Version5.10.0 - Fixing following issue:- WARN | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long

classic Classic list List threaded Threaded
4 messages Options
Sid
Reply | Threaded
Open this post in threaded view
|

Version5.10.0 - Fixing following issue:- WARN | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long

Sid
This post was updated on .
Fixing following issue:- WARN  | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long

Made following change:-
activemq.xml:-
<transportConnector uri="${connection.uri}?wireFormat.maxInactivityDuration=0&amp;keepAlive=true&amp;wireFormat.maxFrameSize=104857600"/>


Client broker url:-
failover:(nio://127.0.0.1:61616?wireFormat.maxInactivityDuration=0&keepAlive=true&wireFormat.maxFrameSize=104857600)?jms.blobTransferPolicy.defaultUploadUrl=http://127.0.0.1:8161/fileserver/&initialReconnectDelay=10000&useExponentialBackOff=false&randomize=false

Please let me know if this is good to go on live..

Thanks,
Sid
Reply | Threaded
Open this post in threaded view
|

Re: Fixing following issue:- WARN | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long

Tim Bain
Setting wireFormat.maxInactivityDuration=0 means you'll never detect that a
client has disappeared without closing the TCP socket cleanly; I'd be very
hesitant about making that change unless you actually need to.  If you're
just trying to avoid getting those lines in your logs, I'm not sure that's
a tradeoff that's worth it.

On Thu, Dec 18, 2014 at 2:14 AM, Sid <[hidden email]> wrote:

>
> Fixing following issue:- WARN  | Transport Connection to:
> blockingQueue_1942843618 failed:
> org.apache.activemq.transport.InactivityIOException: Channel was inactive
> for too (>30000) long
>
> Made following change:-
> activemq.xml:-
> <transportConnector
>
> uri="${connection.uri}?wireFormat.maxInactivityDuration=0&amp;keepAlive=true&amp;wireFormat.maxFrameSize=104857600"/>
>
>
> Client broker url:-
> failover:(nio://
> 127.0.0.1:61616?wireFormat.maxInactivityDuration=0&keepAlive=true&wireFormat.maxFrameSize=104857600)?jms.blobTransferPolicy.defaultUploadUrl=http://127.0.0.1:8161/fileserver/&initialReconnectDelay=10000&useExponentialBackOff=false&randomize=false
>
> Please let me know if this is good to go on live.
>
> Thanks,
> Sid
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/Fixing-following-issue-WARN-Transport-Connection-to-blockingQueue-1942843618-failed-org-apache-activg-tp4689070.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>
Sid
Reply | Threaded
Open this post in threaded view
|

Re: Fixing following issue:- WARN | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long

Sid
This post was updated on .
First of all thanks for responding to this thread.

You are right, I did go through following post:-
http://activemq.2283324.n4.nabble.com/when-should-you-set-maxInactivityDuration-0-td4688281.html

But was wondering whether the setting of flag "keepAlive=true" helps in this case, where wireFormat.maxInactivityDuration=0.

Reference post:-
http://activemq.2283324.n4.nabble.com/ActiveMQ-connection-with-failover-and-prefetch-td3335495.html

I have "failover:()" broker url being configured on producer/consumer(broker clients), hence say I get error " WARN | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long"
on broker then what steps I need to take, because earlier I experienced that after facing above issue, the connectivity is lost among producer, consumer and broker.

So what's recommended next step.

Thanks,
Sid
Reply | Threaded
Open this post in threaded view
|

Re: Fixing following issue:- WARN | Transport Connection to: blockingQueue_1942843618 failed: org.apache.activemq.transport.InactivityIOException: Channel was inactive for too (>30000) long

Tim Bain
Sorry for the very long delay in responding; you may very well have figured
this out since then, but in case not...

As I understand it, the keep-alive is only used if maxInactivityInterval is
non-zero, so I don't think that combination of settings does what you were
thinking it would.  I stand by my earlier recommendation that you're better
off getting these messages in your logs occasionally (and knowing you can
generally ignore them) than turning off inactivity checking entirely...

On Dec 18, 2014 9:30 PM, "Sid" <[hidden email]> wrote:

> First of all thanks for responding to this thread.
>
> Right, I did go through following post:-
>
> http://activemq.2283324.n4.nabble.com/when-should-you-set-maxInactivityDuration-0-td4688281.html
>
> But was wondering whether the setting of flag "keepAlive=true" helps in
> this
> case, where wireFormat.maxInactivityDuration=0.
>
> I have failover broker url being configured on producer/consumer(broker
> clients), hence say I get " WARN | Transport Connection to:
> blockingQueue_1942843618 failed:
> org.apache.activemq.transport.InactivityIOException: Channel was inactive
> for too (>30000) long" error on broker then what steps I need to take,
> because earlier I experienced that after facing above issue, the
> connectivity is lost among producer, consumer and broker.
>
> So what's recommended next step.
>
> Thanks,
> Sid
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/Version5-10-0-Fixing-following-issue-WARN-Transport-Connection-to-blockingQueue-1942843618-failed-org-tp4689070p4689112.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>