https://bz.apache.org/bugzilla/show_bug.cgi?id=63916
Bug ID: 63916
Summary: org.apache.catalina.nonblocking.TestNonBlockingAPI
abysmally slow with NIO/NIO2 on FreeBSD
Product: Tomcat 9
Version: 9.0.x
Hardware: PC
OS: NetBSD
Status: NEW
Severity: major
Priority: P2
Component: Connectors
Assignee: [email protected]
Reporter: [email protected]
Target Milestone: -----
Running all tests regularily from master on
> FreeBSD 12.1-STABLE #1 r353717
on a very capable machine:
> $ sysctl hw.physmem
> hw.physmem: 34200395776
>
> $ sysctl hw.model hw.machine hw.ncpu
> hw.model: Intel(R) Xeon(R) CPU E5-2640 v4 @ 2.40GHz
> hw.machine: amd64
> hw.ncpu: 20
with
> $ for jdk in /usr/local/openjdk* ; do $jdk/bin/java -version; done
> openjdk version "11.0.5" 2019-10-15
> OpenJDK Runtime Environment (build 11.0.5+10-1)
> OpenJDK 64-Bit Server VM (build 11.0.5+10-1, mixed mode)
> openjdk version "12.0.2" 2019-07-16
> OpenJDK Runtime Environment (build 12.0.2+10-3)
> OpenJDK 64-Bit Server VM (build 12.0.2+10-3, mixed mode)
> openjdk version "1.8.0_232"
> OpenJDK Runtime Environment (build 1.8.0_232-b09)
> OpenJDK 64-Bit Server VM (build 25.232-b09, mixed mode)
with the same horrible results over and over again (see tests.csv marked with X
in last col):
> org.apache.catalina.nonblocking.TestNonBlockingAPI;APR;9;0;0;1;37,114;X
> org.apache.catalina.nonblocking.TestNonBlockingAPI;NIO;9;1;0;1;2215,246;X
> org.apache.catalina.nonblocking.TestNonBlockingAPI;NIO2;9;1;0;1;2208,712;X
APR requires less than 40 s to complete, NIO more than 2200 s.
I have attached the test output as well as two thread dumps with 30 s apart
during tests while it is completely blocked.
I wonder wether this issue is present in some form on macOS using the same
KQueue NIO impl at least in Java 8.
It is not really clear to me: is it the test, Tomcat or a JVM on FreeBSD bug. I
am not really skilled in reading thread dumps.
Any pointers I can start digging? I'd collect the appropriate data and contact
the FreeBSD Java port maintainer.
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
