├── gcplot ├── gcserver │ ├── gcplot.properties │ ├── gcserver │ └── logback.xml ├── cassandra │ ├── cassandra-topology.properties │ ├── cassandra-rackdc.properties │ ├── cdb.cql │ ├── jvm.properties │ └── cassandra.yaml ├── nginx │ ├── nginx │ └── gcplot.conf ├── start.sh ├── orientdb │ ├── orientdb.sh │ ├── orientdb-server-log.properties │ └── orientdb-server-config.xml └── Dockerfile └── README.md /gcplot/gcserver/gcplot.properties: -------------------------------------------------------------------------------- 1 | orientdb.connection.string=remote:localhost:2424/gcplot 2 | orientdb.user=admin 3 | orientdb.password=admin 4 | 5 | bootstrap.server.host=0.0.0.0 6 | bootstrap.server.port=9091 7 | 8 | hosts.group=docker 9 | app.hostname=localhost 10 | app.hostaddr=127.0.0.1 11 | worker.tasks.enabled=false 12 | 13 | graphite.sink.enabled=false 14 | graphite.report.every.ms=30000 15 | graphite.host= 16 | graphite.port=2003 17 | 18 | gc.repository.type=cassandra 19 | 20 | cassandra.hosts=localhost 21 | cassandra.port=9042 22 | cassandra.username=cassandra 23 | cassandra.password=cassandra 24 | cassandra.keyspace=gcplot 25 | cassandra.reconnect.delay.ms=100 26 | 27 | gc.cassandra.fetchSize=5000 28 | 29 | s3.resource.bucket= 30 | s3.connector.bucket= 31 | s3.connector.path.base=connector-logs 32 | 33 | files.upload.dir=/tmp/gcserver-file-uploads 34 | 35 | worker.task.distributor.interval.ms=120000 36 | worker.processing.interval.ms=15000 37 | 38 | vertx.worker.pool.size=50 39 | vertx.worker.blocking.pool.size=150 40 | 41 | ui.host.protocol=http 42 | ui.host=localhost 43 | 44 | max.upload.file.size=2147483654 45 | -------------------------------------------------------------------------------- /gcplot/cassandra/cassandra-topology.properties: -------------------------------------------------------------------------------- 1 | # Licensed to the Apache Software Foundation (ASF) under one 2 | # or more contributor license agreements. See the NOTICE file 3 | # distributed with this work for additional information 4 | # regarding copyright ownership. The ASF licenses this file 5 | # to you under the Apache License, Version 2.0 (the 6 | # "License"); you may not use this file except in compliance 7 | # with the License. You may obtain a copy of the License at 8 | # 9 | # http://www.apache.org/licenses/LICENSE-2.0 10 | # 11 | # Unless required by applicable law or agreed to in writing, software 12 | # distributed under the License is distributed on an "AS IS" BASIS, 13 | # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 14 | # See the License for the specific language governing permissions and 15 | # limitations under the License. 16 | 17 | # Cassandra Node IP=Data Center:Rack 18 | 19 | # default for unknown nodes 20 | default=DC1:RAC1 21 | 22 | # Native IPv6 is supported, however you must escape the colon in the IPv6 Address 23 | # Also be sure to comment out JVM_OPTS="$JVM_OPTS -Djava.net.preferIPv4Stack=true" 24 | # in cassandra-env.sh 25 | #fe80\:0\:0\:0\:202\:b3ff\:fe1e\:8329=DC1:RAC3 26 | -------------------------------------------------------------------------------- /gcplot/nginx/nginx: -------------------------------------------------------------------------------- 1 | user www-data; 2 | worker_processes auto; 3 | pid /run/nginx.pid; 4 | 5 | events { 6 | worker_connections 131072; 7 | # multi_accept on; 8 | use epoll; 9 | } 10 | 11 | http { 12 | client_max_body_size 16384M; 13 | ## 14 | # Basic Settings 15 | ## 16 | 17 | sendfile on; 18 | tcp_nopush on; 19 | tcp_nodelay on; 20 | keepalive_timeout 300; 21 | types_hash_max_size 2048; 22 | # server_tokens off; 23 | 24 | # server_names_hash_bucket_size 64; 25 | # server_name_in_redirect off; 26 | 27 | include /etc/nginx/mime.types; 28 | default_type application/octet-stream; 29 | 30 | ## 31 | # Logging Settings 32 | ## 33 | 34 | access_log off; 35 | #access_log /var/log/nginx/access.log; 36 | error_log /var/log/nginx/error.log; 37 | 38 | ## 39 | # Gzip Settings 40 | ## 41 | 42 | gzip on; 43 | gzip_disable "msie6"; 44 | 45 | # gzip_vary on; 46 | # gzip_proxied any; 47 | # gzip_comp_level 6; 48 | # gzip_buffers 16 8k; 49 | # gzip_http_version 1.1; 50 | # gzip_types text/plain text/css application/json application/javascript text/xml application/xml application/xml+rss text/javascript; 51 | 52 | ## 53 | # Virtual Host Configs 54 | ## 55 | 56 | include /etc/nginx/conf.d/*.conf; 57 | include /etc/nginx/sites-enabled/*; 58 | } 59 | -------------------------------------------------------------------------------- /gcplot/cassandra/cassandra-rackdc.properties: -------------------------------------------------------------------------------- 1 | # Licensed to the Apache Software Foundation (ASF) under one 2 | # or more contributor license agreements. See the NOTICE file 3 | # distributed with this work for additional information 4 | # regarding copyright ownership. The ASF licenses this file 5 | # to you under the Apache License, Version 2.0 (the 6 | # "License"); you may not use this file except in compliance 7 | # with the License. You may obtain a copy of the License at 8 | # 9 | # http://www.apache.org/licenses/LICENSE-2.0 10 | # 11 | # Unless required by applicable law or agreed to in writing, software 12 | # distributed under the License is distributed on an "AS IS" BASIS, 13 | # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 14 | # See the License for the specific language governing permissions and 15 | # limitations under the License. 16 | 17 | # These properties are used with GossipingPropertyFileSnitch and will 18 | # indicate the rack and dc for this node 19 | dc=DC1 20 | rack=RAC1 21 | 22 | # Add a suffix to a datacenter name. Used by the Ec2Snitch and Ec2MultiRegionSnitch 23 | # to append a string to the EC2 region name. 24 | #dc_suffix= 25 | 26 | # Uncomment the following line to make this snitch prefer the internal ip when possible, as the Ec2MultiRegionSnitch does. 27 | prefer_local=true 28 | -------------------------------------------------------------------------------- /gcplot/gcserver/gcserver: -------------------------------------------------------------------------------- 1 | #!/bin/bash 2 | SERVICE_NAME=gcserver 3 | PATH_TO_JAR=/home/gcserver/lib/bootstrap.jar 4 | PID_PATH_NAME=/tmp/gcserver.pid 5 | USER=gcserver 6 | ARGS='-server -Xmx{GCPLOT_MEMORY} -Xms{GCPLOT_MEMORY} -XX:+DisableExplicitGC -XX:+PerfDisableSharedMem -XX:+UnlockExperimentalVMOptions -XX:+HeapDumpOnOutOfMemoryError -XX:+UnlockDiagnosticVMOptions' 7 | 8 | function start_service() { 9 | echo "Starting $SERVICE_NAME ..." 10 | stop_service 11 | if [ ! -f $PID_PATH_NAME ]; then 12 | su $USER -c "nohup java $ARGS -Dcurrent.host.name=$HOSTNAME -Dvertx.cacheDirBase=/tmp -Dlogback.configurationFile=/home/gcserver/config/logback.xml -jar $PATH_TO_JAR -cfg /home/gcserver/config >> /dev/null & echo \$! > $PID_PATH_NAME" 13 | sleep 5 14 | echo "$SERVICE_NAME started ..." 15 | else 16 | echo "$SERVICE_NAME is already running ..." 17 | fi 18 | } 19 | 20 | function stop_service() { 21 | if [ -f $PID_PATH_NAME ]; then 22 | PID=$(cat $PID_PATH_NAME); 23 | echo "$SERVICE_NAME stoping ..." 24 | echo "$SERVICE_NAME killing $PID ..." 25 | kill $PID; 26 | wait $PID; 27 | echo "$SERVICE_NAME stopped ..." 28 | rm -rf $PID_PATH_NAME 29 | sleep 5 30 | else 31 | echo "$SERVICE_NAME is not running ..." 32 | fi 33 | } 34 | 35 | case $1 in 36 | start) 37 | start_service 38 | ;; 39 | stop) 40 | stop_service 41 | ;; 42 | restart) 43 | stop_service 44 | start_service 45 | ;; 46 | status) 47 | if [ -f $PID_PATH_NAME ]; then 48 | PID=$(cat $PID_PATH_NAME); 49 | echo "Started $PID." 50 | else 51 | echo "Not started." 52 | fi 53 | ;; 54 | esac 55 | -------------------------------------------------------------------------------- /gcplot/gcserver/logback.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | true 4 | 5 | %d{yyyy/MM/dd HH:mm:ss.SSS} %p [%c{1}] %m%n 6 | 7 | 8 | 9 | /home/gcserver/logs/app.debug.%d{yyyy-MM-dd}.log 10 | 11 | 12 | 31 13 | 30GB 14 | 15 | 16 | 17 | 18 | true 19 | 20 | %d{HH:mm:ss} %p [%c{1}] %m%n 21 | 22 | 23 | 24 | /home/gcserver/logs/analytics.%d{yyyy-MM-dd}.log 25 | 26 | 27 | 31 28 | 20GB 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 43 | 44 | -------------------------------------------------------------------------------- /gcplot/start.sh: -------------------------------------------------------------------------------- 1 | #!/bin/sh 2 | 3 | # Start & Init OrientDB 4 | 5 | sed -i "s/{GCPLOT_MEMORY}/$GCPLOT_MEMORY/g" /etc/init.d/gcserver \ 6 | && sed -i "s/{ORIENTDB_MEMORY}/$ORIENTDB_MEMORY/g" /etc/init.d/orientdb \ 7 | && sed -i "s/{CASSANDRA_MEMORY}/$CASSANDRA_MEMORY/g" /etc/cassandra/jvm.options 8 | 9 | echo "Starting OrientDB ..." 10 | 11 | service orientdb start 12 | 13 | echo "Waiting for the OrientDB 2424 port to be opened ..." 14 | 15 | while ! nc -z localhost 2424; do 16 | sleep 0.1 # wait for 1/10 of the second before check again 17 | done 18 | 19 | echo "Creating OrientDB database ..." 20 | 21 | /opt/orientdb/bin/console.sh "CREATE DATABASE remote:localhost:2424/gcplot admin admin plocal" 22 | 23 | # Start & Init Cassandra 24 | 25 | echo "Starting Cassandra ..." 26 | 27 | service cassandra start 28 | 29 | echo "Waiting for the Cassandra 9042 port to be opened ..." 30 | 31 | while ! nc -z localhost 9042; do 32 | sleep 0.1 # wait for 1/10 of the second before check again 33 | done 34 | 35 | while [ $(grep "Created default superuser role" /var/log/cassandra/debug.log | wc -l) -eq "0" ]; do 36 | sleep 0.1 37 | done 38 | 39 | echo "Performing CQL initialization ..." 40 | 41 | cqlsh --username=cassandra --password=cassandra --file=/etc/cassandra/cdb.cql 42 | 43 | echo "Running gcserver service ..." 44 | 45 | service gcserver start 46 | 47 | # Follow output 48 | 49 | echo "GCPlot instance is initialized" 50 | 51 | while [ $(grep "Trying to reload configuration from main DB" /home/gcserver/logs/app.debug.*.log | wc -l) -eq "0" ]; do 52 | sleep 0.1 53 | done 54 | 55 | echo "Registering admin user in GCPlot ..." 56 | 57 | curl 127.0.0.1:9091/user/register_admin 58 | 59 | echo "Starting nginx service ..." 60 | 61 | service nginx start 62 | 63 | tail -F /home/gcserver/logs/app.debug.*.log 64 | -------------------------------------------------------------------------------- /gcplot/nginx/gcplot.conf: -------------------------------------------------------------------------------- 1 | # Expires map 2 | map $sent_http_content_type $expires { 3 | default off; 4 | text/html epoch; 5 | text/css max; 6 | application/javascript max; 7 | ~image/ max; 8 | } 9 | 10 | server { 11 | listen 80 default_server; 12 | gzip off; 13 | 14 | expires $expires; 15 | charset utf-8; 16 | location / { 17 | root /var/www/landing; 18 | index login.html; 19 | error_page 404 /404.html; 20 | } 21 | location /app { 22 | return 302 $scheme://$http_host$uri/; 23 | } 24 | location /app/ { 25 | root /var/www; 26 | index index.html; 27 | if (!-e $request_filename) { 28 | rewrite ^(.*)$ /app/ break; 29 | } 30 | } 31 | location /assets { 32 | root /var/www/app; 33 | } 34 | location /img { 35 | root /var/www/app; 36 | } 37 | location /rest { 38 | return 302 $scheme://$http_host$uri/; 39 | } 40 | location /rest/ { 41 | proxy_pass http://127.0.0.1:9091/; 42 | proxy_set_header X-Real-IP $remote_addr; 43 | proxy_set_header Host $http_host; 44 | proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for; 45 | add_header 'Access-Control-Allow-Origin' '*'; 46 | add_header 'Access-Control-Allow-Credentials' 'true'; 47 | add_header 'Access-Control-Allow-Methods' 'GET,HEAD,OPTIONS,POST,PUT,DELETE'; 48 | add_header 'Access-Control-Allow-Headers' 'Access-Control-Allow-Headers, Origin,Accept, X-Requested-With, Content-Type, Access-Control-Request-Method, Access-Control-Request-Headers, Transfer-Encoding, X-Firefox-Spdy'; 49 | add_header 'Access-Control-Expose-Headers' 'Transfer-Encoding'; 50 | keepalive_timeout 120s; 51 | keepalive_requests 100; 52 | proxy_connect_timeout 600; 53 | proxy_send_timeout 600; 54 | proxy_read_timeout 600; 55 | send_timeout 600; 56 | proxy_intercept_errors on; 57 | } 58 | } 59 | -------------------------------------------------------------------------------- /gcplot/orientdb/orientdb.sh: -------------------------------------------------------------------------------- 1 | #!/bin/sh 2 | # OrientDB service script 3 | # 4 | # Copyright (c) OrientDB LTD (http://orientdb.com/) 5 | 6 | # chkconfig: 2345 20 80 7 | # description: OrientDb init script 8 | # processname: orientdb.sh 9 | 10 | # You have to SET the OrientDB installation directory here 11 | ORIENTDB_DIR="/opt/orientdb" 12 | ORIENTDB_USER="orientdb" 13 | LOG_DIR="/var/log/orientdb" 14 | export ORIENTDB_OPTS_MEMORY="-Xms{ORIENTDB_MEMORY} -Xmx{ORIENTDB_MEMORY}" 15 | 16 | usage() { 17 | echo "Usage: `basename $0`: " 18 | exit 1 19 | } 20 | 21 | start() { 22 | status 23 | if [ $PID -gt 0 ] 24 | then 25 | echo "OrientDB server daemon was already started. PID: $PID" 26 | return $PID 27 | fi 28 | echo "Starting OrientDB server daemon..." 29 | cd "$ORIENTDB_DIR/bin" 30 | su $ORIENTDB_USER -c "cd \"$ORIENTDB_DIR/bin\"; /usr/bin/nohup ./server.sh 1>$LOG_DIR/orientdb.log 2>$LOG_DIR/orientdb.err &" 31 | } 32 | 33 | stop() { 34 | status 35 | if [ $PID -eq 0 ] 36 | then 37 | echo "OrientDB server daemon is already not running" 38 | return 0 39 | fi 40 | echo "Stopping OrientDB server daemon..." 41 | cd "$ORIENTDB_DIR/bin" 42 | su $ORIENTDB_USER -c "cd \"$ORIENTDB_DIR/bin\"; /usr/bin/nohup ./shutdown.sh 1>>$LOG_DIR/orientdb.log 2>>$LOG_DIR/orientdb.err &" 43 | } 44 | 45 | status() { 46 | PID=` ps auxw | grep 'orientdb.www.path' | grep java | grep -v grep | awk '{print $2}'` 47 | if [ "x$PID" = "x" ] 48 | then 49 | PID=0 50 | fi 51 | 52 | # if PID is greater than 0 then OrientDB is running, else it is not 53 | return $PID 54 | } 55 | 56 | if [ "x$1" = "xstart" ] 57 | then 58 | start 59 | exit 0 60 | fi 61 | 62 | if [ "x$1" = "xstop" ] 63 | then 64 | stop 65 | exit 0 66 | fi 67 | 68 | if [ "x$1" = "xstatus" ] 69 | then 70 | status 71 | if [ $PID -gt 0 ] 72 | then 73 | echo "OrientDB server daemon is running with PID: $PID" 74 | exit 0 75 | else 76 | echo "OrientDB server daemon is NOT running" 77 | exit 3 78 | fi 79 | fi 80 | 81 | usage 82 | -------------------------------------------------------------------------------- /gcplot/orientdb/orientdb-server-log.properties: -------------------------------------------------------------------------------- 1 | # 2 | # /* 3 | # * Copyright 2014 Orient Technologies LTD (info(at)orientechnologies.com) 4 | # * 5 | # * Licensed under the Apache License, Version 2.0 (the "License"); 6 | # * you may not use this file except in compliance with the License. 7 | # * You may obtain a copy of the License at 8 | # * 9 | # * http://www.apache.org/licenses/LICENSE-2.0 10 | # * 11 | # * Unless required by applicable law or agreed to in writing, software 12 | # * distributed under the License is distributed on an "AS IS" BASIS, 13 | # * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. 14 | # * See the License for the specific language governing permissions and 15 | # * limitations under the License. 16 | # * 17 | # * For more information: http://www.orientechnologies.com 18 | # */ 19 | # 20 | 21 | # Specify the handlers to create in the root logger 22 | # (all loggers are children of the root logger) 23 | # The following creates two handlers 24 | handlers = java.util.logging.ConsoleHandler, java.util.logging.FileHandler 25 | 26 | # Set the default logging level for the root logger 27 | .level = INFO 28 | com.orientechnologies.level = INFO 29 | com.orientechnologies.orient.server.distributed.level = INFO 30 | 31 | # Set the default logging level for new ConsoleHandler instances 32 | java.util.logging.ConsoleHandler.level = INFO 33 | # Set the default formatter for new ConsoleHandler instances 34 | java.util.logging.ConsoleHandler.formatter = com.orientechnologies.common.log.OAnsiLogFormatter 35 | 36 | # Set the default logging level for new FileHandler instances 37 | java.util.logging.FileHandler.level = INFO 38 | # Naming style for the output file 39 | java.util.logging.FileHandler.pattern=/var/log/orientdb/orient-server.log 40 | # Set the default formatter for new FileHandler instances 41 | java.util.logging.FileHandler.formatter = com.orientechnologies.common.log.OLogFormatter 42 | # Limiting size of output file in bytes: 43 | java.util.logging.FileHandler.limit=1000000000 44 | # Number of output files to cycle through, by appending an 45 | # integer to the base file name: 46 | java.util.logging.FileHandler.count=10 47 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # GCPlot - All-in-one JVM GC Logs Analyzer 2 | 3 | https://github.com/dmart28/gcplot 4 | 5 | https://github.com/dmart28/gcplot-ui 6 | 7 | GCPlot is a Java Garbage Collector (GC) logs analyzer. Basically, it's an effort to solve all GC logs reading/analyzing problems once and forever. As developers, we were tired about the current situation and efforts needed to just compare some number of GC configurations, so we decided to start from scratch and build a tool that suits best for us. 8 | 9 | The report itself consists of a lot of graphs, measurements, stats, etc about how exactly your GC works. You can also manage the timeline and decide - whether to dig deeper, by analyzing, for example, 2 minutes interval in the most details, or check everything from the bird's eye view by choosing the last month. 10 | 11 | # Installation 12 | 13 | ## Docker Installation 14 | 15 | You can run GCPlot in a Docker container. Docker is supported by most of the modern OS, for more details check official [Docker Installation](https://docs.docker.com/engine/installation/) page. 16 | 17 | In order to run GCPlot as-is without additional configuration, run next command: 18 | 19 | `docker run -d -p 80:80 gcplot/gcplot` 20 | 21 | After that eventually the platform will be accessible from your host machine at `http://127.0.0.1` address. If you would like to use another port, just change it. For example, for `http://127.0.0.1:8080` address, the command will look like: 22 | 23 | `docker run -d -p 8080:80 gcplot/gcplot` 24 | 25 | By default, admin user is already created, with username and password `admin`. Please consider changing it for the best security after the initial log in. 26 | 27 | ### Versions 28 | 29 | You can check the Docker container versions available [here](https://hub.docker.com/r/gcplot/gcplot/tags/). 30 | 31 | ### Memory Settings 32 | 33 | You can control heap size of the services inside container. GCPlot uses Cassandra and OrientDB under the hood, which are also presented inside the container. Default values are: 34 | 35 | ``` 36 | GCPLOT_MEMORY=512m 37 | ORIENTDB_MEMORY=256m 38 | CASSANDRA_MEMORY=1g 39 | ``` 40 | 41 | To give, for example, GCPlot service 1G of heap, and to Cassandra 4G, the command may look like: 42 | 43 | `docker run -d -p 80:80 -e "GCPLOT_MEMORY=1g" -e "CASSANDRA_MEMORY=4g" gcplot/gcplot` 44 | -------------------------------------------------------------------------------- /gcplot/cassandra/cdb.cql: -------------------------------------------------------------------------------- 1 | CREATE KEYSPACE IF NOT EXISTS gcplot 2 | WITH REPLICATION = { 'class' : 'SimpleStrategy', 'replication_factor' : 1 }; 3 | 4 | USE gcplot; 5 | 6 | CREATE TABLE IF NOT EXISTS gc_analyse ( 7 | id uuid, 8 | account_id varchar, 9 | analyse_name varchar, 10 | is_continuous boolean, 11 | start timestamp, 12 | first_event map, 13 | last_event map, 14 | timezone varchar, 15 | jvm_ids set, 16 | jvm_names map, 17 | jvm_versions map, 18 | jvm_gc_types map, 19 | jvm_headers map, 20 | jvm_md_page_size map, 21 | jvm_md_phys_total map, 22 | jvm_md_phys_free map, 23 | jvm_md_swap_total map, 24 | jvm_md_swap_free map, 25 | rc_source_type varchar, 26 | rc_source_config_string varchar, 27 | jvm_rc_source_type map, 28 | jvm_rc_source_config_string map, 29 | configs map, 30 | ext varchar, 31 | PRIMARY KEY (account_id, id) 32 | ) WITH gc_grace_seconds = 60; 33 | 34 | CREATE INDEX IF NOT EXISTS analyse_continuous ON gc_analyse( is_continuous ); 35 | 36 | CREATE TABLE IF NOT EXISTS gc_event ( 37 | id uuid, 38 | parent_id uuid, 39 | analyse_id uuid, 40 | bucket_id varchar, 41 | date varchar, 42 | jvm_id varchar, 43 | description varchar, 44 | written_at timeuuid, 45 | tmstm double, 46 | occurred timestamp, 47 | vm_event_type int, 48 | capacity list, 49 | total_capacity list, 50 | pause_mu bigint, 51 | generations bigint, 52 | phase int, 53 | cause int, 54 | properties bigint, 55 | concurrency int, 56 | gen_cap_before map, 57 | gen_cap_after map, 58 | gen_cap_total map, 59 | user_time double, 60 | sys_time double, 61 | real_time double, 62 | ext varchar, 63 | PRIMARY KEY ((analyse_id, jvm_id, date), written_at) 64 | ) WITH CLUSTERING ORDER BY (written_at DESC) AND gc_grace_seconds = 60; 65 | 66 | CREATE MATERIALIZED VIEW IF NOT EXISTS gc_event_by_bucket 67 | AS SELECT * FROM gc_event 68 | WHERE bucket_id IS NOT NULL AND analyse_id IS NOT NULL AND jvm_id IS NOT NULL AND date IS NOT NULL AND written_at IS NOT NULL 69 | PRIMARY KEY ((bucket_id, analyse_id, jvm_id, date), written_at); 70 | 71 | CREATE TABLE IF NOT EXISTS objects_ages ( 72 | analyse_id uuid, 73 | occurred timestamp, 74 | written_at timeuuid, 75 | jvm_id varchar, 76 | desired_sv_size bigint, 77 | occupied list, 78 | total list, 79 | ext varchar, 80 | PRIMARY KEY ((analyse_id, jvm_id), written_at) 81 | ) WITH CLUSTERING ORDER BY (written_at DESC); 82 | 83 | CREATE INDEX IF NOT EXISTS objects_ages_occurred ON objects_ages( occurred ); 84 | -------------------------------------------------------------------------------- /gcplot/Dockerfile: -------------------------------------------------------------------------------- 1 | # 2 | # GCPlot all-in-one dockerfile 3 | # 4 | 5 | FROM openjdk:8-jdk 6 | 7 | ARG GCPLOT_VERSION=2.0.1 8 | ARG CASSANDRA_VERSION=311x 9 | ARG ORIENTDB_VERSION=2.2.13 10 | 11 | ENV GCPLOT_MEMORY=512m 12 | ENV ORIENTDB_MEMORY=256m 13 | ENV CASSANDRA_MEMORY=1g 14 | 15 | # Install packages and create appropriate user 16 | 17 | RUN \ 18 | apt-get update -y \ 19 | && apt-get install -y wget curl python unzip nginx net-tools netcat \ 20 | && groupadd --system gcserver \ 21 | && useradd -d /home/gcserver -u 1040 -g gcserver -s /bin/bash gcserver \ 22 | && mkdir -p /home/gcserver \ 23 | && mkdir -m 755 -p /home/gcserver/logs \ 24 | && mkdir -m 755 -p /home/gcserver/config \ 25 | && mkdir -m 755 -p /home/gcserver/lib \ 26 | && mkdir -m 777 -p /tmp/gcserver-file-uploads \ 27 | && chown gcserver:gcserver /tmp/gcserver-file-uploads \ 28 | && chown -R gcserver:gcserver /home/gcserver 29 | 30 | COPY gcserver/logback.xml /home/gcserver/config 31 | COPY gcserver/gcplot.properties /home/gcserver/config 32 | COPY gcserver/gcserver /etc/init.d 33 | 34 | RUN chmod 755 /etc/init.d/gcserver \ 35 | && chmod -R 755 /home/gcserver/config/* 36 | 37 | RUN \ 38 | wget "https://downloads.gcplot.com/artifacts/gcserver/${GCPLOT_VERSION}/com.gcplot.web-${GCPLOT_VERSION}-all.jar" -O /home/gcserver/lib/bootstrap.jar \ 39 | && chown -R gcserver:gcserver /home/gcserver 40 | 41 | # Install OrientDB 42 | 43 | RUN \ 44 | groupadd --system orientdb \ 45 | && useradd -d /var/orientdb -u 1041 -g orientdb -s /bin/bash orientdb \ 46 | && mkdir -p /var/orientdb \ 47 | && mkdir -p /var/lib/orientdb \ 48 | && mkdir -p /var/log/orientdb \ 49 | && chown -R orientdb:orientdb /var/orientdb \ 50 | && chown -R orientdb:orientdb /var/lib/orientdb \ 51 | && chown -R orientdb:orientdb /var/log/orientdb \ 52 | && wget "http://orientdb.com/download.php?file=orientdb-community-$ORIENTDB_VERSION.tar.gz&os=multi" -O /tmp/orientdb.tar.gz \ 53 | && tar -zvxf /tmp/orientdb.tar.gz --directory /opt \ 54 | && mv /opt/orientdb-community-$ORIENTDB_VERSION /opt/orientdb \ 55 | && chown -R orientdb:orientdb /opt/orientdb 56 | 57 | COPY orientdb/orientdb-server-config.xml /opt/orientdb/config/orientdb-server-config.xml 58 | COPY orientdb/orientdb-server-log.properties /opt/orientdb/config/orientdb-server-log.properties 59 | COPY orientdb/orientdb.sh /opt/orientdb/bin/orientdb.sh 60 | 61 | RUN cp /opt/orientdb/bin/orientdb.sh /etc/init.d/orientdb \ 62 | && chmod 755 /etc/init.d/orientdb 63 | 64 | # Install Cassandra 65 | 66 | RUN echo "deb http://www.apache.org/dist/cassandra/debian $CASSANDRA_VERSION main" | tee -a /etc/apt/sources.list.d/cassandra.sources.list \ 67 | && curl https://www.apache.org/dist/cassandra/KEYS | apt-key add - \ 68 | && apt-get update -y \ 69 | && apt-get install cassandra -y \ 70 | && sed -i '/ulimit/d' /etc/init.d/cassandra 71 | 72 | COPY cassandra/cassandra-rackdc.properties /etc/cassandra/cassandra-rackdc.properties 73 | COPY cassandra/cassandra-topology.properties /etc/cassandra/cassandra-topology.properties 74 | COPY cassandra/cassandra.yaml /etc/cassandra/cassandra.yaml 75 | COPY cassandra/jvm.properties /etc/cassandra/jvm.options 76 | COPY cassandra/cdb.cql /etc/cassandra/cdb.cql 77 | COPY start.sh /start.sh 78 | 79 | RUN chmod 755 /start.sh 80 | 81 | # Install UI 82 | 83 | RUN wget "https://downloads.gcplot.com/artifacts/ui/$GCPLOT_VERSION/gcplot-ui-$GCPLOT_VERSION.tar.gz" -O /tmp/gcplot-ui.tar.gz \ 84 | && tar -xvzf /tmp/gcplot-ui.tar.gz -C /tmp \ 85 | && sed -i "s/\${CACHE_BUSTER}/$(date +%s)/g" /tmp/dist/index.html \ 86 | && mkdir /var/www/app \ 87 | && mkdir /var/www/landing \ 88 | && mv /tmp/dist/* /var/www/app \ 89 | && mv /tmp/landing/* /var/www/landing \ 90 | && chmod -R 755 /var/www/* 91 | 92 | # Configuring nginx 93 | 94 | COPY nginx/gcplot.conf /etc/nginx/sites-available/gcplot.conf 95 | COPY nginx/nginx /etc/nginx/nginx.conf 96 | RUN ln -s /etc/nginx/sites-available/gcplot.conf /etc/nginx/sites-enabled/gcplot.conf \ 97 | && rm -rf /etc/nginx/sites-enabled/default \ 98 | && rm -rf /etc/nginx/sites-available/default 99 | 100 | VOLUME ["/var/lib/cassandra", "/var/lib/orientdb"] 101 | EXPOSE 9042 2424 2480 80 102 | CMD ["/start.sh"] 103 | -------------------------------------------------------------------------------- /gcplot/orientdb/orientdb-server-config.xml: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | 32 | 33 | 34 | 35 | 36 | 37 | 38 | 39 | 40 | 41 | 42 | 43 | 44 | 45 | 46 | 47 | 48 | 49 | 50 | 51 | 52 | 53 | 54 | 55 | 56 | 57 | 58 | 59 | 60 | 61 | 62 | 63 | 64 | 65 | 66 | 67 | 68 | 69 | 70 | 71 | 72 | 73 | 74 | 75 | 76 | 77 | 78 | 79 | 80 | 81 | 82 | 83 | 84 | 85 | true 86 | 87 | -------------------------------------------------------------------------------- /gcplot/cassandra/jvm.properties: -------------------------------------------------------------------------------- 1 | ########################################################################### 2 | # jvm.options # 3 | # # 4 | # - all flags defined here will be used by cassandra to startup the JVM # 5 | # - one flag should be specified per line # 6 | # - lines that do not start with '-' will be ignored # 7 | # - only static flags are accepted (no variables or parameters) # 8 | # - dynamic flags will be appended to these on cassandra-env # 9 | ########################################################################### 10 | 11 | ###################### 12 | # STARTUP PARAMETERS # 13 | ###################### 14 | 15 | # Uncomment any of the following properties to enable specific startup parameters 16 | 17 | # In a multi-instance deployment, multiple Cassandra instances will independently assume that all 18 | # CPU processors are available to it. This setting allows you to specify a smaller set of processors 19 | # and perhaps have affinity. 20 | #-Dcassandra.available_processors=number_of_processors 21 | 22 | # The directory location of the cassandra.yaml file. 23 | #-Dcassandra.config=directory 24 | 25 | # Sets the initial partitioner token for a node the first time the node is started. 26 | #-Dcassandra.initial_token=token 27 | 28 | # Set to false to start Cassandra on a node but not have the node join the cluster. 29 | #-Dcassandra.join_ring=true|false 30 | 31 | # Set to false to clear all gossip state for the node on restart. Use when you have changed node 32 | # information in cassandra.yaml (such as listen_address). 33 | #-Dcassandra.load_ring_state=true|false 34 | 35 | # Enable pluggable metrics reporter. See Pluggable metrics reporting in Cassandra 2.0.2. 36 | #-Dcassandra.metricsReporterConfigFile=file 37 | 38 | # Set the port on which the CQL native transport listens for clients. (Default: 9042) 39 | #-Dcassandra.native_transport_port=port 40 | 41 | # Overrides the partitioner. (Default: org.apache.cassandra.dht.Murmur3Partitioner) 42 | #-Dcassandra.partitioner=partitioner 43 | 44 | # To replace a node that has died, restart a new node in its place specifying the address of the 45 | # dead node. The new node must not have any data in its data directory, that is, it must be in the 46 | # same state as before bootstrapping. 47 | #-Dcassandra.replace_address=listen_address or broadcast_address of dead node 48 | 49 | # Allow restoring specific tables from an archived commit log. 50 | #-Dcassandra.replayList=table 51 | 52 | # Allows overriding of the default RING_DELAY (1000ms), which is the amount of time a node waits 53 | # before joining the ring. 54 | #-Dcassandra.ring_delay_ms=ms 55 | 56 | # Set the port for the Thrift RPC service, which is used for client connections. (Default: 9160) 57 | #-Dcassandra.rpc_port=port 58 | 59 | # Set the SSL port for encrypted communication. (Default: 7001) 60 | #-Dcassandra.ssl_storage_port=port 61 | 62 | # Enable or disable the native transport server. See start_native_transport in cassandra.yaml. 63 | # cassandra.start_native_transport=true|false 64 | 65 | # Enable or disable the Thrift RPC server. (Default: true) 66 | #-Dcassandra.start_rpc=true/false 67 | 68 | # Set the port for inter-node communication. (Default: 7000) 69 | #-Dcassandra.storage_port=port 70 | 71 | # Set the default location for the trigger JARs. (Default: conf/triggers) 72 | #-Dcassandra.triggers_dir=directory 73 | 74 | # For testing new compaction and compression strategies. It allows you to experiment with different 75 | # strategies and benchmark write performance differences without affecting the production workload. 76 | #-Dcassandra.write_survey=true 77 | 78 | # To disable configuration via JMX of auth caches (such as those for credentials, permissions and 79 | # roles). This will mean those config options can only be set (persistently) in cassandra.yaml 80 | # and will require a restart for new values to take effect. 81 | #-Dcassandra.disable_auth_caches_remote_configuration=true 82 | 83 | ######################## 84 | # GENERAL JVM SETTINGS # 85 | ######################## 86 | 87 | # enable thread priorities, primarily so we can give periodic tasks 88 | # a lower priority to avoid interfering with client workload 89 | -XX:+UseThreadPriorities 90 | 91 | # allows lowering thread priority without being root on linux - probably 92 | # not necessary on Windows but doesn't harm anything. 93 | # see http://tech.stolsvik.com/2010/01/linux-java-thread-priorities-workar 94 | -XX:ThreadPriorityPolicy=42 95 | 96 | # Enable heap-dump if there's an OOM 97 | -XX:+HeapDumpOnOutOfMemoryError 98 | 99 | # Per-thread stack size. 100 | -Xss256k 101 | 102 | # Larger interned string table, for gossip's benefit (CASSANDRA-6410) 103 | -XX:StringTableSize=1000003 104 | 105 | # Make sure all memory is faulted and zeroed on startup. 106 | # This helps prevent soft faults in containers and makes 107 | # transparent hugepage allocation more effective. 108 | -XX:+AlwaysPreTouch 109 | 110 | # Disable biased locking as it does not benefit Cassandra. 111 | -XX:-UseBiasedLocking 112 | 113 | # Enable thread-local allocation blocks and allow the JVM to automatically 114 | # resize them at runtime. 115 | -XX:+UseTLAB 116 | -XX:+ResizeTLAB 117 | 118 | # http://www.evanjones.ca/jvm-mmap-pause.html 119 | -XX:+PerfDisableSharedMem 120 | 121 | # Prefer binding to IPv4 network intefaces (when net.ipv6.bindv6only=1). See 122 | # http://bugs.sun.com/bugdatabase/view_bug.do?bug_id=6342561 (short version: 123 | # comment out this entry to enable IPv6 support). 124 | -Djava.net.preferIPv4Stack=true 125 | 126 | ### Debug options 127 | 128 | # uncomment to enable flight recorder 129 | #-XX:+UnlockCommercialFeatures 130 | #-XX:+FlightRecorder 131 | 132 | # uncomment to have Cassandra JVM listen for remote debuggers/profilers on port 1414 133 | #-agentlib:jdwp=transport=dt_socket,server=y,suspend=n,address=1414 134 | 135 | # uncomment to have Cassandra JVM log internal method compilation (developers only) 136 | #-XX:+UnlockDiagnosticVMOptions 137 | #-XX:+LogCompilation 138 | 139 | ################# 140 | # HEAP SETTINGS # 141 | ################# 142 | 143 | # Heap size is automatically calculated by cassandra-env based on this 144 | # formula: max(min(1/2 ram, 1024MB), min(1/4 ram, 8GB)) 145 | # That is: 146 | # - calculate 1/2 ram and cap to 1024MB 147 | # - calculate 1/4 ram and cap to 8192MB 148 | # - pick the max 149 | # 150 | # For production use you may wish to adjust this for your environment. 151 | # If that's the case, uncomment the -Xmx and Xms options below to override the 152 | # automatic calculation of JVM heap memory. 153 | # 154 | # It is recommended to set min (-Xms) and max (-Xmx) heap sizes to 155 | # the same value to avoid stop-the-world GC pauses during resize, and 156 | # so that we can lock the heap in memory on startup to prevent any 157 | # of it from being swapped out. 158 | -Xms{CASSANDRA_MEMORY} 159 | -Xmx{CASSANDRA_MEMORY} 160 | 161 | # Young generation size is automatically calculated by cassandra-env 162 | # based on this formula: min(100 * num_cores, 1/4 * heap size) 163 | # 164 | # The main trade-off for the young generation is that the larger it 165 | # is, the longer GC pause times will be. The shorter it is, the more 166 | # expensive GC will be (usually). 167 | # 168 | # It is not recommended to set the young generation size if using the 169 | # G1 GC, since that will override the target pause-time goal. 170 | # More info: http://www.oracle.com/technetwork/articles/java/g1gc-1984535.html 171 | # 172 | # The example below assumes a modern 8-core+ machine for decent 173 | # times. If in doubt, and if you do not particularly want to tweak, go 174 | # 100 MB per physical CPU core. 175 | #-Xmn800M 176 | 177 | ################# 178 | # GC SETTINGS # 179 | ################# 180 | 181 | ### CMS Settings 182 | 183 | -XX:+UseParNewGC 184 | -XX:+UseConcMarkSweepGC 185 | -XX:+CMSParallelRemarkEnabled 186 | -XX:SurvivorRatio=8 187 | -XX:MaxTenuringThreshold=1 188 | -XX:CMSInitiatingOccupancyFraction=75 189 | #-XX:+UseCMSInitiatingOccupancyOnly 190 | #-XX:CMSWaitDuration=10000 191 | #-XX:+CMSParallelInitialMarkEnabled 192 | -XX:+CMSEdenChunksRecordAlways 193 | # some JVMs will fill up their heap when accessed via JMX, see CASSANDRA-6541 194 | #-XX:+CMSClassUnloadingEnabled 195 | 196 | 197 | ### G1 Settings (experimental, comment previous section and uncomment section below to enable) 198 | 199 | ## Use the Hotspot garbage-first collector. 200 | # -XX:+UseG1GC 201 | # 202 | ## Have the JVM do less remembered set work during STW, instead 203 | ## preferring concurrent GC. Reduces p99.9 latency. 204 | #-XX:G1RSetUpdatingPauseTimePercent=5 205 | # 206 | ## Main G1GC tunable: lowering the pause target will lower throughput and vise versa. 207 | ## 200ms is the JVM default and lowest viable setting 208 | ## 1000ms increases throughput. Keep it smaller than the timeouts in cassandra.yaml. 209 | # -XX:MaxGCPauseMillis=80 210 | 211 | ## Optional G1 Settings 212 | 213 | # Save CPU time on large (>= 16GB) heaps by delaying region scanning 214 | # until the heap is 70% full. The default in Hotspot 8u40 is 40%. 215 | -XX:InitiatingHeapOccupancyPercent=70 216 | 217 | # For systems with > 8 cores, the default ParallelGCThreads is 5/8 the number of logical cores. 218 | # Otherwise equal to the number of cores when 8 or less. 219 | # Machines with > 10 cores should try setting these to <= full cores. 220 | #-XX:ParallelGCThreads=16 221 | # By default, ConcGCThreads is 1/4 of ParallelGCThreads. 222 | # Setting both to the same value can reduce STW durations. 223 | #-XX:ConcGCThreads=16 224 | 225 | ### GC logging options -- uncomment to enable 226 | 227 | #-XX:+PrintGCDetails 228 | #-XX:+PrintGCDateStamps 229 | #-XX:+PrintHeapAtGC 230 | #-XX:+PrintTenuringDistribution 231 | #-XX:+PrintGCApplicationStoppedTime 232 | #-XX:+PrintPromotionFailure 233 | #-XX:PrintFLSStatistics=1 234 | #-Xloggc:/var/log/cassandra/gc.log 235 | #-XX:+UseGCLogFileRotation 236 | #-XX:NumberOfGCLogFiles=10 237 | #-XX:GCLogFileSize=200M 238 | -------------------------------------------------------------------------------- /gcplot/cassandra/cassandra.yaml: -------------------------------------------------------------------------------- 1 | # Cassandra storage config YAML 2 | 3 | # NOTE: 4 | # See http://wiki.apache.org/cassandra/StorageConfiguration for 5 | # full explanations of configuration directives 6 | # /NOTE 7 | 8 | # The name of the cluster. This is mainly used to prevent machines in 9 | # one logical cluster from joining another. 10 | cluster_name: 'gcp_csd_cluster' 11 | 12 | # This defines the number of tokens randomly assigned to this node on the ring 13 | # The more tokens, relative to other nodes, the larger the proportion of data 14 | # that this node will store. You probably want all nodes to have the same number 15 | # of tokens assuming they have equal hardware capability. 16 | # 17 | # If you leave this unspecified, Cassandra will use the default of 1 token for legacy compatibility, 18 | # and will use the initial_token as described below. 19 | # 20 | # Specifying initial_token will override this setting on the node's initial start, 21 | # on subsequent starts, this setting will apply even if initial token is set. 22 | # 23 | # If you already have a cluster with 1 token per node, and wish to migrate to 24 | # multiple tokens per node, see http://wiki.apache.org/cassandra/Operations 25 | num_tokens: 256 26 | 27 | # Triggers automatic allocation of num_tokens tokens for this node. The allocation 28 | # algorithm attempts to choose tokens in a way that optimizes replicated load over 29 | # the nodes in the datacenter for the replication strategy used by the specified 30 | # keyspace. 31 | # 32 | # The load assigned to each node will be close to proportional to its number of 33 | # vnodes. 34 | # 35 | # Only supported with the Murmur3Partitioner. 36 | # allocate_tokens_for_keyspace: KEYSPACE 37 | 38 | # initial_token allows you to specify tokens manually. While you can use # it with 39 | # vnodes (num_tokens > 1, above) -- in which case you should provide a 40 | # comma-separated list -- it's primarily used when adding nodes # to legacy clusters 41 | # that do not have vnodes enabled. 42 | # initial_token: 43 | 44 | # See http://wiki.apache.org/cassandra/HintedHandoff 45 | # May either be "true" or "false" to enable globally 46 | hinted_handoff_enabled: true 47 | # When hinted_handoff_enabled is true, a black list of data centers that will not 48 | # perform hinted handoff 49 | #hinted_handoff_disabled_datacenters: 50 | # - DC1 51 | # - DC2 52 | # this defines the maximum amount of time a dead host will have hints 53 | # generated. After it has been dead this long, new hints for it will not be 54 | # created until it has been seen alive and gone down again. 55 | max_hint_window_in_ms: 10800000 # 3 hours 56 | 57 | # Maximum throttle in KBs per second, per delivery thread. This will be 58 | # reduced proportionally to the number of nodes in the cluster. (If there 59 | # are two nodes in the cluster, each delivery thread will use the maximum 60 | # rate; if there are three, each will throttle to half of the maximum, 61 | # since we expect two nodes to be delivering hints simultaneously.) 62 | hinted_handoff_throttle_in_kb: 1024 63 | 64 | # Number of threads with which to deliver hints; 65 | # Consider increasing this number when you have multi-dc deployments, since 66 | # cross-dc handoff tends to be slower 67 | max_hints_delivery_threads: 2 68 | 69 | # Directory where Cassandra should store hints. 70 | # If not set, the default directory is $CASSANDRA_HOME/data/hints. 71 | hints_directory: /var/lib/cassandra/hints 72 | 73 | # How often hints should be flushed from the internal buffers to disk. 74 | # Will *not* trigger fsync. 75 | hints_flush_period_in_ms: 10000 76 | 77 | # Maximum size for a single hints file, in megabytes. 78 | max_hints_file_size_in_mb: 512 79 | 80 | # Compression to apply to the hint files. If omitted, hints files 81 | # will be written uncompressed. LZ4, Snappy, and Deflate compressors 82 | # are supported. 83 | #hints_compression: 84 | # - class_name: LZ4Compressor 85 | # parameters: 86 | # - 87 | 88 | # Maximum throttle in KBs per second, total. This will be 89 | # reduced proportionally to the number of nodes in the cluster. 90 | batchlog_replay_throttle_in_kb: 1024 91 | 92 | # Authentication backend, implementing IAuthenticator; used to identify users 93 | # Out of the box, Cassandra provides org.apache.cassandra.auth.{AllowAllAuthenticator, 94 | # PasswordAuthenticator}. 95 | # 96 | # - AllowAllAuthenticator performs no checks - set it to disable authentication. 97 | # - PasswordAuthenticator relies on username/password pairs to authenticate 98 | # users. It keeps usernames and hashed passwords in system_auth.credentials table. 99 | # Please increase system_auth keyspace replication factor if you use this authenticator. 100 | # If using PasswordAuthenticator, CassandraRoleManager must also be used (see below) 101 | authenticator: PasswordAuthenticator 102 | 103 | # Authorization backend, implementing IAuthorizer; used to limit access/provide permissions 104 | # Out of the box, Cassandra provides org.apache.cassandra.auth.{AllowAllAuthorizer, 105 | # CassandraAuthorizer}. 106 | # 107 | # - AllowAllAuthorizer allows any action to any user - set it to disable authorization. 108 | # - CassandraAuthorizer stores permissions in system_auth.permissions table. Please 109 | # increase system_auth keyspace replication factor if you use this authorizer. 110 | authorizer: CassandraAuthorizer 111 | 112 | # Part of the Authentication & Authorization backend, implementing IRoleManager; used 113 | # to maintain grants and memberships between roles. 114 | # Out of the box, Cassandra provides org.apache.cassandra.auth.CassandraRoleManager, 115 | # which stores role information in the system_auth keyspace. Most functions of the 116 | # IRoleManager require an authenticated login, so unless the configured IAuthenticator 117 | # actually implements authentication, most of this functionality will be unavailable. 118 | # 119 | # - CassandraRoleManager stores role data in the system_auth keyspace. Please 120 | # increase system_auth keyspace replication factor if you use this role manager. 121 | role_manager: CassandraRoleManager 122 | 123 | # Validity period for roles cache (fetching permissions can be an 124 | # expensive operation depending on the authorizer). Granted roles are cached for 125 | # authenticated sessions in AuthenticatedUser and after the period specified 126 | # here, become eligible for (async) reload. 127 | # Defaults to 2000, set to 0 to disable. 128 | # Will be disabled automatically for AllowAllAuthenticator. 129 | roles_validity_in_ms: 2000 130 | 131 | # Refresh interval for roles cache (if enabled). 132 | # After this interval, cache entries become eligible for refresh. Upon next 133 | # access, an async reload is scheduled and the old value returned until it 134 | # completes. If roles_validity_in_ms is non-zero, then this must be 135 | # also. 136 | # Defaults to the same value as roles_validity_in_ms. 137 | # roles_update_interval_in_ms: 1000 138 | 139 | # Validity period for permissions cache (fetching permissions can be an 140 | # expensive operation depending on the authorizer, CassandraAuthorizer is 141 | # one example). Defaults to 2000, set to 0 to disable. 142 | # Will be disabled automatically for AllowAllAuthorizer. 143 | permissions_validity_in_ms: 2000 144 | 145 | # Refresh interval for permissions cache (if enabled). 146 | # After this interval, cache entries become eligible for refresh. Upon next 147 | # access, an async reload is scheduled and the old value returned until it 148 | # completes. If permissions_validity_in_ms is non-zero, then this must be 149 | # also. 150 | # Defaults to the same value as permissions_validity_in_ms. 151 | # permissions_update_interval_in_ms: 1000 152 | 153 | # The partitioner is responsible for distributing groups of rows (by 154 | # partition key) across nodes in the cluster. You should leave this 155 | # alone for new clusters. The partitioner can NOT be changed without 156 | # reloading all data, so when upgrading you should set this to the 157 | # same partitioner you were already using. 158 | # 159 | # Besides Murmur3Partitioner, partitioners included for backwards 160 | # compatibility include RandomPartitioner, ByteOrderedPartitioner, and 161 | # OrderPreservingPartitioner. 162 | # 163 | partitioner: org.apache.cassandra.dht.Murmur3Partitioner 164 | 165 | # Directories where Cassandra should store data on disk. Cassandra 166 | # will spread data evenly across them, subject to the granularity of 167 | # the configured compaction strategy. 168 | # If not set, the default directory is $CASSANDRA_HOME/data/data. 169 | data_file_directories: 170 | - /var/lib/cassandra/data 171 | 172 | # commit log. when running on magnetic HDD, this should be a 173 | # separate spindle than the data directories. 174 | # If not set, the default directory is $CASSANDRA_HOME/data/commitlog. 175 | commitlog_directory: /var/lib/cassandra/commitlog 176 | 177 | # policy for data disk failures: 178 | # die: shut down gossip and client transports and kill the JVM for any fs errors or 179 | # single-sstable errors, so the node can be replaced. 180 | # stop_paranoid: shut down gossip and client transports even for single-sstable errors, 181 | # kill the JVM for errors during startup. 182 | # stop: shut down gossip and client transports, leaving the node effectively dead, but 183 | # can still be inspected via JMX, kill the JVM for errors during startup. 184 | # best_effort: stop using the failed disk and respond to requests based on 185 | # remaining available sstables. This means you WILL see obsolete 186 | # data at CL.ONE! 187 | # ignore: ignore fatal errors and let requests fail, as in pre-1.2 Cassandra 188 | disk_failure_policy: stop 189 | 190 | # policy for commit disk failures: 191 | # die: shut down gossip and Thrift and kill the JVM, so the node can be replaced. 192 | # stop: shut down gossip and Thrift, leaving the node effectively dead, but 193 | # can still be inspected via JMX. 194 | # stop_commit: shutdown the commit log, letting writes collect but 195 | # continuing to service reads, as in pre-2.0.5 Cassandra 196 | # ignore: ignore fatal errors and let the batches fail 197 | commit_failure_policy: stop 198 | 199 | # Maximum size of the key cache in memory. 200 | # 201 | # Each key cache hit saves 1 seek and each row cache hit saves 2 seeks at the 202 | # minimum, sometimes more. The key cache is fairly tiny for the amount of 203 | # time it saves, so it's worthwhile to use it at large numbers. 204 | # The row cache saves even more time, but must contain the entire row, 205 | # so it is extremely space-intensive. It's best to only use the 206 | # row cache if you have hot rows or static rows. 207 | # 208 | # NOTE: if you reduce the size, you may not get you hottest keys loaded on startup. 209 | # 210 | # Default value is empty to make it "auto" (min(5% of Heap (in MB), 100MB)). Set to 0 to disable key cache. 211 | key_cache_size_in_mb: 212 | 213 | # Duration in seconds after which Cassandra should 214 | # save the key cache. Caches are saved to saved_caches_directory as 215 | # specified in this configuration file. 216 | # 217 | # Saved caches greatly improve cold-start speeds, and is relatively cheap in 218 | # terms of I/O for the key cache. Row cache saving is much more expensive and 219 | # has limited use. 220 | # 221 | # Default is 14400 or 4 hours. 222 | key_cache_save_period: 14400 223 | 224 | # Number of keys from the key cache to save 225 | # Disabled by default, meaning all keys are going to be saved 226 | # key_cache_keys_to_save: 100 227 | 228 | # Row cache implementation class name. 229 | # Available implementations: 230 | # org.apache.cassandra.cache.OHCProvider Fully off-heap row cache implementation (default). 231 | # org.apache.cassandra.cache.SerializingCacheProvider This is the row cache implementation availabile 232 | # in previous releases of Cassandra. 233 | # row_cache_class_name: org.apache.cassandra.cache.OHCProvider 234 | 235 | # Maximum size of the row cache in memory. 236 | # Please note that OHC cache implementation requires some additional off-heap memory to manage 237 | # the map structures and some in-flight memory during operations before/after cache entries can be 238 | # accounted against the cache capacity. This overhead is usually small compared to the whole capacity. 239 | # Do not specify more memory that the system can afford in the worst usual situation and leave some 240 | # headroom for OS block level cache. Do never allow your system to swap. 241 | # 242 | # Default value is 0, to disable row caching. 243 | row_cache_size_in_mb: 256 244 | 245 | # Duration in seconds after which Cassandra should save the row cache. 246 | # Caches are saved to saved_caches_directory as specified in this configuration file. 247 | # 248 | # Saved caches greatly improve cold-start speeds, and is relatively cheap in 249 | # terms of I/O for the key cache. Row cache saving is much more expensive and 250 | # has limited use. 251 | # 252 | # Default is 0 to disable saving the row cache. 253 | row_cache_save_period: 0 254 | 255 | # Number of keys from the row cache to save. 256 | # Specify 0 (which is the default), meaning all keys are going to be saved 257 | # row_cache_keys_to_save: 100 258 | 259 | # Maximum size of the counter cache in memory. 260 | # 261 | # Counter cache helps to reduce counter locks' contention for hot counter cells. 262 | # In case of RF = 1 a counter cache hit will cause Cassandra to skip the read before 263 | # write entirely. With RF > 1 a counter cache hit will still help to reduce the duration 264 | # of the lock hold, helping with hot counter cell updates, but will not allow skipping 265 | # the read entirely. Only the local (clock, count) tuple of a counter cell is kept 266 | # in memory, not the whole counter, so it's relatively cheap. 267 | # 268 | # NOTE: if you reduce the size, you may not get you hottest keys loaded on startup. 269 | # 270 | # Default value is empty to make it "auto" (min(2.5% of Heap (in MB), 50MB)). Set to 0 to disable counter cache. 271 | # NOTE: if you perform counter deletes and rely on low gcgs, you should disable the counter cache. 272 | counter_cache_size_in_mb: 273 | 274 | # Duration in seconds after which Cassandra should 275 | # save the counter cache (keys only). Caches are saved to saved_caches_directory as 276 | # specified in this configuration file. 277 | # 278 | # Default is 7200 or 2 hours. 279 | counter_cache_save_period: 7200 280 | 281 | # Number of keys from the counter cache to save 282 | # Disabled by default, meaning all keys are going to be saved 283 | # counter_cache_keys_to_save: 100 284 | 285 | # saved caches 286 | # If not set, the default directory is $CASSANDRA_HOME/data/saved_caches. 287 | saved_caches_directory: /var/lib/cassandra/saved_caches 288 | 289 | # commitlog_sync may be either "periodic" or "batch." 290 | # 291 | # When in batch mode, Cassandra won't ack writes until the commit log 292 | # has been fsynced to disk. It will wait 293 | # commitlog_sync_batch_window_in_ms milliseconds between fsyncs. 294 | # This window should be kept short because the writer threads will 295 | # be unable to do extra work while waiting. (You may need to increase 296 | # concurrent_writes for the same reason.) 297 | # 298 | # commitlog_sync: batch 299 | # commitlog_sync_batch_window_in_ms: 2 300 | # 301 | # the other option is "periodic" where writes may be acked immediately 302 | # and the CommitLog is simply synced every commitlog_sync_period_in_ms 303 | # milliseconds. 304 | commitlog_sync: periodic 305 | commitlog_sync_period_in_ms: 10000 306 | 307 | # The size of the individual commitlog file segments. A commitlog 308 | # segment may be archived, deleted, or recycled once all the data 309 | # in it (potentially from each columnfamily in the system) has been 310 | # flushed to sstables. 311 | # 312 | # The default size is 32, which is almost always fine, but if you are 313 | # archiving commitlog segments (see commitlog_archiving.properties), 314 | # then you probably want a finer granularity of archiving; 8 or 16 MB 315 | # is reasonable. 316 | # Max mutation size is also configurable via max_mutation_size_in_kb setting in 317 | # cassandra.yaml. The default is half the size commitlog_segment_size_in_mb * 1024. 318 | # 319 | # NOTE: If max_mutation_size_in_kb is set explicitly then commitlog_segment_size_in_mb must 320 | # be set to at least twice the size of max_mutation_size_in_kb / 1024 321 | # 322 | commitlog_segment_size_in_mb: 32 323 | 324 | # Compression to apply to the commit log. If omitted, the commit log 325 | # will be written uncompressed. LZ4, Snappy, and Deflate compressors 326 | # are supported. 327 | #commitlog_compression: 328 | # - class_name: LZ4Compressor 329 | # parameters: 330 | # - 331 | 332 | # any class that implements the SeedProvider interface and has a 333 | # constructor that takes a Map of parameters will do. 334 | seed_provider: 335 | # Addresses of hosts that are deemed contact points. 336 | # Cassandra nodes use this list of hosts to find each other and learn 337 | # the topology of the ring. You must change this if you are running 338 | # multiple nodes! 339 | - class_name: org.apache.cassandra.locator.SimpleSeedProvider 340 | parameters: 341 | # seeds is actually a comma-delimited list of addresses. 342 | # Ex: ",," 343 | - seeds: 127.0.0.1 344 | 345 | # For workloads with more data than can fit in memory, Cassandra's 346 | # bottleneck will be reads that need to fetch data from 347 | # disk. "concurrent_reads" should be set to (16 * number_of_drives) in 348 | # order to allow the operations to enqueue low enough in the stack 349 | # that the OS and drives can reorder them. Same applies to 350 | # "concurrent_counter_writes", since counter writes read the current 351 | # values before incrementing and writing them back. 352 | # 353 | # On the other hand, since writes are almost never IO bound, the ideal 354 | # number of "concurrent_writes" is dependent on the number of cores in 355 | # your system; (8 * number_of_cores) is a good rule of thumb. 356 | concurrent_reads: 96 357 | concurrent_writes: 96 358 | concurrent_counter_writes: 96 359 | 360 | # For materialized view writes, as there is a read involved, so this should 361 | # be limited by the less of concurrent reads or concurrent writes. 362 | concurrent_materialized_view_writes: 96 363 | 364 | # Maximum memory to use for pooling sstable buffers. Defaults to the smaller 365 | # of 1/4 of heap or 512MB. This pool is allocated off-heap, so is in addition 366 | # to the memory allocated for heap. Memory is only allocated as needed. 367 | # file_cache_size_in_mb: 512 368 | 369 | # Flag indicating whether to allocate on or off heap when the sstable buffer 370 | # pool is exhausted, that is when it has exceeded the maximum memory 371 | # file_cache_size_in_mb, beyond which it will not cache buffers but allocate on request. 372 | 373 | # buffer_pool_use_heap_if_exhausted: true 374 | 375 | # The strategy for optimizing disk read 376 | # Possible values are: 377 | # ssd (for solid state disks, the default) 378 | # spinning (for spinning disks) 379 | # disk_optimization_strategy: ssd 380 | 381 | # Total permitted memory to use for memtables. Cassandra will stop 382 | # accepting writes when the limit is exceeded until a flush completes, 383 | # and will trigger a flush based on memtable_cleanup_threshold 384 | # If omitted, Cassandra will set both to 1/4 the size of the heap. 385 | # memtable_heap_space_in_mb: 2048 386 | # memtable_offheap_space_in_mb: 2048 387 | 388 | # Ratio of occupied non-flushing memtable size to total permitted size 389 | # that will trigger a flush of the largest memtable. Larger mct will 390 | # mean larger flushes and hence less compaction, but also less concurrent 391 | # flush activity which can make it difficult to keep your disks fed 392 | # under heavy write load. 393 | # 394 | # memtable_cleanup_threshold defaults to 1 / (memtable_flush_writers + 1) 395 | # memtable_cleanup_threshold: 0.11 396 | 397 | # Specify the way Cassandra allocates and manages memtable memory. 398 | # Options are: 399 | # heap_buffers: on heap nio buffers 400 | # offheap_buffers: off heap (direct) nio buffers 401 | memtable_allocation_type: heap_buffers 402 | 403 | # Total space to use for commit logs on disk. 404 | # 405 | # If space gets above this value, Cassandra will flush every dirty CF 406 | # in the oldest segment and remove it. So a small total commitlog space 407 | # will tend to cause more flush activity on less-active columnfamilies. 408 | # 409 | # The default value is the smaller of 8192, and 1/4 of the total space 410 | # of the commitlog volume. 411 | # 412 | # commitlog_total_space_in_mb: 8192 413 | 414 | # This sets the amount of memtable flush writer threads. These will 415 | # be blocked by disk io, and each one will hold a memtable in memory 416 | # while blocked. 417 | # 418 | # memtable_flush_writers defaults to the smaller of (number of disks, 419 | # number of cores), with a minimum of 2 and a maximum of 8. 420 | # 421 | # If your data directories are backed by SSD, you should increase this 422 | # to the number of cores. 423 | #memtable_flush_writers: 8 424 | 425 | # A fixed memory pool size in MB for for SSTable index summaries. If left 426 | # empty, this will default to 5% of the heap size. If the memory usage of 427 | # all index summaries exceeds this limit, SSTables with low read rates will 428 | # shrink their index summaries in order to meet this limit. However, this 429 | # is a best-effort process. In extreme conditions Cassandra may need to use 430 | # more than this amount of memory. 431 | index_summary_capacity_in_mb: 432 | 433 | # How frequently index summaries should be resampled. This is done 434 | # periodically to redistribute memory from the fixed-size pool to sstables 435 | # proportional their recent read rates. Setting to -1 will disable this 436 | # process, leaving existing index summaries at their current sampling level. 437 | index_summary_resize_interval_in_minutes: 60 438 | 439 | # Whether to, when doing sequential writing, fsync() at intervals in 440 | # order to force the operating system to flush the dirty 441 | # buffers. Enable this to avoid sudden dirty buffer flushing from 442 | # impacting read latencies. Almost always a good idea on SSDs; not 443 | # necessarily on platters. 444 | trickle_fsync: false 445 | trickle_fsync_interval_in_kb: 10240 446 | 447 | # TCP port, for commands and data 448 | # For security reasons, you should not expose this port to the internet. Firewall it if needed. 449 | storage_port: 7000 450 | 451 | # SSL port, for encrypted communication. Unused unless enabled in 452 | # encryption_options 453 | # For security reasons, you should not expose this port to the internet. Firewall it if needed. 454 | ssl_storage_port: 7001 455 | 456 | # Address or interface to bind to and tell other Cassandra nodes to connect to. 457 | # You _must_ change this if you want multiple nodes to be able to communicate! 458 | # 459 | # Set listen_address OR listen_interface, not both. Interfaces must correspond 460 | # to a single address, IP aliasing is not supported. 461 | # 462 | # Leaving it blank leaves it up to InetAddress.getLocalHost(). This 463 | # will always do the Right Thing _if_ the node is properly configured 464 | # (hostname, name resolution, etc), and the Right Thing is to use the 465 | # address associated with the hostname (it might not be). 466 | # 467 | # Setting listen_address to 0.0.0.0 is always wrong. 468 | # 469 | # If you choose to specify the interface by name and the interface has an ipv4 and an ipv6 address 470 | # you can specify which should be chosen using listen_interface_prefer_ipv6. If false the first ipv4 471 | # address will be used. If true the first ipv6 address will be used. Defaults to false preferring 472 | # ipv4. If there is only one address it will be selected regardless of ipv4/ipv6. 473 | listen_interface: lo 474 | # listen_interface_prefer_ipv6: false 475 | 476 | # Address to broadcast to other Cassandra nodes 477 | # Leaving this blank will set it to the same value as listen_address 478 | broadcast_address: 127.0.0.1 479 | 480 | # When using multiple physical network interfaces, set this 481 | # to true to listen on broadcast_address in addition to 482 | # the listen_address, allowing nodes to communicate in both 483 | # interfaces. 484 | # Ignore this property if the network configuration automatically 485 | # routes between the public and private networks such as EC2. 486 | # listen_on_broadcast_address: false 487 | 488 | # Internode authentication backend, implementing IInternodeAuthenticator; 489 | # used to allow/disallow connections from peer nodes. 490 | # internode_authenticator: org.apache.cassandra.auth.AllowAllInternodeAuthenticator 491 | 492 | # Whether to start the native transport server. 493 | # Please note that the address on which the native transport is bound is the 494 | # same as the rpc_address. The port however is different and specified below. 495 | start_native_transport: true 496 | # port for the CQL native transport to listen for clients on 497 | # For security reasons, you should not expose this port to the internet. Firewall it if needed. 498 | native_transport_port: 9042 499 | # Enabling native transport encryption in client_encryption_options allows you to either use 500 | # encryption for the standard port or to use a dedicated, additional port along with the unencrypted 501 | # standard native_transport_port. 502 | # Enabling client encryption and keeping native_transport_port_ssl disabled will use encryption 503 | # for native_transport_port. Setting native_transport_port_ssl to a different value 504 | # from native_transport_port will use encryption for native_transport_port_ssl while 505 | # keeping native_transport_port unencrypted. 506 | # native_transport_port_ssl: 9142 507 | # The maximum threads for handling requests when the native transport is used. 508 | # This is similar to rpc_max_threads though the default differs slightly (and 509 | # there is no native_transport_min_threads, idle threads will always be stopped 510 | # after 30 seconds). 511 | # native_transport_max_threads: 128 512 | # 513 | # The maximum size of allowed frame. Frame (requests) larger than this will 514 | # be rejected as invalid. The default is 256MB. If you're changing this parameter, 515 | # you may want to adjust max_value_size_in_mb accordingly. 516 | # native_transport_max_frame_size_in_mb: 256 517 | 518 | # The maximum number of concurrent client connections. 519 | # The default is -1, which means unlimited. 520 | # native_transport_max_concurrent_connections: -1 521 | 522 | # The maximum number of concurrent client connections per source ip. 523 | # The default is -1, which means unlimited. 524 | # native_transport_max_concurrent_connections_per_ip: -1 525 | 526 | # Whether to start the thrift rpc server. 527 | start_rpc: false 528 | 529 | # The address or interface to bind the Thrift RPC service and native transport 530 | # server to. 531 | # 532 | # Set rpc_address OR rpc_interface, not both. Interfaces must correspond 533 | # to a single address, IP aliasing is not supported. 534 | # 535 | # Leaving rpc_address blank has the same effect as on listen_address 536 | # (i.e. it will be based on the configured hostname of the node). 537 | # 538 | # Note that unlike listen_address, you can specify 0.0.0.0, but you must also 539 | # set broadcast_rpc_address to a value other than 0.0.0.0. 540 | # 541 | # For security reasons, you should not expose this port to the internet. Firewall it if needed. 542 | # 543 | # If you choose to specify the interface by name and the interface has an ipv4 and an ipv6 address 544 | # you can specify which should be chosen using rpc_interface_prefer_ipv6. If false the first ipv4 545 | # address will be used. If true the first ipv6 address will be used. Defaults to false preferring 546 | # ipv4. If there is only one address it will be selected regardless of ipv4/ipv6. 547 | # rpc_address: 0.0.0.0 548 | rpc_interface: lo 549 | # rpc_interface_prefer_ipv6: false 550 | 551 | # port for Thrift to listen for clients on 552 | rpc_port: 9160 553 | 554 | # RPC address to broadcast to drivers and other Cassandra nodes. This cannot 555 | # be set to 0.0.0.0. If left blank, this will be set to the value of 556 | # rpc_address. If rpc_address is set to 0.0.0.0, broadcast_rpc_address must 557 | # be set. 558 | broadcast_rpc_address: 127.0.0.1 559 | 560 | # enable or disable keepalive on rpc/native connections 561 | rpc_keepalive: true 562 | 563 | # Cassandra provides two out-of-the-box options for the RPC Server: 564 | # 565 | # sync -> One thread per thrift connection. For a very large number of clients, memory 566 | # will be your limiting factor. On a 64 bit JVM, 180KB is the minimum stack size 567 | # per thread, and that will correspond to your use of virtual memory (but physical memory 568 | # may be limited depending on use of stack space). 569 | # 570 | # hsha -> Stands for "half synchronous, half asynchronous." All thrift clients are handled 571 | # asynchronously using a small number of threads that does not vary with the amount 572 | # of thrift clients (and thus scales well to many clients). The rpc requests are still 573 | # synchronous (one thread per active request). If hsha is selected then it is essential 574 | # that rpc_max_threads is changed from the default value of unlimited. 575 | # 576 | # The default is sync because on Windows hsha is about 30% slower. On Linux, 577 | # sync/hsha performance is about the same, with hsha of course using less memory. 578 | # 579 | # Alternatively, can provide your own RPC server by providing the fully-qualified class name 580 | # of an o.a.c.t.TServerFactory that can create an instance of it. 581 | rpc_server_type: sync 582 | 583 | # Uncomment rpc_min|max_thread to set request pool size limits. 584 | # 585 | # Regardless of your choice of RPC server (see above), the number of maximum requests in the 586 | # RPC thread pool dictates how many concurrent requests are possible (but if you are using the sync 587 | # RPC server, it also dictates the number of clients that can be connected at all). 588 | # 589 | # The default is unlimited and thus provides no protection against clients overwhelming the server. You are 590 | # encouraged to set a maximum that makes sense for you in production, but do keep in mind that 591 | # rpc_max_threads represents the maximum number of client requests this server may execute concurrently. 592 | # 593 | # rpc_min_threads: 16 594 | # rpc_max_threads: 2048 595 | 596 | # uncomment to set socket buffer sizes on rpc connections 597 | # rpc_send_buff_size_in_bytes: 598 | # rpc_recv_buff_size_in_bytes: 599 | 600 | # Uncomment to set socket buffer size for internode communication 601 | # Note that when setting this, the buffer size is limited by net.core.wmem_max 602 | # and when not setting it it is defined by net.ipv4.tcp_wmem 603 | # See: 604 | # /proc/sys/net/core/wmem_max 605 | # /proc/sys/net/core/rmem_max 606 | # /proc/sys/net/ipv4/tcp_wmem 607 | # /proc/sys/net/ipv4/tcp_wmem 608 | # and: man tcp 609 | # internode_send_buff_size_in_bytes: 610 | # internode_recv_buff_size_in_bytes: 611 | 612 | # Frame size for thrift (maximum message length). 613 | thrift_framed_transport_size_in_mb: 15 614 | 615 | # Set to true to have Cassandra create a hard link to each sstable 616 | # flushed or streamed locally in a backups/ subdirectory of the 617 | # keyspace data. Removing these links is the operator's 618 | # responsibility. 619 | incremental_backups: false 620 | 621 | # Whether or not to take a snapshot before each compaction. Be 622 | # careful using this option, since Cassandra won't clean up the 623 | # snapshots for you. Mostly useful if you're paranoid when there 624 | # is a data format change. 625 | snapshot_before_compaction: false 626 | 627 | # Whether or not a snapshot is taken of the data before keyspace truncation 628 | # or dropping of column families. The STRONGLY advised default of true 629 | # should be used to provide data safety. If you set this flag to false, you will 630 | # lose data on truncation or drop. 631 | auto_snapshot: true 632 | 633 | # When executing a scan, within or across a partition, we need to keep the 634 | # tombstones seen in memory so we can return them to the coordinator, which 635 | # will use them to make sure other replicas also know about the deleted rows. 636 | # With workloads that generate a lot of tombstones, this can cause performance 637 | # problems and even exaust the server heap. 638 | # (http://www.datastax.com/dev/blog/cassandra-anti-patterns-queues-and-queue-like-datasets) 639 | # Adjust the thresholds here if you understand the dangers and want to 640 | # scan more tombstones anyway. These thresholds may also be adjusted at runtime 641 | # using the StorageService mbean. 642 | tombstone_warn_threshold: 1000 643 | tombstone_failure_threshold: 100000 644 | 645 | # Granularity of the collation index of rows within a partition. 646 | # Increase if your rows are large, or if you have a very large 647 | # number of rows per partition. The competing goals are these: 648 | # 1) a smaller granularity means more index entries are generated 649 | # and looking up rows withing the partition by collation column 650 | # is faster 651 | # 2) but, Cassandra will keep the collation index in memory for hot 652 | # rows (as part of the key cache), so a larger granularity means 653 | # you can cache more hot rows 654 | column_index_size_in_kb: 64 655 | 656 | 657 | # Log WARN on any batch size exceeding this value. 5kb per batch by default. 658 | # Caution should be taken on increasing the size of this threshold as it can lead to node instability. 659 | batch_size_warn_threshold_in_kb: 5 660 | 661 | # Fail any batch exceeding this value. 50kb (10x warn threshold) by default. 662 | batch_size_fail_threshold_in_kb: 50 663 | 664 | # Log WARN on any batches not of type LOGGED than span across more partitions than this limit 665 | unlogged_batch_across_partitions_warn_threshold: 10 666 | 667 | # Number of simultaneous compactions to allow, NOT including 668 | # validation "compactions" for anti-entropy repair. Simultaneous 669 | # compactions can help preserve read performance in a mixed read/write 670 | # workload, by mitigating the tendency of small sstables to accumulate 671 | # during a single long running compactions. The default is usually 672 | # fine and if you experience problems with compaction running too 673 | # slowly or too fast, you should look at 674 | # compaction_throughput_mb_per_sec first. 675 | # 676 | # concurrent_compactors defaults to the smaller of (number of disks, 677 | # number of cores), with a minimum of 2 and a maximum of 8. 678 | # 679 | # If your data directories are backed by SSD, you should increase this 680 | # to the number of cores. 681 | #concurrent_compactors: 1 682 | 683 | # Throttles compaction to the given total throughput across the entire 684 | # system. The faster you insert data, the faster you need to compact in 685 | # order to keep the sstable count down, but in general, setting this to 686 | # 16 to 32 times the rate you are inserting data is more than sufficient. 687 | # Setting this to 0 disables throttling. Note that this account for all types 688 | # of compaction, including validation compaction. 689 | compaction_throughput_mb_per_sec: 16 690 | 691 | # Log a warning when compacting partitions larger than this value 692 | compaction_large_partition_warning_threshold_mb: 100 693 | 694 | # When compacting, the replacement sstable(s) can be opened before they 695 | # are completely written, and used in place of the prior sstables for 696 | # any range that has been written. This helps to smoothly transfer reads 697 | # between the sstables, reducing page cache churn and keeping hot rows hot 698 | sstable_preemptive_open_interval_in_mb: 50 699 | 700 | # Throttles all outbound streaming file transfers on this node to the 701 | # given total throughput in Mbps. This is necessary because Cassandra does 702 | # mostly sequential IO when streaming data during bootstrap or repair, which 703 | # can lead to saturating the network connection and degrading rpc performance. 704 | # When unset, the default is 200 Mbps or 25 MB/s. 705 | # stream_throughput_outbound_megabits_per_sec: 200 706 | 707 | # Throttles all streaming file transfer between the datacenters, 708 | # this setting allows users to throttle inter dc stream throughput in addition 709 | # to throttling all network stream traffic as configured with 710 | # stream_throughput_outbound_megabits_per_sec 711 | # When unset, the default is 200 Mbps or 25 MB/s 712 | # inter_dc_stream_throughput_outbound_megabits_per_sec: 200 713 | 714 | # How long the coordinator should wait for read operations to complete 715 | read_request_timeout_in_ms: 5000 716 | # How long the coordinator should wait for seq or index scans to complete 717 | range_request_timeout_in_ms: 10000 718 | # How long the coordinator should wait for writes to complete 719 | write_request_timeout_in_ms: 2000 720 | # How long the coordinator should wait for counter writes to complete 721 | counter_write_request_timeout_in_ms: 5000 722 | # How long a coordinator should continue to retry a CAS operation 723 | # that contends with other proposals for the same row 724 | cas_contention_timeout_in_ms: 1000 725 | # How long the coordinator should wait for truncates to complete 726 | # (This can be much longer, because unless auto_snapshot is disabled 727 | # we need to flush first so we can snapshot before removing the data.) 728 | truncate_request_timeout_in_ms: 60000 729 | # The default timeout for other, miscellaneous operations 730 | request_timeout_in_ms: 10000 731 | 732 | # Enable operation timeout information exchange between nodes to accurately 733 | # measure request timeouts. If disabled, replicas will assume that requests 734 | # were forwarded to them instantly by the coordinator, which means that 735 | # under overload conditions we will waste that much extra time processing 736 | # already-timed-out requests. 737 | # 738 | # Warning: before enabling this property make sure to ntp is installed 739 | # and the times are synchronized between the nodes. 740 | cross_node_timeout: false 741 | 742 | # Set socket timeout for streaming operation. 743 | # The stream session is failed if no data/ack is received by any of the participants 744 | # within that period, which means this should also be sufficient to stream a large 745 | # sstable or rebuild table indexes. 746 | # Default value is 86400000ms, which means stale streams timeout after 24 hours. 747 | # A value of zero means stream sockets should never time out. 748 | # streaming_socket_timeout_in_ms: 86400000 749 | 750 | # phi value that must be reached for a host to be marked down. 751 | # most users should never need to adjust this. 752 | phi_convict_threshold: 12 753 | 754 | # endpoint_snitch -- Set this to a class that implements 755 | # IEndpointSnitch. The snitch has two functions: 756 | # - it teaches Cassandra enough about your network topology to route 757 | # requests efficiently 758 | # - it allows Cassandra to spread replicas around your cluster to avoid 759 | # correlated failures. It does this by grouping machines into 760 | # "datacenters" and "racks." Cassandra will do its best not to have 761 | # more than one replica on the same "rack" (which may not actually 762 | # be a physical location) 763 | # 764 | # IF YOU CHANGE THE SNITCH AFTER DATA IS INSERTED INTO THE CLUSTER, 765 | # YOU MUST RUN A FULL REPAIR, SINCE THE SNITCH AFFECTS WHERE REPLICAS 766 | # ARE PLACED. 767 | # 768 | # IF THE RACK A REPLICA IS PLACED IN CHANGES AFTER THE REPLICA HAS BEEN 769 | # ADDED TO A RING, THE NODE MUST BE DECOMMISSIONED AND REBOOTSTRAPPED. 770 | # 771 | # Out of the box, Cassandra provides 772 | # - SimpleSnitch: 773 | # Treats Strategy order as proximity. This can improve cache 774 | # locality when disabling read repair. Only appropriate for 775 | # single-datacenter deployments. 776 | # - GossipingPropertyFileSnitch 777 | # This should be your go-to snitch for production use. The rack 778 | # and datacenter for the local node are defined in 779 | # cassandra-rackdc.properties and propagated to other nodes via 780 | # gossip. If cassandra-topology.properties exists, it is used as a 781 | # fallback, allowing migration from the PropertyFileSnitch. 782 | # - PropertyFileSnitch: 783 | # Proximity is determined by rack and data center, which are 784 | # explicitly configured in cassandra-topology.properties. 785 | # - Ec2Snitch: 786 | # Appropriate for EC2 deployments in a single Region. Loads Region 787 | # and Availability Zone information from the EC2 API. The Region is 788 | # treated as the datacenter, and the Availability Zone as the rack. 789 | # Only private IPs are used, so this will not work across multiple 790 | # Regions. 791 | # - Ec2MultiRegionSnitch: 792 | # Uses public IPs as broadcast_address to allow cross-region 793 | # connectivity. (Thus, you should set seed addresses to the public 794 | # IP as well.) You will need to open the storage_port or 795 | # ssl_storage_port on the public IP firewall. (For intra-Region 796 | # traffic, Cassandra will switch to the private IP after 797 | # establishing a connection.) 798 | # - RackInferringSnitch: 799 | # Proximity is determined by rack and data center, which are 800 | # assumed to correspond to the 3rd and 2nd octet of each node's IP 801 | # address, respectively. Unless this happens to match your 802 | # deployment conventions, this is best used as an example of 803 | # writing a custom Snitch class and is provided in that spirit. 804 | # 805 | # You can use a custom Snitch by setting this to the full class name 806 | # of the snitch, which will be assumed to be on your classpath. 807 | endpoint_snitch: GossipingPropertyFileSnitch 808 | 809 | # controls how often to perform the more expensive part of host score 810 | # calculation 811 | dynamic_snitch_update_interval_in_ms: 100 812 | # controls how often to reset all host scores, allowing a bad host to 813 | # possibly recover 814 | dynamic_snitch_reset_interval_in_ms: 600000 815 | # if set greater than zero and read_repair_chance is < 1.0, this will allow 816 | # 'pinning' of replicas to hosts in order to increase cache capacity. 817 | # The badness threshold will control how much worse the pinned host has to be 818 | # before the dynamic snitch will prefer other replicas over it. This is 819 | # expressed as a double which represents a percentage. Thus, a value of 820 | # 0.2 means Cassandra would continue to prefer the static snitch values 821 | # until the pinned host was 20% worse than the fastest. 822 | dynamic_snitch_badness_threshold: 0.1 823 | 824 | # request_scheduler -- Set this to a class that implements 825 | # RequestScheduler, which will schedule incoming client requests 826 | # according to the specific policy. This is useful for multi-tenancy 827 | # with a single Cassandra cluster. 828 | # NOTE: This is specifically for requests from the client and does 829 | # not affect inter node communication. 830 | # org.apache.cassandra.scheduler.NoScheduler - No scheduling takes place 831 | # org.apache.cassandra.scheduler.RoundRobinScheduler - Round robin of 832 | # client requests to a node with a separate queue for each 833 | # request_scheduler_id. The scheduler is further customized by 834 | # request_scheduler_options as described below. 835 | request_scheduler: org.apache.cassandra.scheduler.NoScheduler 836 | 837 | # Scheduler Options vary based on the type of scheduler 838 | # NoScheduler - Has no options 839 | # RoundRobin 840 | # - throttle_limit -- The throttle_limit is the number of in-flight 841 | # requests per client. Requests beyond 842 | # that limit are queued up until 843 | # running requests can complete. 844 | # The value of 80 here is twice the number of 845 | # concurrent_reads + concurrent_writes. 846 | # - default_weight -- default_weight is optional and allows for 847 | # overriding the default which is 1. 848 | # - weights -- Weights are optional and will default to 1 or the 849 | # overridden default_weight. The weight translates into how 850 | # many requests are handled during each turn of the 851 | # RoundRobin, based on the scheduler id. 852 | # 853 | # request_scheduler_options: 854 | # throttle_limit: 80 855 | # default_weight: 5 856 | # weights: 857 | # Keyspace1: 1 858 | # Keyspace2: 5 859 | 860 | # request_scheduler_id -- An identifier based on which to perform 861 | # the request scheduling. Currently the only valid option is keyspace. 862 | # request_scheduler_id: keyspace 863 | 864 | # Enable or disable inter-node encryption 865 | # Default settings are TLS v1, RSA 1024-bit keys (it is imperative that 866 | # users generate their own keys) TLS_RSA_WITH_AES_128_CBC_SHA as the cipher 867 | # suite for authentication, key exchange and encryption of the actual data transfers. 868 | # Use the DHE/ECDHE ciphers if running in FIPS 140 compliant mode. 869 | # NOTE: No custom encryption options are enabled at the moment 870 | # The available internode options are : all, none, dc, rack 871 | # 872 | # If set to dc cassandra will encrypt the traffic between the DCs 873 | # If set to rack cassandra will encrypt the traffic between the racks 874 | # 875 | # The passwords used in these options must match the passwords used when generating 876 | # the keystore and truststore. For instructions on generating these files, see: 877 | # http://download.oracle.com/javase/6/docs/technotes/guides/security/jsse/JSSERefGuide.html#CreateKeystore 878 | # 879 | server_encryption_options: 880 | internode_encryption: none 881 | keystore: conf/.keystore 882 | keystore_password: cassandra 883 | truststore: conf/.truststore 884 | truststore_password: cassandra 885 | # More advanced defaults below: 886 | # protocol: TLS 887 | # algorithm: SunX509 888 | # store_type: JKS 889 | # cipher_suites: [TLS_RSA_WITH_AES_128_CBC_SHA,TLS_RSA_WITH_AES_256_CBC_SHA,TLS_DHE_RSA_WITH_AES_128_CBC_SHA,TLS_DHE_RSA_WITH_AES_256_CBC_SHA,TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA,TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA] 890 | # require_client_auth: false 891 | 892 | # enable or disable client/server encryption. 893 | client_encryption_options: 894 | enabled: false 895 | # If enabled and optional is set to true encrypted and unencrypted connections are handled. 896 | optional: false 897 | keystore: conf/.keystore 898 | keystore_password: cassandra 899 | # require_client_auth: false 900 | # Set trustore and truststore_password if require_client_auth is true 901 | # truststore: conf/.truststore 902 | # truststore_password: cassandra 903 | # More advanced defaults below: 904 | # protocol: TLS 905 | # algorithm: SunX509 906 | # store_type: JKS 907 | # cipher_suites: [TLS_RSA_WITH_AES_128_CBC_SHA,TLS_RSA_WITH_AES_256_CBC_SHA,TLS_DHE_RSA_WITH_AES_128_CBC_SHA,TLS_DHE_RSA_WITH_AES_256_CBC_SHA,TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA,TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA] 908 | 909 | # internode_compression controls whether traffic between nodes is 910 | # compressed. 911 | # can be: all - all traffic is compressed 912 | # dc - traffic between different datacenters is compressed 913 | # none - nothing is compressed. 914 | internode_compression: all 915 | 916 | # Enable or disable tcp_nodelay for inter-dc communication. 917 | # Disabling it will result in larger (but fewer) network packets being sent, 918 | # reducing overhead from the TCP protocol itself, at the cost of increasing 919 | # latency if you block for cross-datacenter responses. 920 | inter_dc_tcp_nodelay: false 921 | 922 | # TTL for different trace types used during logging of the repair process. 923 | tracetype_query_ttl: 86400 924 | tracetype_repair_ttl: 604800 925 | 926 | # GC Pauses greater than gc_warn_threshold_in_ms will be logged at WARN level 927 | # Adjust the threshold based on your application throughput requirement 928 | # By default, Cassandra logs GC Pauses greater than 200 ms at INFO level 929 | gc_warn_threshold_in_ms: 1000 930 | 931 | # UDFs (user defined functions) are disabled by default. 932 | # As of Cassandra 3.0 there is a sandbox in place that should prevent execution of evil code. 933 | enable_user_defined_functions: false 934 | 935 | # Enables scripted UDFs (JavaScript UDFs). 936 | # Java UDFs are always enabled, if enable_user_defined_functions is true. 937 | # Enable this option to be able to use UDFs with "language javascript" or any custom JSR-223 provider. 938 | # This option has no effect, if enable_user_defined_functions is false. 939 | enable_scripted_user_defined_functions: false 940 | 941 | # The default Windows kernel timer and scheduling resolution is 15.6ms for power conservation. 942 | # Lowering this value on Windows can provide much tighter latency and better throughput, however 943 | # some virtualized environments may see a negative performance impact from changing this setting 944 | # below their system default. The sysinternals 'clockres' tool can confirm your system's default 945 | # setting. 946 | windows_timer_interval: 1 947 | 948 | # Maximum size of any value in SSTables. Safety measure to detect SSTable corruption 949 | # early. Any value size larger than this threshold will result into marking an SSTable 950 | # as corrupted. 951 | # max_value_size_in_mb: 256 952 | 953 | ############################################################# 954 | ### Lekane changes #### 955 | ############################################################# 956 | #auto_bootstrap: false 957 | --------------------------------------------------------------------------------