Resolved [SOLVED] Possible port number conflict - how to change?

Hello,

I encountered a problem today. We've made a copy of our Progress server, changing the obvious things like IP address, etc.
Normally we would be able to start the server, start the DB and Apps server and would be able to use it for testing.
One thing to point out, is that the copy server would be on a different network, so for example the live one would be 172.18.201.32
The copy/test one would be 172.18.200.32. They are normally in physically different locations.

The new copy server is on 172.18.16.32. THis machine is in the same physical location as the Live one.

I know, it's probably odd that I think this may be a conflict. But... I managed to start the Admin Server "proadsv -start".
I managed to start the database:

proserve package -pf broker.pf # Serve the database
probiw package # Enterprise Progress
proapw package # Enterprise Progress
proapw package # Enterprise Progress
proserve package -pf sqlbroker.pf # SQL - Unixware problem
prowdog package # Enterprise Progress

However, when trying to start the Apps server broker:

asbman -i AccordTest -start # App server

It would throw these messages:
OpenEdge Release 11.6.1 as of Fri Feb 19 18:20:45 EST 2016


Connecting to Progress AdminServer using rmi://localhost:20931/Chimera (8280)
Unable to connect to Progress AdminServer at rmi://localhost:20931/Chimera (8287)
Unable to find AccordTest (8281)


One very peculiar thing occurred - the Live Apps server also was affected, basically it went down on 172.18.201.32 .


My question is - is it possible there is a conflict on the side of the port numbers the brokers are using?

It may just be coincidence that the Live server went down as I started the copy one for testing.

Is there any way I can find how to change the port number and change it? For the Application server I found this article:

AdminServer
  • AdminServer Listening Port: -port 20931

I have *.pf files and the port number 20931 is not in any of them. I found the folder /bcp/dlc/properties which looks like the installation
where the Progress sits. I have not found this number in any of the files. So where it the file that would have to be changed?

More bizarrely, here are the date and times of my database being started:

[root@accordvwell test]# ls -ltr package*
-rw-rw-rw-. 1 root root 202 Jan 5 2018 package.st
-rw-r--r--. 1 root root 655360 Feb 14 10:54 package.db
-rw-r--r--. 1 root root 10619977728 Feb 14 10:54 package_7.d1
-r--r--r--. 1 root root 1420768 Feb 14 11:00 package.lic
-rw-r--r--. 1 root root 610402304 Feb 14 11:43 package.b1
-rw-r--r--. 1 root root 11381768192 Feb 14 11:43 package_9.d1
-rw-r--r--. 1 root root 45386563584 Feb 14 11:43 package_8.d1
-rw-r--r--. 1 root root 16252928 Feb 14 11:43 package.d1
-rw-r--r--. 1 root root 16329956 Feb 14 11:43 package.lg

Yet my attempt to start the brokers wasn't even logged:
-rw-rw-rw-. 1 root root 1024021 Feb 3 16:50 AccordLive.broker.019973.log
-rw-rw-rw-. 1 root root 1258040 Feb 3 16:52 AccordLive.server.052239.log
-rw-rw-rw-. 1 root root 2651961 Feb 4 08:58 AccordTest.broker.log
-rw-rw-rw-. 1 root root 5236581 Feb 4 08:58 AccordTest.server.log
-rw-rw-rw-. 1 root root 104571 Feb 4 09:04 AccordLive.broker.019974.log
-rw-rw-rw-. 1 root root 259060 Feb 4 09:04 AccordLive.server.052240.log

I hope this makes sense.


Thanks,
Richard
 
Last edited:

Cecil

19+ years progress programming and still learning.
Just to tick things off. What is the subnet mask i.e. 255.255.0.0?
 

Rob Fitzpatrick

ProgressTalk.com Sponsor
As @Cecil said, with IP addresses but no subnet masks, we don't know whether these hosts share a network segment. The first thing I would do in this situation is remove the test server from production infrastructure. Testing should be done in a separate test environment.

This article describes how to change the AdminServer listening port.
 
Hello,

the test server is on this:

$ ifconfig -a
ens192: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 172.18.16.32 netmask 255.255.252.0 broadcast 172.18.19.255
inet6 fe80::20c:29ff:fe6a:be prefixlen 64 scopeid 0x20<link>
ether 00:0c:29:6a:00:be txqueuelen 1000 (Ethernet)


ens224: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 172.18.206.32 netmask 255.255.255.0 broadcast 172.18.206.255
inet6 fe80::20c:29ff:fe6a:c8 prefixlen 64 scopeid 0x20<link>
ether 00:0c:29:6a:00:c8 txqueuelen 1000 (Ethernet)

The Live server is on:

$ ifconfig -a
ens192: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 172.18.201.32 netmask 255.255.255.0 broadcast 172.18.201.255
inet6 fe80::20c:29ff:fe6a:b4 prefixlen 64 scopeid 0x20<link>
ether 00:0c:29:6a:00:b4 txqueuelen 1000 (Ethernet)

What I didn't realise is that for the Live we have one IP address 172.18.201.32 to connect, the test server has two different interfaces, but were told to connect via 172.18.16.32.

The new machine is a physically different new hardware and the OS is a VM guest.

Thanks you
Richard
 
Last edited:

Cecil

19+ years progress programming and still learning.
i going to ask a stupid question, but there is no static host entry for localhost resolving to 172.18.201.32 in the host file.
 
i going to ask a stupid question, but there is no static host entry for localhost resolving to 172.18.201.32 in the host file.
Hello,

you silly question was well justified, it turned out this had nothing to do with port numbers conflict, etc. Our third party who were supposed to do all the checks had the Live server IP address on this test server.

Thank you all for all the responses.

Richard
 
Top