We are using version 12.9.
Recently there have been occurrences of scoreboard showing phantom tickets i.e. scoreboard shows ticket count which is 1 or 2 more than the actual tickets seen when clicked on the scoreboard node.
The correct count is shown automatically after some time.
Any idea what could be causing this?
Hi Chirag - is this happening on an out-of-the-box scoreboard node or is this a custom scoreboard query that is having the issue? If so, can you share the custom scoreboard query that is being used by that node?
2 different nodes have the issue. These are custom nodes having below queries (have replaced ids with ***):
When the admin opens the dashboard by changing the role, the correct count is shown. However, all agents having that specific role face the issue.
(active = 1) AND status IN (\'OP\') AND (type = \'I\') AND (log_agent = U\'***\' OR log_agent = U\'***\') AND NOT tenant.name IN (\'***\') AND (group!=U\'***\' AND group!=U\'***\')
group=U\'***\' AND status=\'OP\' AND type = \'I\'
These nodes haven't been modified recently and have been created few months ago. The issue is being seen only now. So not sure if it has something to do with the queries.
Plesse check if you habe monitor joins inactive.
If so, this common behaviour.
hi, the "tenant.name" might be the problem when monitor_joins is off.
So please check in the nx.env file for this parameter:
# Set this variable to yes if you want to enable refreshing lists when an# update is made to a table referenced by a join in the where clause.@NX_MONITOR_JOINS=No
Please verify if the 'problem' is solved when you change it to Yes.
But keep in your mind, this will have performance impacts.
Thanks. This is good to know though I wouldn't go for this option since there are many articles advising against it. Performance impact is not something that would be affordable.
I see the same issue but in a different way. We have an interface system between CA and an SM9 system. Sometimes the interface doesn't always work correctly and we use SQL to manually update statuses of tickets to sync the tickets because we do not want another XML sent across the system. If we do not refresh the call_req table after updating the status (usually resolved to closed as well as updating times and inactive status) the scoreboard can reflect a different number than what is displayed upon clicking the query for the results.
pdm_cache_refresh -t Call_Req
Personally I would try to refresh the ca_contact table and see if that works to fix your scoreboard problem. (or any of the other associated tables)
pdm_cache_refresh -t ca_contact
I had this in mind for some time but haven't got a chance to do it yet since the issue keeps disappearing.
Will execute these commands the next time I catch the issue.
Thanks for sharing!