Grafana Documentation
Grafana is an open source metric analytics & visualization suite. In this we can create various
dashboards to view and analyze various important website metrics. Grafana uses Graphite as
back end source in our case.
Dashboards created:
1. TCS-ATG-ENDECA-Dashboard
In this dashboard we can monitor various metrics like
ATG-Raw order rate
Desktop Order rate
Skava (Mobile site) order rate
Active session (both app server and bot server)
Order Processing Aux to Manh.
Mobile(vfill)
Desktop Order (vFill)
Endeca Server Response time
ATG server response time
Bot Server response time
Web Servers response time
The URL for this dashboard is:
http://pwegraph01.prod.bn-web.com:3000/dashboard/db/tcs-atg-endeca-dashboard
The date and time range needs to be changed to see the data for the required time period
by using the drop down present on the top right corner of the dashboard screen
There are some predefined relative time ranges also we can set custom time range.
A brief description of important panels available in this dashboard
ATG-Raw order rate
This is the overall order rate of Barnes and Noble ecommerce website. This graph
shows the number of orders placed on the website (both desktop and mobile) in every
two minutes.
We monitor the order rate trend and report to support team in case of any abnormality.
This panel is also used to find the peak hour for maximum number of order placed. This
information is used in preparation of monitoring daily report.
Desktop order rate
This panel shows order trend for desktop website (www.barnesandnoble.com). Similar to
the above panel this panel also shows order count per two minutes.
Skava order rate
The skava order rate panel shows the trend for orders placed per two minutes on mobile
website (m.barnesandnoble.com)
• Physical Servers – FrontEnd
In this panel we can check number of active sessions on a particular server. Currently all
App servers (pweatgap01 to pweatgap14, both store 01 and store 02) are configured in
this panel.
If the active session on a particular server goes high or low then we need to report to
production support team with NOC in cc to look into that server.
ATG app servers are restarted between 2:00 AM EST to 3:00 AM EST daily on a rolling
basis so in this time frame active sessions see a fluctuation.
• ATG-BOT-Active Sessions
This panel shows active sessions on the bot servers (pwebotva01 to pwebotva22). Any
abnormality in the trend needs to be reported.
• Order Processing Aux to Manh.
The Order Processing Aux to Manhattan panel in the dashboard shows the order trend
which goes to Manhattan for processing.
It goes to zero every day around 8:00 AM EST. If at any other time it goes to zero we
need to report.
The order trend in this panel follows a similar trend to ATG- Raw Order Rate.
• Mobile(vfill)
The Mobile(vfill) panel in the dashboard shows the mobile website order trend which
goes to Manhattan for processing.
The order trend in this panel follows a similar trend to Skava Order rate.
• Desktop Order (vFill)
The Desktop Order(vfill) panel in the dashboard shows the Desktop website order trend
which goes to Manhattan for processing. The graph trend is similar to Desktop Order
trend.
• Endeca Server Response time
This panel shows the response time of Endeca servers (pweendmx*)
• ATG server response time
This panel shows the response time of ATG App servers (pweatgap*)
• Bot Server response time
This panel shows the response time of ATG BOT servers (pwebotva*)
• Web Servers response time
This panel shows the response time of ATG Web servers (pweatgws*)
