You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@activemq.apache.org by jliao <jl...@riotgames.com> on 2013/06/27 00:53:22 UTC

ActiveMQ transport error when I do a telnet check on the port

I'm using ActiveMQ 5.5, and I have some monitoring going against the broker
port to perform a TCP port check, and somehow I'm getting all these errors
in my logs:


WARN  | 06-26-2013 15:00:41.055 | Async Exception Handler |
org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
Connection to: /10.20.30.40:51534 failed:
org.apache.activemq.transport.InactivityIOException: Cannot send, channel
has already failed: /10.20.30.40:51534
WARN  | 06-26-2013 15:01:40.715 | ActiveMQ NIO Worker |
org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
Connection to: /10.20.30.40:51545 failed: java.io.EOFException
WARN  | 06-26-2013 15:01:40.715 | Async Exception Handler |
org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
Connection to: /10.20.30.40:51545 failed:
org.apache.activemq.transport.InactivityIOException: Cannot send, channel
has already failed: /10.20.30.40:51545
WARN  | 06-26-2013 15:02:40.502 | ActiveMQ NIO Worker |
org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
Connection to: /10.20.30.40:51561 failed: java.io.EOFException


The 10.20.30.40 is the IP for the ActiveMQ broker, and the TCP port checking
is 61656. This error can be repro by just simply doing a "telnet 10.20.30.40
61656". So does anyone know what might be causing this error?



--
View this message in context: http://activemq.2283324.n4.nabble.com/ActiveMQ-transport-error-when-I-do-a-telnet-check-on-the-port-tp4668611.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re:Re: ActiveMQ transport error when I do a telnet check on the port

Posted by SuoNayi <su...@163.com>.
Or just giving a higher logging level to suppress those annoying logs?
We also use TCP port check to see if our brokers are available and I just ignore those logs.


At 2013-06-27 08:21:46,jliao <jl...@riotgames.com> wrote:
>are you referring to the transportConnectors config in the ActiveMQ itself?
>Unfortunately I don't think I can change it. Is there any other alternative?
>
><transportConnectors>
>    <transportConnector name="openwirenio"
>uri="nio://0.0.0.0:61656?useQueueForAccept=false"/>
></transportConnectors>
>
>
>
>--
>View this message in context: http://activemq.2283324.n4.nabble.com/ActiveMQ-transport-error-when-I-do-a-telnet-check-on-the-port-tp4668611p4668620.html
>Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: ActiveMQ transport error when I do a telnet check on the port

Posted by jliao <jl...@riotgames.com>.
are you referring to the transportConnectors config in the ActiveMQ itself?
Unfortunately I don't think I can change it. Is there any other alternative?

<transportConnectors>
    <transportConnector name="openwirenio"
uri="nio://0.0.0.0:61656?useQueueForAccept=false"/>
</transportConnectors>



--
View this message in context: http://activemq.2283324.n4.nabble.com/ActiveMQ-transport-error-when-I-do-a-telnet-check-on-the-port-tp4668611p4668620.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.

Re: ActiveMQ transport error when I do a telnet check on the port

Posted by Christian Posta <ch...@gmail.com>.
when you open a socket to the broke port, there are a series of protocol
negotiations expected, and this can cause errors if the protocol isn't
followed. you may try stomp transport for your telnet session?
a tcp port check could cause those errors because port is opened/closed or
even just half-opened (SYN port scan) before the broker can respond.


On Wed, Jun 26, 2013 at 3:53 PM, jliao <jl...@riotgames.com> wrote:

> I'm using ActiveMQ 5.5, and I have some monitoring going against the broker
> port to perform a TCP port check, and somehow I'm getting all these errors
> in my logs:
>
>
> WARN  | 06-26-2013 15:00:41.055 | Async Exception Handler |
> org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
> Connection to: /10.20.30.40:51534 failed:
> org.apache.activemq.transport.InactivityIOException: Cannot send, channel
> has already failed: /10.20.30.40:51534
> WARN  | 06-26-2013 15:01:40.715 | ActiveMQ NIO Worker |
> org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
> Connection to: /10.20.30.40:51545 failed: java.io.EOFException
> WARN  | 06-26-2013 15:01:40.715 | Async Exception Handler |
> org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
> Connection to: /10.20.30.40:51545 failed:
> org.apache.activemq.transport.InactivityIOException: Cannot send, channel
> has already failed: /10.20.30.40:51545
> WARN  | 06-26-2013 15:02:40.502 | ActiveMQ NIO Worker |
> org.apache.activemq.broker.TransportConnection.Transport(202) - Transport
> Connection to: /10.20.30.40:51561 failed: java.io.EOFException
>
>
> The 10.20.30.40 is the IP for the ActiveMQ broker, and the TCP port
> checking
> is 61656. This error can be repro by just simply doing a "telnet
> 10.20.30.40
> 61656". So does anyone know what might be causing this error?
>
>
>
> --
> View this message in context:
> http://activemq.2283324.n4.nabble.com/ActiveMQ-transport-error-when-I-do-a-telnet-check-on-the-port-tp4668611.html
> Sent from the ActiveMQ - User mailing list archive at Nabble.com.
>



-- 
*Christian Posta*
http://www.christianposta.com/blog
twitter: @christianposta

Re: ActiveMQ transport error when I do a telnet check on the port

Posted by Tim Bain <tb...@alumni.duke.edu>.
Great, thanks for sharing the solution that worked for you.

Tim

On Mon, Apr 9, 2018, 6:01 AM jroel <je...@alsic.be> wrote:

> Hello,
>
> We 've resolved this issue with a custom bash script involving the
> following
> command "nmap -sS -p PORT IPADDRESS" instead of the telnet variant.
>
> This command does not cause the errors in the logging.
>
>
>
>
>
> --
> Sent from:
> http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html
>

Re: ActiveMQ transport error when I do a telnet check on the port

Posted by jroel <je...@alsic.be>.
Hello,

We 've resolved this issue with a custom bash script involving the following
command "nmap -sS -p PORT IPADDRESS" instead of the telnet variant.

This command does not cause the errors in the logging.





--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html

Re: ActiveMQ transport error when I do a telnet check on the port

Posted by Tim Bain <tb...@alumni.duke.edu>.
As Christian Posta wrote in the first response, if you (or your load
balancer) are trying to determine the health of the broker by simply
opening a TCP socket to the OpenWire port and then closing it, you're going
to get these warnings in the broker logs.

Unfortunately, I'm not sure there's a great built-in option for your load
balancer to check that won't do this. Christian suggested trying a health
check on the STOMP port, which you could try, or you could open a port for
the HTTP transport and check that. I haven't personally done either of
those things, so you'd have to see if they did what you wanted.

Tim

On Thu, Apr 5, 2018 at 7:56 AM, jroelens <je...@alsic.be> wrote:

> Hello,
>
> Did you ever found a resolution?
> We're having the same problem....
>
>
>
> --
> Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-
> f2341805.html
>

Re: ActiveMQ transport error when I do a telnet check on the port

Posted by jroelens <je...@alsic.be>.
Hello, 

Did you ever found a resolution?
We're having the same problem....



--
Sent from: http://activemq.2283324.n4.nabble.com/ActiveMQ-User-f2341805.html