We have been looking into it, but do also have other tasks.

The stacktrace shows the client was trying to ready a TextMessage for
delivery via onMessage but was unable to do so as it failed to decode
the text content as UTF-8 while trying.

The stacktrace indicates the text payload was contained within a raw
Data section, which is why it only failed at this location rather than
earlier where the overall message was first decoded and a failure of a
more typical string typed payload would have been handled differently.
Though not particularly important, this implies the message wasn't
sent with Qpid JMS originally since it typically doesnt send those
sections other than for BytesMessage, and if so the message was
indicating in some way that it contained textual Data for the client
to treat it as a TextMessage (e.g through a content-type or a message
type annotation).

When investigating this I did notice that this particular type of
failure would trigger invocation of the ExceptionListener (if there is
one) whilst holding the delivery lock and though that isnt really a
bug, in concert with the mish mash of your application locking and its
cross-connection thread usage it likely contributed to your observed
triple-thread deadlock. It is something we are looking into possibly
improving.

Ultimately the root cause is the message payload and it not being able
to be decoded. You would need to look at what is being sent / received
and identify in what way it is unexpected. You can try tools such as
Wireshark or perhaps the clients logging, e.g see
http://qpid.apache.org/releases/qpid-jms-0.51.0/docs/index.html#logging
for frame tracing details. You can control the length of transfer
frame payload proton-j outputs using the
proton.trace_frame_payload_length system property, if you need to
increase the output shown to see the affected content. You can also
have raw traffic bytes logged from Netty by setting
"transport.traceBytes=true" URI option and having logger
org.apache.qpid.jms.transports.netty.NettyTcpTransport set to debug
level.

Robbie

On Wed, 13 May 2020 at 15:28, akabhishek1
<[email protected]> wrote:
>
> Hi Rob/Team,
>
> Could you please take a look on this issue?
>
> At this moment, i am not able to replicate this issue.
>
> Can you please provide any clue to generate  'Cannot decode String in UTF-8'
> error?
>
> So i can generate 'Suspected deadlocked threads' error and share with you
> details.
>
> Regards,
> Abhishek
>
>
>
> --
> Sent from: http://qpid.2158936.n2.nabble.com/Apache-Qpid-users-f2158936.html
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to