"Database Not Responding" Message Appears in Global Scan

This knowledge base article provides steps of one approach which may resolve a "Database Not Responding" (DNR) message in the Global Scan.

When a DNR message appears on the Global Scan page, check the uptime.log file for messages similar to the following line (numbers will likely be different from example):

011-03-29 11:40:48,873 ERROR (ProtocolHandler:79) - In command entities=168,169,171,175,176,177,179|commandType=query|commandQueryName=GET_GLOBAL_SCAN_ENTITIES

This type of message likely indicates that the Global Scan page is attempting to display recent performance data from an element that has been removed or deleted from up.time.  To verify this condition, run the following queries to see if there are entries in the entity_latest_data_sample table that do not correspond to an existing element in the entity table:

select host_id from entity_latest_data_sample where host_id not in (select entity_id from entity);

For each ID that is listed from the above query, run the following two queries (replace <ID> with the ID number).  The first query should return an empty set but the second should have one or more results.

select entity_id from entity where entity_id = <ID>;
select * from entity_latest_data_sample where host_id = <ID>;

To remove the erroneous performance data that is causing this issue, run the following delete statement.

delete from entity_latest_data_sample where host_id = <ID>;

Once the delete statement has been run for each of the ID's from the first query, view the GlobalScan page to verify that the issue has been resolved.