Performance of Oracle CLOB reads in EAP
Issue
Several thread dumps taken over the course of a few minutes show nearly identical stacks (for one or more threads) like the below
http-/10.0.0.x:8080-1" daemon prio=10 tid=0x00007f42144b2800 ...
java.lang.Thread.State: RUNNABLE
at java.net.SocketInputStream.socketRead0(Native Method)
...
at oracle.net.ns.NetInputStream.read(NetInputStream.java:260)
at oracle.jdbc.driver.T4CSocketInputStreamWrapper.read(T4CSocketInputStreamWrapper.java:105)
...
at oracle.jdbc.driver.T4C8TTIClob.read(T4C8TTIClob.java:246)
at oracle.jdbc.driver.T4CConnection.getChars(T4CConnection.java:3187)
- locked <0x0000000787a94c10> (a oracle.jdbc.driver.T4CConnection)
at oracle.sql.CLOB.getChars(CLOB.java:459)
at oracle.sql.CLOB.getSubString(CLOB.java:321)
...
Environment
- Red Hat JBoss Enterprise Application Platform (EAP) 6
- Oracle
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.