I was referred here by the Chat help.
We seem to have issues with large spikes in response times and corresponding drops in traffic and as the JVM virtually hangs. This seems to be happening at “coldfusion.server.j2ee.pool.ObjectPool:checkIn” “coldfusion.server.j2ee.sql.pool.JDBCPool:checkOut “ and “coldfusion.sql.CFDataSource:getConnection” which can take a ridiculous amount of time. Our Data Centre people are adamant that there is nothing wrong with the Oracle Database (10g) side of things. JVM restarts can on occasion help, but not always.
We are running CF10 HotFix 8, JAVA 1.7.0_60, with Apache 2.2.27 underlying.
Response time vs call drop off.
This reflects in very slow connection requests:
We see a variety of timed_waiting and blocked threads during this time but nothing that indicates the source issue, with most being blocked on “coldfusion.server.j2ee.pool.ObjectPool:checkIn”
Pool setting are as follows from CF Admin:
Any advice on possible causes/solutions would be appreciated?
Did you ever find an answer to this? I've been seeing this on CF2016 for months and have been unable to track down the issue. It doesn't seem to be related to GC as mentioned here https://www.bennadel.com/blog/2580-coldfusion-threads-hanging-on-jdbc-pool-checkin-checkout.htm. Any luck on your end?