What should I know about the Viz Agent?

Relying on periodic observations of data has several advantages over mining for artifact history: 

1. Reduced API load. When Viz initially receives artifact information, the data is typically accurate and only needs to be captured once — reducing the API load on your tools. 
2. No complex time zone calculations. Because you can define the time zone for Viz and that will determine the timestamp of artifact information, there is no need for complex time zone calculations. 
3. Consistent data collection. Though many tools do not provide historical API’s, Viz can consistently retrieve the current state of artifacts to determine changes over time.

A few things need to be in place to send artifact data to Viz:

1. Viz must be running properly.
2. The agent must be running properly.
3. There must be a connection from the agent to external tools.
4. There must be a connection from the agent to Viz.

If any of the above are not in place, you may encounter issues when sending artifact data. 

Note: Planview is responsible for Viz and cloud agent uptime and the cloud agent’s connection to Viz. The customer is responsible for on-prem agent uptime and connections between the agent and tools. The customer is also responsible for ensuring their network allows the on-prem agent to connect to Viz.

If you’d like to learn more, see here.

Below, you can learn more about certain scenarios where data might be lost. 

• Server Uptime: If the server running the on-prem Viz agent crashes, the agent cannot query tools and cannot send data to Viz, resulting in missing data.
• On-prem Agent Connectivity to Viz: If your network does not allow connectivity from the on-prem agent to Viz, data cannot be sent to Viz, thus resulting in missing data. If you’d like to learn more, see here. 
• Improper Credentials: For Viz to query your tools for artifact data, you must first enter credentials for your tools. If you enter the wrong credentials or the credentials do not have proper permissions, data cannot be sent to Viz — resulting in inaccurate Flow Metrics and missing data.
• Improper Certificate: To prevent missing data when using on-prem tools, you may need to import your proxy’s certificate chain to the Viz agent’s JRE keystore if the certificate on the tool or proxy is not trusted. This is only necessary if the certificate on the tool or proxy is not trusted by the out-of-the-box JRE truststore. If the certificate is signed by a common certificate authority and the entire certification trust chain is served, importing the JRE truststore should not be necessary. If you’d like to learn more, see here.
  • Note: If you need to import the proxy certificate chain into your on-prem agent, you will need to connect to your on-prem tools via a non-secure connection. You can select this option on the tool connection screen in Viz. 

The implications of not sending data to Viz depends heavily on the timeframe involved.

While most customers group their Flow Metrics weekly or monthly, some opt to group their data into quarterly buckets. Rarely do we see customers viewing daily Flow Velocity or Flow Load changes. This implies that the granularity of relevant data is at the week or month level. When connectivity interruptions occur at these levels, there may be some noticeable changes in your Flow Metrics.

Below, you can see some implications for various periods when the agent isn’t sending data to Viz.

• Minutes to hours: If the agent cannot query or send data to Viz for a few minutes to a few hours, there should not be a noticeable impact on your Flow Metrics as the agent only queries your tools once every four hours. Simply put, if the agent cannot query or send data during a period when it’s not collecting data, there is no data loss. 
• Multiple hours: If the agent cannot query or send data to Viz for multiple hours, there is a possibility that a query will be missed. However, as soon as the agent can query your tools again, Viz will receive accurate data. For the most part, even if you group your Flow Metrics in Viz at the most granular level, you should not see any difference in the data. 
• A day: If the agent cannot query or send data to Viz for a day, you may notice some changes in your Flow Metrics, although it’s unlikely. If grouping your Flow Metrics by week, your completed work will still contribute to the same week, even if delayed by a day, thus not impacting your Flow Metrics. 
• Days to Weeks: If the agent cannot query or send data to Viz for multiple days to weeks, there will be a noticeable change in your Flow Metrics. Viz will assume that artifacts remained in their last known state for the duration of the outage. Then, once the agent can query and send data to Viz, Viz will assume that any changed Flow Items have changed when Viz sees the change — resulting in inaccurate Flow Metrics.

There are two main differences between the on-prem agent and the cloud agent:

1. The cloud agent does not require you to provision and run your server, whereas the on-prem agent does require provisioning. 
2. The on-prem agent requires that you perform manual, periodic upgrades. With the cloud agent, Viz will automatically upgrade the agent for you when there are changes.

Note: Planview recommends that you use the cloud agent to connect to your tools. 

If you’d like to learn more, click here.