We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
the owner of ES is root and yes 755 permissions and owned by root. it isnt
going to some other place - there isnt any files/folders called
elasticsearch but the ones in /etc/default and /etc/init.d.
/data/moloch/logs is set for the LOG_DIR
As for the plugin - I put a post on the ES site under github and was told
to ask my questions here. Is there someplace else I should post the
question about the plugin?
On Thursday, February 12, 2015 at 4:36:11 PM UTC-5, Mark Walkom wrote:
You'd have to ask the author of head why it doesn't align.
Does ES have permissions to write to the new log directory?
We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
the owner of ES is root and yes 755 permissions and owned by root. it
isnt going to some other place - there isnt any files/folders called
elasticsearch but the ones in /etc/default and /etc/init.d.
/data/moloch/logs is set for the LOG_DIR
As for the plugin - I put a post on the ES site under github and was told
to ask my questions here. Is there someplace else I should post the
question about the plugin?
On Thursday, February 12, 2015 at 4:36:11 PM UTC-5, Mark Walkom wrote:
You'd have to ask the author of head why it doesn't align.
Does ES have permissions to write to the new log directory?
We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
I have changed the defaults file so it is running as root.
ES_USER=root
ES_GROUP=root
and below is the process - so you can see it is running as root and has the
es.default.path.logs set for /data/moloch/logs.
On Thursday, February 12, 2015 at 1:47:38 PM UTC-5, caspertz wrote:
We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
I have changed the defaults file so it is running as root.
ES_USER=root
ES_GROUP=root
and below is the process - so you can see it is running as root and has
the es.default.path.logs set for /data/moloch/logs.
On Thursday, February 12, 2015 at 1:47:38 PM UTC-5, caspertz wrote:
We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
I have changed the defaults file so it is running as root.
ES_USER=root
ES_GROUP=root
and below is the process - so you can see it is running as root and has
the es.default.path.logs set for /data/moloch/logs.
On Thursday, February 12, 2015 at 1:47:38 PM UTC-5, caspertz wrote:
We have a 4 node ES cluster running version 1.4.1. We are using Ubuntu
and have it set up as a service. We have defined our location for logs as
a specific directory other than /var/log/elasticsearch which is the
default. When this starts, the process shows the value we set for the log
area but no logs are created. It is not due to disk being full.
-Des.default.path.logs=/data/moloch/logs - shows in the process list
Also have found that the /_plugin/head and the _cluster/health do not
always match up. Have see where the plugin shows green while the cluster
health shows yellow. Is this a bug or by design?
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.