests used randomly allocated ports. But there are no reliable way to predict whether you can take this port or not after you close it. This is what I observed in my case. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. IIRC think this is what is currently done in RMI tests. The RMI tests are still suffering from this problem, unfortunately. The RMI test library gets a random port with new ServerSocket(0), gets the port number, closes the socket, then returns the port to the caller. The caller then assumes that it can use that port as it wishes. That's when the BindException can occur. There are about 10 RMI test bugs in the database that all seem to have this as their root cause. There is some retry logic in RMI's test library, but that's to avoid the so-called reserved ports that specific RMI tests use, or if new ServerSocket(0) fails. It doesn't have anything to do with the BindException that occurs when the caller attempts to reuse the port with another socket. My observation is also that setting SO_REUSEADDR has no effect. I haven't tried SO_LINGER. My hunch is that it won't have any effect, since the sockets in question aren't actually going into TIME_WAIT state. But I suppose it's worth a try. I don't have any solution for this; we're still discussing the issue. I think the best approach would be to refactor the code so that the eventual user of the socket opens it up on an ephemeral port in the first place. That avoids the open/close/reopen business. Unfortunately that doesn't help the case where you want to tell another JVM to run a service on a specific port. We don't have a solution for that case yet. The second-best approach (not really a solution) is to open/close a serversocket to get the port, sleep for a little bit, then return the port number to the caller. This might give the kernel a chance to clean up the socket after the close. Of course, this still has a race condition, but it might reduce the incidence of problems to an acceptable level. I'll let you know if we come up with anything better. s'marks
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
Stuart, As soon as you close socket, you open a door for the race. So you need another communication channel to pass a port number (or bind result) between a client and a server without closing a socket on the server side. Typical scenario used by network related code is: 1. Server opens the socket 2. Server binds to port(0) 3. Server gets port number assigned by OS 4. Server informs client (e.g. write the port down to known file, broadcast it etc) 5. Client establishes connection. If the server is a blackbox and have to get a port number from outside, scenario looks like: WHILE(!success and !timeout) 1. Driver chooses random port number 2. Driver runs a server with this number 3. Driver checks that server is actually listening on this port (e.g. try to connect by it self) WEND 4. Driver runs a client with this port number or bails out with descriptive error message. -Dmitry On 2014-12-11 20:53, Stuart Marks wrote: On 12/11/14 7:09 AM, olivier.lagn...@oracle.com wrote: On 11/12/2014 15:43, Dmitry Samersoff wrote: You can set SO_LINGER to zero, in this case socket will be closed immediately without waiting in TIME_WAIT SO-LINGER did not help either in my case (see my previous mail to Jaroslav). That ended-up in using another hard-coded (supposedly free) port. Note that was before RMI tests used randomly allocated ports. But there are no reliable way to predict whether you can take this port or not after you close it. This is what I observed in my case. So the only valid solution is to try to connect to a random port and if this attempt fails try another random port. Everything else will cause more or less frequent intermittent failures. IIRC think this is what is currently done in RMI tests. The RMI tests are still suffering from this problem, unfortunately. The RMI test library gets a random port with new ServerSocket(0), gets the port number, closes the socket, then returns the port to the caller. The caller then assumes that it can use that port as it wishes. That's when the BindException can occur. There are about 10 RMI test bugs in the database that all seem to have this as their root cause. There is some retry logic in RMI's test library, but that's to avoid the so-called reserved ports that specific RMI tests use, or if new ServerSocket(0) fails. It doesn't have anything to do with the BindException that occurs when the caller attempts to reuse the port with another socket. My observation is also that setting SO_REUSEADDR has no effect. I haven't tried SO_LINGER. My hunch is that it won't have any effect, since the sockets in question aren't actually going into TIME_WAIT state. But I suppose it's worth a try. I don't have any solution for this; we're still discussing the issue. I think the best approach would be to refactor the code so that the eventual user of the socket opens it up on an ephemeral port in the first place. That avoids the open/close/reopen business. Unfortunately that doesn't help the case where you want to tell another JVM to run a service on a specific port. We don't have a solution for that case yet. The second-best approach (not really a solution) is to open/close a serversocket to get the port, sleep for a little bit, then return the port number to the caller. This might give the kernel a chance to clean up the socket after the close. Of course, this still has a race condition, but it might reduce the incidence of problems to an acceptable level. I'll let you know if we come up with anything better. s'marks -- Dmitry Samersoff Oracle Java development team, Saint Petersburg, Russia * I would love to change the world, but they won't give me the sources.
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 12/09/2014 01:39 AM, Stuart Marks wrote: On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? Hm, this is exactly what happened with this test :( The problem is that the port is reported as available while it is still occupied and RMI registry attempts to start using that port. If setting SO_REUSEADDR does not work then the only solution would be to retry the test case when this exception occurs. -JB- s'marks
RFR 8066708: JMXStartStopTest fails to connect to port 38112
Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Thanks, -JB-
Re: RFR 8066708: JMXStartStopTest fails to connect to port 38112
On 12/8/14 12:35 PM, Jaroslav Bachorik wrote: Please, review the following test change Issue : https://bugs.openjdk.java.net/browse/JDK-8066708 Webrev: http://cr.openjdk.java.net/~jbachorik/8066708/webrev.00 The test fails very intermittently when RMI registry is trying to bind to a port previously used in the test (via ServerSocket). This seems to be caused by the sockets created via `new ServerSocket(0)` and being in reusable mode. The fix attempts to prevent this by explicitly forbidding the reusable mode. Hi Jaroslav, I happened to see this fly by, and there are (I think) some similar issues going on in the RMI tests. But first I'll note that I don't think setReuseAddress() will have the effect that you want. Typically it's set to true before binding a socket, so that a subsequent bind operation will succeed even if the address/port is already in use. ServerSockets created with new ServerSocket(0) are already bound, and I'm not sure what calling setReuseAddress(false) will do on such sockets. The spec says behavior is undefined, but my bet is that it does nothing. I guess it doesn't hurt to try this out to see if it makes a difference, but I don't have much confidence it will help. The potential similarity to the RMI tests is exemplified by JDK-8049202 (sorry, this bug report isn't open) but briefly this tests the RMI registry as follows: 1. Opens port 1099 using new ServerSocket(1099) [1099 is the default RMI registry port] in order to ensure that 1099 isn't in use by something else already; 2. If this succeeds, it immediately closes the ServerSocket. 3. Then it creates a new RMI registry on port 1099. In principle, this should succeed, yet it fails around 10% of the time on some systems. The error is port already in use. My best theory is that even though the socket has just been closed by a user program, the kernel has to run the socket through some of the socket states such as FIN_WAIT_1, FIN_WAIT_2, or CLOSING before the socket is actually closed and is available for reuse. If a program -- even the same one -- attempts to open a socket on the same port before the socket has reached its final state, it will get an already in use error. If this is true I don't believe that setting SO_REUSEADDR will work if the socket is in one of these final states. (I remember reading this somewhere but I'm not sure where at the moment. I can try to dig it up if there is interest.) I admit this is just a theory and I'm open to alternatives, and I'm also open to hearing about ways to deal with this problem. Could something similar be going on with this JMX test? s'marks
