Occasional High CPU Usage from beam.smp with no activity
Since 1.8 we've noticed the occasional high CPU usage from one box out of three in our couchbase cluster due to beam.smp. These servers get little to no use since the project we were going to use couchbase for was basically scrapped, however it's still running for legacy purposes and we occasionally get CPU alerts. I thought upgrading to 2.0 might solve this issue, but it's still happening. I had our alerts log out a "ps aux --sort pcpu" and this is what I keep seeing:
deploy 5938 30.7 1.7 1736088 1138120 ? Sl Mar14 1298:44 /opt/couchbase/lib/erlang/erts-5.8.5/bin/beam.smp -S 16:16 -sbt u -P 327680 -K true -- -root /opt/couchbase/lib/erlang -progname erl -- -home /home/ptravers -- -smp enable -setcookie nocookie -kernel inet_dist_listen_min 21100 inet_dist_listen_max 21299 error_logger false -sasl sasl_error_logger false -noshell -noinput -noshell -noinput -run ns_bootstrap -- -couch_ini /opt/couchbase/etc/couchdb/default.ini /opt/couchbase/etc/couchdb/default.d/capi.ini /opt/couchbase/etc/couchdb/default.d/geocouch.ini /opt/couchbase/etc/couchdb/local.ini -ns_server config_path "/opt/couchbase/etc/couchbase/static_config" -ns_server pidfile "/opt/couchbase/var/lib/couchbase/couchbase-server.pid" -ns_server nodefile "/opt/couchbase/var/lib/couchbase/couchbase-server.node" -ns_server cookiefile "/opt/couchbase/var/lib/couchbase/couchbase-server.cookie"
Any ideas what might be causing this?
We do use cpu periodically from various background activities. Like gathering stats or gossiping between nodes. Some cpu usage is expected. We are aware that it's a bit on high side especially for idle or mostly idle setups and especially on lower-end hardware. We're working on that. BTW upcoming 2.0.1 already has some improvements over 2.0. And 2.0 was a bit of regression compared to 1.8.1.