[ERROR]OpenJDK 64-Bit Server VM warning: Option UseConcMarkSweepGC was deprecated in version 9.0 and will likely be removed in a future release

Elastic Stack Elasticsearch
1

I just set up a GKE ES cluster with 3 master eligible nodes, 3 data nodes, and 2 coordinator nodes.

I haven't created any index yet.

Following is the logging from Stackdriver:

[2019-09-24 23:01:24.158 PDT]
OpenJDK 64-Bit Server VM warning: Option UseConcMarkSweepGC was deprecated in version 9.0 and will likely be removed in a future release.

  • severity: "ERROR"
  • textPayload: "OpenJDK 64-Bit Server VM warning: Option UseConcMarkSweepGC was deprecated in version 9.0 and will likely be removed in a future release.

[2019-09-24 23:01:30.636 PDT]
version[7.3.0], pid[1], build[default/docker/de777fa/2019-07-24T18:30:11.767338Z], OS[Linux/4.14.127+/amd64], JVM[Oracle Corporation/OpenJDK 64-Bit Server VM/12.0.1/12.0.1+12]

I noticed that every time this error occurred, the nodes will leave and rejoin, which happens about 8 ~ 10 times per day.

For example:
[2019-09-24 22:59:15.254 PDT]
node-left[{elasticsearch-pr-master-2}{WewdseokSy6AXRBE-ORPbw}{dAI_yzFcSVSdsO8Drc7QRg}{10.4.4.8}{10.4.4.8:9300}{m}{xpack.installed=true} disconnected], term: 3, version: 31, reason: removed {{elasticsearch-pr-master-2}{WewdseokSy6AXRBE-ORPbw}{dAI_yzFcSVSdsO8Drc7QRg}{10.4.4.8}{10.4.4.8:9300}{m}{xpack.installed=true},}

[2019-09-24 22:59:14.515 PDT]
"stacktrace": ["org.elasticsearch.transport.NodeDisconnectedException: [elasticsearch-pr-master-2][10.4.4.8:9300][disconnected] disconnected"] }

[2019-09-24 22:59:14.564 PDT]
"Caused by: java.net.ConnectException: Connection refused",

[2019-09-24 22:59:14.579 PDT]
"stacktrace": ["org.elasticsearch.transport.ConnectTransportException: [elasticsearch-pr-master-2][10.4.4.8:9300] connect_exception",

[2019-09-24 22:59:14.579 PDT]
"Caused by: io.netty.channel.AbstractChannel$AnnotatedConnectException: Connection refused: 10.4.4.8/10.4.4.8:9300",

[2019-09-24 22:59:14.612 PDT]
"Caused by: java.net.ConnectException: Connection refused",

May I get help on how to fix this?

Thanks!

2

That log line itself will not affect production. It means that a configuration set by Elasticsearch on startup may vanish at some point in the future. If that happens, Elasticsearch will have an alternative configured properly.

However, this log line occuring more than one once has a meaning. This line is only written out, when the JVM (and thus Elasticsearch) is started - this basically means, some component is restarting your container resulting in that log line. This also explains nodes joining and leaving.

3

This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.