First off, hope everyone and their families are doing well.
I am having a difficult time interpreting what is happening with a Database Connection Monitor that I am using. I want to get further insight into how I can troubleshoot this from a Solarwinds perspective. What I am being told from the DB team, is that they are seeing nothing go down and have 100% uptime. What I am being told, is they are performing a check for the DBs uptime. My thought is that the DB can have 100% uptime, but does not mean it is always accessible. My example would be that my Cisco 9300 has 100% uptime, but does not mean I was able to access it 100% of the time from a network connection. I am still very new with DBs so my example may be a poor one, or not even applicable, so disregard if needed.
For the specifics on this component monitor, this component only goes down for five minutes 99% of the time. My polling frequency and polling timeout are both set to five minutes for this Application Monitor. This Database Component Monitor has a query timeout of 30 seconds, the connection string is Driver={%omitted%};Database=%omitted%;Hostname=${IP};Port=50000; Protocol=TCPIP;Uid=${USER};Pwd=${PASSWORD};, has a SQL query of Select count(*) from syscat.tables with ur and its Preferred Polling Method for this application is Agent.
Would increasing the SQL query timeout from 30 seconds to a higher amount help this issue? I know I could adjust the alert to require this threshold to be met for ten minutes, ultimately making this alert trigger less often and make the DB team happier, plus make their availability numbers increase, but I would like to understand this issue as best as I can before trying to push the time limit farther out.
Also, I was getting a little stumped by the Polling Frequency and Polling Timeout, but I believe I understand this correctly now. For example, I poll now and then five minutes later, if I get no response/timeout, Solarwinds considers the Application down. Then the poll would occur again due to the five minute time frame elapsing. Then I believe the process would repeat as normal without any fast polling mechanism. I do not believe that the SAM Application Monitor have a fast polling mechanism to them, but I wanted to mention this in case they do.
Any information to help me would be greatly appreciated. Thank you!
