Greetings,
Last night Aru and I were able to track down the source of the five queries I mentioned above.
One query was the result of a long-standing bug in our GCVS object name resolution code. I’ve implemented a patch for this, but we haven’t deployed it as the same issue appears in multiple locations in our code. We’re refactoring the code right now.
The other four queries were caused by bots indexing the (paginated) data search application, resulting in hundred of queries against the AID per bot. Aru has made a change that mitigates this issue.
This morning I also found the VSX server’s network connection was down. This machine hosts several of our APIs, so it likely caused VStar and TA to not function correctly. This is a really odd issue because VSX is a virtualized server in a major datacenter. Nevertheless, a reboot got it up and running.
At the moment, the database, www, apps, and vsx are all functional. Please let me know if any further issues persist.
Again, I apologize for the service outage and appreciate your understanding regarding these matters.
Brian