WFLYPRT0023: Could not connect to remote://10.111.10.123:9990. The connection timed out

Latest response

I am using JBoss EAP 7.3 update 3.

RHEL 7.x 64-bit O/S

JBoss implementation in domain mode between two servers: master host lnx-08, slave host lnx-07

When starting the JBoss servers on the master server all servers start fine. No issues. When starting the Jboss servers on lnx-07 (slave) I receive the following error in the process-controller.log:
WFLYPRT0023: Could not connect to remote://10.111.10.123:9990. The connection timed out.

Here are the interface entries in the domain.xml:

10.111.10.123 is the ip for lnx-08.

As you can see I can ping lnx-08 from lnx-07 successfully:
[jboss@lnx-isat-shiv07 log]$ ping 10.101.10.123
PING 10.101.10.123 (10.111.10.123) 56(84) bytes of data.
64 bytes from 10.111.10.123: icmp_seq=1 ttl=64 time=0.462 ms
64 bytes from 10.111.10.123: icmp_seq=2 ttl=64 time=0.413 ms

And I can telnet over port 9990, but I do see an HTTP/1.1 400 Bad Request message:
[jboss@lnx-isat-07 log]$ telnet 10.101.10.123 9990

Trying 10.111.10.123...
Connected to 10.101.10.123.
Escape character is '^]'.
HTTP/1.1 400 Bad Request
Content-Length: 0
Connection: close
Connection closed by foreign host.

Firewall / port 9990 is open on lnx-07 (and lnx-08):
[root@lnx-isat-shiv07 ~]# firewall-cmd --list-all
public (active)
target: default
icmp-block-inversion: no
interfaces: ens161 ens192 ens224 ens256
sources:
services: dhcpv6-client ssh
ports: 7001/tcp 9990/tcp
protocols:
masquerade: no
forward-ports:
source-ports:
icmp-blocks:
rich rules:

And finally, as you can see, the Jboss servers on the master (lnx-08) are running:
[jboss@lnx-isat-shiv08 system]$ netstat -ln | grep ':9990 ' | grep 'LISTEN'
tcp 0 0 10.111.10.123:9990 0.0.0.0:* LISTEN

Any suggestions why this may be happening? This is a development env and was working until a reboot of the lnx-07 server. I dont have Linux support.

Thanks in advance
Andy

Responses

Apologies, I missed the interfaces section. Details below:

....

Hi Andrew,

It is safer to remove the real ip-address and to replace the with the phrase "myserver-ip"

Regards,

Jan Gerrit

Hi Andrew,

It must an issue in the JBoss setup, for it is clear that on tcp level the connection is active.

A http error means that a connection has been established, but disconnected due to an unexpected http request.

Regards,

Jan Gerrit

test the multicast connection.

Hi Andrew,

As Ian Laws is mentioning multicast connection you may need to get your network management team involved to check if multicast is still working.

Regards,

Jan Gerrit

Hi Jan The ip addresses are not real, I've replaced the values, I just thought it would be easier to understand presented like this. So initially you say it must be an issue in the jboss setup, but I think you are now saying check the multicast-connection first. Is this right? Thanks Andrew

Hi Andrew,

If the ip's are "faked", than there is no issue.

Which issue to tackle first: * multi cast connection * JBoss issues

I cannot determine, where a change may have been triggered due to the reboot I do not know.

My gut feeling says a JBoss deployment might have been triggered.

A network issue could be, if the network change has been done during your reboot.

Regards,

Jan Gerrit