The xml I prepare contains all the thousands of devices being monitored regardless it is monitored for alarms or QOS.
<Devices><Device><PrimaryDnsName>Tentent1 SHORT HOSTNAME1</PrimaryDnsName><UserAlias>cm_data_import_SHORT HOSTNAME1</UserAlias><Origin>TenantOrigin1</Origin><PrimaryIPV4Address>18.104.22.168</PrimaryIPV4Address><PrimaryDnsName>Tentent2 SHORT HOSTNAME3</PrimaryDnsName><UserAlias>cm_data_import_SHORT HOSTNAME3</UserAlias><Origin>TenantOrigin2</Origin><PrimaryIPV4Address>22.214.171.124</PrimaryIPV4Address></Device></Devices>
But the Launch Relationship Viewer from the Actions menu in the Unified Service Manager (USM) portlet is missing.
Currently it is not showing up at the action menu for all the groups, but when I add a Relationship Viewer portlet to a UMP page it shows the topology default view where I can find lots of device in the groups and not in the groups.
Today I use Probe Utility remove_all_master_devices and rerun restart discovery_server probe and waited for 40 min for the auto_discovery Inventory to be populated. Then I use raw config on the discovery_server to set purged from the database from default 30 to 0 days then rerun the discovery_agent without importing xml through cm_data_import still the Launch Relationship Viewer from the Actions menu did not appear. The devices I remove by remove_master_devices_by_cskeys still seem to show up in the topology . I am not sure how the topology completely get it data from to be populated.
I would like to know how do I rebuild or clean up topology data totally.
It seem cm_computer_system and cm_device are not where it getting the data from.
I am running UIM 8.5 , discovery_server and discovery agent version 8.5, relationship services and topology_agent version 1.72.
Here are the steps for resetting your topology/relationship database and kicking off a new topology rediscovery allowing for the capture of traces that can be used to identify any problems should there be no changes in the connections in your topology map in the UMP RelationshipViewer portlet. 1. For all of your topology_agent probes from the probe's Raw Configure GUI, select the setup folder from the left-hand pane, then Modify/Add the following key values from the right-hand pane: loglevel = 5 logsize = 80000 Then select the task-config folder from the left-hand pane and Modify the following keys in the right-hand pane: discover-on-startup = true snmp-walk-enable = true Apply the changes, 2. Deactivate all topology_agents on all of your hubs 3. Deactivate the fault_correlation_engine on the primary hub 4. From the relationship_services Raw Configure GUI, select the setup folder from the left-hand pane, then Modify/Add the following keys in the right-hand pane: loglevel = 5 logsize = 80000 Apply the changes. 5. Deactivate the relationship_services on the primary hub 6. Delete/Rename the <installPath>\Nimsoft\probes\service\relationship_services\database directory on the primary hub (this is the topology/relationship database we will rebuild) 7. Activate relationship_services on the primary hub 8. Activate the fault_correlation_engine on the primary hub 9. Activate the topology_agent on each of your hubs. 10. Allow a full discovery cycle to complete. This can take some time, one to three hours is generally enough. To verify that it has completed, look for the message “total discovery elapsed time” in the topology_agent.log files.
This might also help to re-building topology data.
1. Deactivate topology_agent probe.
2. Deactivate relationship_services probe.
3. Move(or remove) the following files.
4. Change the following key (as follows) in topology_agent.cfg
<task-config>/discover-on-startup = true
5. Activate relationship_services probe.
6. Activate topology_agent probe.
I don't have access by terminals of Windows or command line to one of the hub where one of the topology_agent .
If topology is dependent on files in /probes/service/topology_agent/ files I will need to delete the topology_agent and redeploy to that hub.
Will this work in removing the files from the directory on that hub ?
Unfortunately, probe deletion operation (via Product GUI) does not guarantee all files in the probe folder got cleared.
Thanks.I will try it out tomorrow morning when I start work again.
It seems relationship_services holds the topology default database. Topology_agent only provides data provided by the discovery_agent discovery successful protocol discovery.
I would like to know can I still use cm_data_import approach that I use in How can I change the discovery entries origin attribute to correlation the values ( by probe example net_connect dev_id) and discovery_agent dev_id with a new dev_id by cm_data_import in the cm_device table to correct wrong naming, ip address and naming in the USM to display as intended list device naming , ip address and origin without affect Launch Relationship Viewer from the Actions menu in the Unified Service Manager (USM) portlet.?
Thanks. I have rebuild the topology and can see that through relationship viewer portlet that there is changes as it is being build but the link for Launch Relationship Viewer from the Actions menu still does not show up for each USM group.
When I add a Relationship Viewer to a Portal page it show devices and robots but I don't see the usual network connectivity for Relationship Selector.