One of the most common issues experienced by our customers is with collecting endpoint data using our Endpoint Monitor solution using the new Vyopta Collector and the new Vyopta Admin Portal (VAP). For this article, we will walk through a set of troubleshooting steps that can be used to assist with resolving your endpoint collection issues before needing to contact Vyopta Support for assistance.
Pre-checks to be performed before troubleshooting
- Verify that you have access to the Vyopta Admin Portal and that your Vyopta account has been granted the Administrator role.
- Verify that you can RDP into your collector server (or SSH/VNC if you are using the collector on a linux server).
- Please review our Management & Configuration Courses . This is a free web based training course that will provide you an introduction of the new Vyopta Admin Portal. We would strongly suggest checking out our other courses in the Vyopta Learning Center if you haven't already done so.
Review your data collector settings
With the release of our new collector and Vyopta Admin Portal, we have been making several 'behind the scenes' improvements. Before starting troubleshooting, you will need to verify that your collector server is sending data to our new URL. Here are the steps you will need to take to verify that your collector is using the new URL.
1) RDP into your Windows server thats hosting the Collector.
2) Open the file 'vyoptacollector.xml,' which is located in the same folder as 'vyoptacollector.exe', with a text editor. If you are not sure where the file is, you should be able to see the location in the VAP. Clicking on the 'Vyopta Collector' status dropdown should show the Collector Directory, as in the screenshot below.
Once you open the file, you should see an entry for 'server.url' and a 'broker.url' as shown in the screenshot below. If the server.url and broker.url are pointing to the same location as shown in the screenshot, you can close the file, and skip to the next section.
If your server.url is pointing to apps.vyopta.com, please check that you are able to access api.vyopta.com from a web browser on your collector server, since some customers have restrictive access to the internet from the collector server. In the example below the following message can be seen when trying to navigate to https://api.vyopta.com from Chrome. If you see the same message this indicates you are able to connect to the new URL
If you try connecting to api.vyopta.com using Internet Explorer, you may get another message that the webpage cannot be found (below). This is expected since, due to the age of the browser, Internet Explorer will not be able to process the return contents from api.vyopta.com - the Internet Explorer is no longer being developed by MicroSoft. Please check our Supported Browsers for CPM to learn about the latest browsers we support with our CPM Product Suite.
If you are not able to connect to api.vyopta.com, you will need to reach out to your infrastructure/network team to allow access to api.vyopta.com, before making the change to your vyoptacollector.xml file.
Following the same steps above, please test the broker URL from your browser (https://galaxy-admin-mq.vyopta.com) You should see the following response.
If you are able to access both URLs with your browser, please make the change to your server.URL entry as shown in the screenshot below.
The New VAP Endpoint Screen
Below is a screenshot of the new Endpoint screen in the VAP.
This version incorporates new items to assist with troubleshooting endpoints - these are referenced above.
The black arrow is showing a new column called Auth. This column will identify if a username and password pair has been assigned for the endpoint. For us to successfully monitor the endpoint credentials will need to be supplied for the endpoint - if we have credentials, you will see a checkmark for the endpoint. If there is an X in the auth column that indicates there are partial credentials (username or password) or no credentials supplied.
If you click on the validation checkmark for an endpoint that is missing credentials, you will see the following dialog.
If you click on the validation checkmark for an endpoint that is not part of a monitoring group, you will see the following dialog
The white validation arrow (shown by the blue arrow) indicates the endpoint is not part of an access group, or is not able to be validated.
Finally in the STATUS column, when using our new 4.6 collector or later, you will see the actual connection status, and not a generic 'NoMessage' error in this field. It is possible that you will still see a 'NoMessage' status if the endpoint is not able to send us data back through the HTTP Feedback slot - this condition will be addressed later in this document.
We have also enhanced the filtering for endpoints with the new screen. Instead of a single generic search of all endpoints that are registered with Vyopta, you have the ability to filter based on
- Endpoints that have a specific status
- Endpoints that are part of a specific monitoring group
- Endpoints that have complete, or incomplete auth status (missing credentials)
With the new filtering, you can select multiple statuses or groups. For each of the new filters, you can click on the checkbox to filter on the range of status or groups as shown in the screenshot below. In this example, I have selected all endpoints with a status of NoMessages and OK, and Monitoring group of DM and DemoGroup (and one additional group) And for Authentication, where you can select Complete or Missing (or both)
How can I troubleshoot some of the common endpoint issues?
For each of the endpoints you want to monitor with our Endpoint Monitor, you will need to make sure every endpoint entry in the Vyopta Admin Portal has the following
- An IP address.
- Credentials for connecting to the endpoint for monitoring purposes.
- The endpoint entry is assigned to a monitoring group.
- You are able to SSH to the endpoint using the credentials from the server running the collector.
- You are able to access the endpoint from the browser from the collector server, using the supplied credentials. If your IP address for the endpoint is 192.168.4.4, you should be able to connect to https://192.168.4.4/status.xml
1) Registered / Managed endpoints - Check that your Management System is showing the endpoint as available.
For the customers using managed endpoints in CUCM/TMS, the first step is to check that we are getting status data from CUCM/TMS for the registered endpoint. The easiest way to check is by looking at the Endpoint dataset in CPM Monitoring.
In the above screenshot, you will see the status of NotAvailable. If this is the case, you will need to check the management system that the endpoint is registered to, and verify the current status of the endpoint. If the management system is reporting the endpoint as not available (or in some cases as none), we will not be able to directly monitor the endpoint. You will need to work with your team to address the endpoint availability in your management system before you troubleshoot the ability to monitor your endpoint with our endpoint monitor.
VAP Status - And next steps to troubleshoot.
Below is a list of the common statuses we report on for the endpoint, along with next steps you can take to address the issue
OK - We are able to monitor the endpoint successful, no action is required
Blank status - The endpoint is missing credentials or a monitoring group, and therefore the endpoint cannot be monitored. Please verify the endpoint has credentials and is assigned to a monitoring group. If all the endpoints in the monitoring group are reporting a blank status, verify that your service account has been assigned to the monitoring group.
Codec unreachable / NotReachable - This status indicates that we were not able to access the endpoint via SSH or HTTP. Typically this is due to a couple items.
- The endpoint is not available (if managed, check TMS/CUCM on availability)
- The endpoint ports are being blocked between the endpoint and the collector server (test connecting with SSH or a web browser to confirm you are able to connect to the endpoint)
NoMessages- This status message indicates that we are able to connect to the endpoint, but we are not receiving any data from the endpoint. With our new collector we are taking advantage of the HTTPFeedback slots to report streaming data from the endpoint back to the data collector. If you are seeing this message, you will need to check to see if we able to register the data collector with you endpoint.
The easiest way to check is to login to your web console on your endpoint and select Setup | Status, and then select HttpFeedback as shown in the screenshot below
The URL should be pointing back to your collector server's IP address and port 22181. You may see problems if the collector server is using DHCP, the IP address changes on the collector and the endpoint is using the old IP address of the collector as part of the feedback URL, there are no feedback slots available on the endpoint, or if the feedback slots are not enabled on the endpoint.