Overview

The remote collector deployment should assist in 2 scenarios:

Data should be collected across several deployments but a consolidate single view is required (only one Local nodes is required).
When a Local Node is reaching a CPU limit and an offload of work is required (can offload up to 20% CPU in high load).

In order to setup a new Remote Collector server you will need to install another new DPOD server based on the prerequisites below. The Node that will contain the Data and the console will be called "Local Node" and the second installation (contains only the Syslog and WS-M agent) will be called "remote collector".

Prerequisites

Two DPOD installations must be with the same version (minimum version is v1.0.7 )
The remote collector DPOD installations should be configured with the "medium" architecture type as detailed in the Hardware and Software Requirements
Each installation will requires some different ports to be opened in the firewall - see table1
There are no requirements regarding the Environment name of each DPOD installation
The two DPODs need to be able to communicate with each other and with the monitored DataPower devices

Setup steps

In order to configure the local node and remote collector(s), run the following script in the local node once per remote collector .

configure_local_node.sh -a <IP address of the remote collector>
For example: configure_local_node.sh -a 192.168.0.5

The script will configure both the local node and remote collector.
Run this script once for each remote collector that you want to add - e.g. if you want to add two remote collectors, run the script twice (in the local node), first time with the IP address of the first remote collector, and second time with the IP of the second remote collector.

Optional parameters:

configure_local_node.sh -a <IP address of the remote collector> -s <initial syslog agents port> -w <initial WSM agents port>
For example: configure_local_node.sh -a 192.168.0.5 -s 70000 -w 70050

The defaults are port 60000 for the initial syslog agents port and 60020 for the initial WSM agents port

Output

Example for a successful execution - note that the script writes two log file, one in the local node and one in the remote collector, the log file names are mentioned in the script's output.

Example for a failed execution, you will need to check the log file for further information.
in case of a failure, the script will try to rollback the configuration changes it made, so you can try to fix the problem and run it again.

After a successful execution, you will be able to see the new remote collectors in the Manage → System → Nodes page,
For example, if we added two remote collectors:

Also, the new agents will be shown in the agents in the Manage → Internal Health → Agents page.
For example, we have one local node with two agents and two remote collectors with two agents each, the page will show six agents:

Configure The Monitored Device to Remote Collector's Agents

It is possible to configure entire monitored device to remote collector's agent or just a specific domain.

To configure monitored device / specific domain please follow instructions on Adding Monitored Devices

Manual Setup Steps

We recommend using the script described in the previous section.
There is no need to take any manual steps if you already run the script.

The following communication and ports are used in a remote collector deployment scenario (table 1). Perform the following commands to accomplish this task on each DPOD local firewall:

Run in Local Node -
Change the XXXX to the IP of the Remote Collector

iptables -I INPUT -p tcp -s XXXX/24 --dport 9300:9309 -j ACCEPT
service iptables save
service iptables restart

After running the commands, run the following command and search the output for two entries showing port 9300 (shown in red in the below screenshot)

iptables -L -n

table 1

From	To	Ports (Defaults)	Protocol	Usage
Local Node DPOD Appliance	Each Monitored Device	5550 (TCP)	HTTP/S	Monitored Device administration management interface
Local Node DPOD Appliance	DNS Server	TCP and UDP 53	DNS	DNS services
Local Node DPOD Appliance	NTP Server	123 (UDP)	NTP	Time synchronization
Local Node DPOD Appliance	Organizational mail server	25 (TCP)	SMTP	Send reports by email
Local Node DPOD Appliance	LDAP	TCP 389 / 636 (SSL). TCP 3268 / 3269 (SSL)	LDAP	Authentication & authorization. Can be over SSL
NTP Server	Local Node DPOD Appliance	123 (UDP)	NTP	Time synchronization
Each Monitored Device	Local Node DPOD Appliance	60000-60009 (TCP)	TCP	SYSLOG Data
Each Monitored Device	Local Node DPOD Appliance	60020-60029 (TCP)	HTTP/S	WS-M Payloads
FROM Users IPs	Local Node DPOD Appliance	443 (TCP)	HTTP/S	Access to with IBM DataPower Operations Dashboard Console
FROM Admins IPs	Local Node DPOD Appliance	22 (TCP)	TCP	SSH
Remote Collector DPOD Appliance	Local Node DPOD Appliance	9300-9309	TCP	DPOD's Store communication
Remote Collector DPOD Appliance	Each Monitored Device	5550 (TCP)	HTTP/S	Monitored Device administration management interface
Remote Collector DPOD Appliance	DNS Server	TCP and UDP 53	DNS	DNS services
Remote Collector DPOD Appliance	NTP Server	123 (UDP)	NTP	Time synchronization
Remote Collector DPOD Appliance	Organizational mail server	25 (TCP)	SMTP	Send reports by email
Remote Collector DPOD Appliance	LDAP	TCP 389 / 636 (SSL). TCP 3268 / 3269 (SSL)	LDAP	Authentication & authorization. Can be over SSL
NTP Server	Remote Collector DPOD Appliance	123 (UDP)	NTP	Time synchronization
Each Monitored Device	Remote Collector DPOD Appliance	60000-60009 (TCP)	TCP	SYSLOG Data
Each Monitored Device	Remote Collector DPOD Appliance	60020-60029 (TCP)	HTTP/S	WS-M Payloads
FROM Users IPs	Remote Collector DPOD Appliance	443 (TCP)	HTTP/S	Access to with IBM DataPower Operations Dashboard Console
FROM Admins IPs	Remote Collector DPOD Appliance	22 (TCP)	TCP	SSH

From the Local Node's UI, go to the Manage menu, select "Nodes" under "System" and click "Edit"

Enter the IP address of the Remote Collector device and click "Update", you can leave the "Agents DNS Address" empty
In the Local Node
Connect to the Local Node DPOD via ssh as root user (using putty or any other ssh client)
Using the Command Line Interface choose option 2 - "Stop All", and wait until all the services are stopped, this may take a few minutes to complete.

In the Local Node
Using putty or any other ssh client, issue the following command:

sed -i -e "s/^SERVICES_SIXTH_GROUP=\".*MonTier-SyslogAgent-1 MonTier-HK-WdpServiceResources MonTier-HK-WdpDeviceResources/SERVICES_SIXTH_GROUP=\"MonTier-HK-WdpServiceResources MonTier-HK-WdpDeviceResources/g" /etc/sysconfig/MonTier

In the Local Node
Using putty or any other ssh client, issue the following command:

mv /etc/init.d/MonTier-SyslogAgent-1 /etc/init.d/Disabled-MonTier-SyslogAgent-1
mv /etc/init.d/MonTier-SyslogAgent-2 /etc/init.d/Disabled-MonTier-SyslogAgent-2
mv /etc/init.d/MonTier-SyslogAgent-3 /etc/init.d/Disabled-MonTier-SyslogAgent-3
mv /etc/init.d/MonTier-SyslogAgent-4 /etc/init.d/Disabled-MonTier-SyslogAgent-4
mv /etc/init.d/MonTier-SyslogAgent-5 /etc/init.d/Disabled-MonTier-SyslogAgent-5
mv /etc/init.d/MonTier-SyslogAgent-6 /etc/init.d/Disabled-MonTier-SyslogAgent-6
mv /etc/init.d/MonTier-SyslogAgent-7 /etc/init.d/Disabled-MonTier-SyslogAgent-7
mv /etc/init.d/MonTier-SyslogAgent-8 /etc/init.d/Disabled-MonTier-SyslogAgent-8
mv /etc/init.d/MonTier-SyslogAgent-9 /etc/init.d/Disabled-MonTier-SyslogAgent-9
mv /etc/init.d/MonTier-SyslogAgent-10 /etc/init.d/Disabled-MonTier-SyslogAgent-10


mv /etc/init.d/MonTier-WsmAgent-1 /etc/init.d/Disabled-MonTier-WsmAgent-1
mv /etc/init.d/MonTier-WsmAgent-2 /etc/init.d/Disabled-MonTier-WsmAgent-2
mv /etc/init.d/MonTier-WsmAgent-3 /etc/init.d/Disabled-MonTier-WsmAgent-3
mv /etc/init.d/MonTier-WsmAgent-4 /etc/init.d/Disabled-MonTier-WsmAgent-4
mv /etc/init.d/MonTier-WsmAgent-5 /etc/init.d/Disabled-MonTier-WsmAgent-5

Note: some errors might appear for services that are not exists in your specific deployment architecture type - for example "mv: cannot stat ‘/etc/init.d/Disabled-MonTier-SyslogAgent-10’: No such file or directory"

In the Local Node
Using any text editor (like vi), edit /etc/hosts files (e.g. vi /etc/hosts)
Change the following entries:
montier-es from 127.0.0.1 to the IP of the Local node device
montier-syslog and montier-wsm to the IP of the remote collector device

you should save the changes when exit (e.g wq)
In the Local Node
Using the Command Line Interface - Select option 1 "Start All", this may take a few minutes to complete
Connect to the Remote Collector DPOD via ssh as root user (using putty or any other ssh client)
Using the Command Line Interface choose option 2 - "Stop All", and wait until all the services are stopped, this may take a few minutes to complete.

In the Remote Collector
Using putty or any other ssh client, issue the following commands:

mv /etc/init.d/MonTier-es-raw-trans-Node-1 /etc/init.d/Disabled-MonTier-es-raw-trans-Node-1
mv /etc/init.d/MonTier-es-raw-trans-Node-2 /etc/init.d/Disabled-MonTier-es-raw-trans-Node-2
mv /etc/init.d/MonTier-es-raw-trans-Node-3 /etc/init.d/Disabled-MonTier-es-raw-trans-Node-3
mv /etc/init.d/MonTier-es-raw-trans-Node-4 /etc/init.d/Disabled-MonTier-es-raw-trans-Node-4

mv /etc/init.d/MonTier-Derby /etc/init.d/Disabled-MonTier-Derby

mv /etc/init.d/MonTier-HK-ESRetention /etc/init.d/Disabled-MonTier-HK-ESRetention

mv /etc/init.d/MonTier-HK-SyslogKeepalive /etc/init.d/Disabled-MonTier-HK-SyslogKeepalive
mv /etc/init.d/MonTier-HK-WsmKeepalive /etc/init.d/Disabled-MonTier-HK-WsmKeepalive

mv /etc/init.d/MonTier-HK-WdpDeviceResources /etc/init.d/Disabled-MonTier-HK-WdpDeviceResources
mv /etc/init.d/MonTier-HK-WdpServiceResources /etc/init.d/Disabled-MonTier-HK-WdpServiceResources

mv /etc/init.d/MonTier-Reports /etc/init.d/Disabled-MonTier-Reports

mv /etc/init.d/MonTier-UI /etc/init.d/Disabled-MonTier-UI

sed -i -e "s/^SERVICES_FIRST_GROUP=\".*/SERVICES_FIRST_GROUP=\"\"/g" /etc/sysconfig/MonTier
sed -i -e "s/^SERVICES_SECOND_GROUP=\".*/SERVICES_SECOND_GROUP=\"\"/g" /etc/sysconfig/MonTier
sed -i -e "s/^SERVICES_THIRD_GROUP=\".*/SERVICES_THIRD_GROUP=\"\"/g" /etc/sysconfig/MonTier
sed -i -e "s/\MonTier-HK-WdpServiceResources MonTier-HK-WdpDeviceResources//g" /etc/sysconfig/MonTier
sed -i -e "s/^SERVICES_SEVENTH_GROUP=\".*/SERVICES_SEVENTH_GROUP=\"\"/g" /etc/sysconfig/MonTier

Note: some errors might appear for services that are not exists in your specific deployment architecture type - for example "mv: cannot stat ‘/etc/init.d/MonTier-es-raw-trans-Node-4’: No such file or directory"

In the Remote Collector
Using any text editor (like vi), edit /etc/hosts files (e.g. vi /etc/hosts)
Change the following entries:
montier-es from 127.0.0.1 to the ip of the Local Node device
In the Remote Collector
Using the Command Line Interface choose option 1 - "Start All", and wait until all the services are stopped, this may take a few minutes to complete.
Verify in the console in Management → Internal health → Agents that all agents are in green state.
Run the following two scripts, you will need to obtain them from IBM support:
in the Local Node - configure_local_node.sh
in the Remote Collector - configure_remote_collector.sh
In the Local Node - !! Only if DPOD was already attached to DataPower Gateways !!
you will need to reconfigure again all the the attached device.

After the setup is complete - DPOD's web console will not longer be available for the Remote Collector, The only way to connect to the Remote Collector will be via ssh client

Browser not supported