I came across a issue today in a 2 node cluster where both nodes were
occupying around 100% CPU usage for no reason. No indexing is happening and
no search queries are being executed.
$sudo strace -p 3003
Process 3003 attached - interrupt to quit
futex(0x7f3696d879d0, FUTEX_WAIT, 3005, NULL
^C <unfinished ...>
Process 3003 detached
When I used "perf top" to identify, I see libjvm.so occupying 90% CPU
usage. I'm using elasticsearch-0.90.0-1.noarch version. Any idea what might
be happening?
I came across a issue today in a 2 node cluster where both nodes were
occupying around 100% CPU usage for no reason. No indexing is happening and
no search queries are being executed.
$sudo strace -p 3003
Process 3003 attached - interrupt to quit
futex(0x7f3696d879d0, FUTEX_WAIT, 3005, NULL
^C <unfinished ...>
Process 3003 detached
When I used "perf top" to identify, I see libjvm.so occupying 90% CPU
usage. I'm using elasticsearch-0.90.0-1.noarch version. Any idea what might
be happening?
Thanks for your reply. Sorry, I forgot to mention the java version. And
yes, I'm using java-1.7.0-openjdk-1.7.0.19-2.3.9.1.el6_4.x86_64.
I'll move to oracle Java 7 now but I'm not sure if that's the problem
because this thing happens only rarely. (Last time it happened a month
ago). Any other thing that might be going wrong here or a known bug or
something?
I came across a issue today in a 2 node cluster where both nodes were
occupying around 100% CPU usage for no reason. No indexing is happening and
no search queries are being executed.
$sudo strace -p 3003
Process 3003 attached - interrupt to quit
futex(0x7f3696d879d0, FUTEX_WAIT, 3005, NULL
^C <unfinished ...>
Process 3003 detached
When I used "perf top" to identify, I see libjvm.so occupying 90% CPU
usage. I'm using elasticsearch-0.90.0-1.noarch version. Any idea what might
be happening?
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.