logicalclocks · gibchikafa · Sep 19, 2023 · Sep 19, 2023 · Sep 21, 2023 · Sep 25, 2023
diff --git a/Berksfile b/Berksfile
@@ -3,29 +3,28 @@ source 'https://supermarket.chef.io'
 
 metadata
 
-cookbook 'conda', github: "logicalclocks/conda-chef", branch: "master"
-cookbook 'kagent', github: "logicalclocks/kagent-chef", branch: "master"
-cookbook 'hops', github: "logicalclocks/hops-hadoop-chef", branch: "master"
-cookbook 'ndb', github: "logicalclocks/ndb-chef", branch: "master"
-cookbook 'hadoop_spark', github: "logicalclocks/spark-chef", branch: "master"
-cookbook 'flink', github: "logicalclocks/flink-chef", branch: "master"
-cookbook 'livy', github: "logicalclocks/livy-chef", branch: "master"
-cookbook 'epipe', github: "logicalclocks/epipe-chef", branch: "master"
-cookbook 'tensorflow', github: "logicalclocks/tensorflow-chef", branch: "master"
-cookbook 'kzookeeper', github: "logicalclocks/kzookeeper", branch: "master"
-cookbook 'kkafka', github: "logicalclocks/kafka-cookbook", branch: "master"
-cookbook 'elastic', github: "logicalclocks/elasticsearch-chef", branch: "master"
-cookbook 'hopslog', github: "logicalclocks/hopslog-chef", branch: "master"
-cookbook 'hopsmonitor', github: "logicalclocks/hopsmonitor-chef", branch: "master"
-cookbook 'hops_airflow', github: "logicalclocks/airflow-chef", branch: "master"
-cookbook 'hive2', github: "logicalclocks/hive-chef", branch: "master"
-cookbook 'consul', github: "logicalclocks/consul-chef", branch: "master"
-cookbook 'kube-hops', github: "logicalclocks/kube-hops-chef", branch: "master"
-cookbook 'cloud', github: "logicalclocks/cloud-chef", branch: "master"
-cookbook 'onlinefs', github: "logicalclocks/onlinefs-chef", branch: "master"
-cookbook 'flyingduck', github: "logicalclocks/flyingduck-chef", branch: "master"
+cookbook 'conda', github: "logicalclocks/conda-chef", branch: "3.4"
+cookbook 'kagent', github: "logicalclocks/kagent-chef", branch: "3.4"
+cookbook 'hops', github: "logicalclocks/hops-hadoop-chef", branch: "3.4"
+cookbook 'ndb', github: "logicalclocks/ndb-chef", branch: "3.4"
+cookbook 'hadoop_spark', github: "logicalclocks/spark-chef", branch: "3.4"
+cookbook 'flink', github: "logicalclocks/flink-chef", branch: "3.4"
+cookbook 'livy', github: "logicalclocks/livy-chef", branch: "3.4"
+cookbook 'epipe', github: "logicalclocks/epipe-chef", branch: "3.4"
+cookbook 'tensorflow', github: "logicalclocks/tensorflow-chef", branch: "3.4"
+cookbook 'kzookeeper', github: "logicalclocks/kzookeeper", branch: "3.4"
+cookbook 'kkafka', github: "logicalclocks/kafka-cookbook", branch: "3.4"
+cookbook 'elastic', github: "logicalclocks/elasticsearch-chef", branch: "3.4"
+cookbook 'hopslog', github: "logicalclocks/hopslog-chef", branch: "3.4"
+cookbook 'hopsmonitor', github: "logicalclocks/hopsmonitor-chef", branch: "3.4"
+cookbook 'hops_airflow', github: "logicalclocks/airflow-chef", branch: "3.4"
+cookbook 'hive2', github: "logicalclocks/hive-chef", branch: "3.4"
+cookbook 'consul', github: "logicalclocks/consul-chef", branch: "3.4"
+cookbook 'kube-hops', github: "logicalclocks/kube-hops-chef", branch: "3.4"
+cookbook 'cloud', github: "logicalclocks/cloud-chef", branch: "3.4"
+cookbook 'onlinefs', github: "logicalclocks/onlinefs-chef", branch: "3.4"
+cookbook 'flyingduck', github: "logicalclocks/flyingduck-chef", branch: "3.4"
 
-cookbook 'ulimit', github: "logicalclocks/chef-ulimit", branch: "master"
-cookbook 'glassfish', github: "logicalclocks/chef-glassfish", branch: "master"
+cookbook 'glassfish', github: "logicalclocks/chef-glassfish", branch: "3.4"
 cookbook 'java', github: "logicalclocks/java", branch: "v7.0.0-1"
 cookbook 'packagecloud', '= 1.0.1'
diff --git a/Karamelfile b/Karamelfile
@@ -38,4 +38,7 @@ dependencies:
   - recipe: hopsworks::loadbalancer
     global:
       - hopsworks::config_node
-
+  - recipe: hopsworks::reindex
+    global:
+      - hopsworks::default
+      - elastic::default
diff --git a/attributes/default.rb b/attributes/default.rb
@@ -521,4 +521,34 @@
 default['judge']['port']      = "1111"
 default['judge']['home']      = "#{node['install']['dir']}/judge"
 default['judge']['etc']       = "#{node['judge']['home']}/etc"
-default['judge']['logs']      = "#{node['judge']['home']}/logs"
+default['judge']['logs']      = "#{node['judge']['home']}/logs"
+
+
+default['hopsworks']['rstudio_dir'] = node['hopsworks']['dir'] + "/rstudio"
+#rstudio configuration variables
+default["hopsworks"]['rstudio_host']                 = "localhost"
+default["hopsworks"]['rstudio_origin_scheme']        = "https"
+default["hopsworks"]["rstudio_www_address"]          = "0.0.0.0"
+default["hopsworks"]["rstudio_session_timeout_minutes"] = 360
+default["hopsworks"]["rstudio_logging_level"]        = "info"
+default["hopsworks"]["rstudio_logger_type"]          = "file"
+default["hopsworks"]["rstudio_log_file_max_size"]    = 512
+default["hopsworks"]["rstudio_default_cran_repo"]    = "https://cloud.r-project.org/"
+
+default['rstudio']['base_dir']                       = node['install']['dir'].empty? ? node['hopsworks']['dir'] + "/rstudio" : node['install']['dir'] + "/rstudio"
+default['rstudio']['shutdown_timer_interval']          = "30m"
+
+# CRAN
+default['rstudio']['cran']['mirror'] = 'http://cran.rstudio.com/'
+
+# APT configuration for Ubuntu or Debian installs.
+case node["platform"].downcase
+when "ubuntu"
+  default['rstudio']['apt']['key'] = 'E084DAB9'
+  default['rstudio']['apt']['keyserver'] = 'keyserver.ubuntu.com'
+  default['rstudio']['apt']['uri'] = 'http://cran.stat.ucla.edu/bin/linux/ubuntu'
+when "debian"
+  default['rstudio']['apt']['key'] = '381BA480'
+  default['rstudio']['apt']['keyserver'] = 'subkeys.pgp.net'
+  default['rstudio']['apt']['uri'] = 'http://cran.stat.ucla.edu/bin/linux/debian'
+end
diff --git a/files/default/hopsworks_templates/config_template.json b/files/default/hopsworks_templates/config_template.json
@@ -50,6 +50,7 @@
     "executorCores": ${conf.executorCores?c},
     "executorMemory": "${conf.executorMemory}",
     "proxyUser": "${conf.hdfsUser}",
+    "name": "${conf.livySessionName}",
     "queue": "${conf.yarnQueue}",
     "conf": {
 ${conf.sparkConfiguration}

diff --git a/files/default/hopsworks_templates/rstudio_dockerfile_template b/files/default/hopsworks_templates/rstudio_dockerfile_template
@@ -0,0 +1,6 @@
+FROM ${conf.baseImage}
+
+COPY ${conf.buildScript} ./
+RUN chmod +x ${conf.buildScript}
+RUN ./${conf.buildScript}
+RUN rm ${conf.buildScript}
diff --git a/files/default/hopsworks_templates/rstudio_logging_config_template b/files/default/hopsworks_templates/rstudio_logging_config_template
@@ -0,0 +1,9 @@
+[*]
+log-level=warn
+logger-type=syslog
+
+[@rserver]
+log-level=${conf.logLevel}
+logger-type=${conf.loggerType}
+max-size-mb=${conf.maxSizeMb}
+log-file-include-pid=${conf.includePid}
diff --git a/files/default/hopsworks_templates/rstudio_rserver_config_template b/files/default/hopsworks_templates/rstudio_rserver_config_template
@@ -0,0 +1,5 @@
+rsession-which-r=${conf.versionPath}
+www-address=${conf.ipAddress}
+www-port=${conf.port}
+www-root-path=${conf.rootPath}
+server-user=${conf.serverUser}
diff --git a/files/default/hopsworks_templates/rstudio_rsession_config_template b/files/default/hopsworks_templates/rstudio_rsession_config_template
@@ -0,0 +1,2 @@
+session-timeout-minutes=${conf.sessionTimeoutMinutes}
+r-cran-repos=${conf.cranRepo}
diff --git a/files/default/hopsworks_templates/sparklyr_config.yml b/files/default/hopsworks_templates/sparklyr_config.yml
@@ -0,0 +1,10 @@
+default:
+  livy.driverCores: ${conf.driverCores}
+  livy.driverMemory: "${conf.driverMemory}"
+  livy.numExecutors: ${conf.numExecutors}
+  livy.executorCores: ${conf.executorCores}
+  livy.executorMemory: "${conf.executorMemory}"
+  livy.proxyUser: "${conf.proxyUser}"
+  livy.queue: "${conf.yarnQueue}"
+  livy.name: "${conf.livyAppName}"
+${conf.sparkConfiguration}
diff --git a/files/default/sql/ddl/3.4.0__initial_tables.sql b/files/default/sql/ddl/3.4.0__initial_tables.sql
@@ -824,6 +824,7 @@ CREATE TABLE `project` (
                            `topic_name` VARCHAR(255) DEFAULT NULL,
                            `python_env_id` int(11) DEFAULT NULL,
                            `creation_status` tinyint(1) NOT NULL DEFAULT '0',
+                           `rstudio_docker_image` varchar(255) COLLATE latin1_general_cs DEFAULT NULL,
                            PRIMARY KEY (`id`),
                            UNIQUE KEY `projectname` (`projectname`),
                            KEY `user_idx` (`username`),
@@ -1084,17 +1085,21 @@ CREATE TABLE `rstudio_interpreter` (
 /*!40101 SET @saved_cs_client     = @@character_set_client */;
 /*!40101 SET character_set_client = utf8 */;
 CREATE TABLE `rstudio_project` (
-                                   `port` int(11) NOT NULL,
+                                   `port` int NOT NULL,
+                                   `hdfs_user_id` int NOT NULL,
                                    `created` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP,
+                                   `expires` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP,
                                    `last_accessed` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
-                                   `host_ip` varchar(255) COLLATE latin1_general_cs NOT NULL,
-                                   `token` varchar(255) COLLATE latin1_general_cs NOT NULL,
-                                   `secret` varchar(64) COLLATE latin1_general_cs NOT NULL,
-                                   `pid` bigint(20) NOT NULL,
-                                   `project_id` int(11) NOT NULL,
+                                   `secret` varchar(64) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                   `pid` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                   `project_id` int NOT NULL,
+                                   `login_password` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                   `login_username` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
                                    PRIMARY KEY (`port`),
+                                   KEY `hdfs_user_idx` (`hdfs_user_id`),
                                    KEY `project_id` (`project_id`),
-                                   CONSTRAINT `FK_284_578` FOREIGN KEY (`project_id`) REFERENCES `project` (`id`) ON DELETE CASCADE ON UPDATE NO ACTION
+                                   CONSTRAINT `FK_103_577` FOREIGN KEY (`hdfs_user_id`) REFERENCES `hops`.`hdfs_users` (`id`) ON DELETE CASCADE,
+                                   CONSTRAINT `FK_284_578` FOREIGN KEY (`project_id`) REFERENCES `project` (`id`) ON DELETE CASCADE
 ) ENGINE=ndbcluster DEFAULT CHARSET=latin1 COLLATE=latin1_general_cs;
 /*!40101 SET character_set_client = @saved_cs_client */;
 
@@ -1107,39 +1112,43 @@ CREATE TABLE `rstudio_project` (
 CREATE TABLE `rstudio_settings` (
                                     `project_id` int(11) NOT NULL,
                                     `team_member` varchar(150) COLLATE latin1_general_cs NOT NULL,
-                                    `num_tf_ps` int(11) DEFAULT '1',
-                                    `num_tf_gpus` int(11) DEFAULT '0',
-                                    `num_mpi_np` int(11) DEFAULT '1',
-                                    `appmaster_cores` int(11) DEFAULT '1',
-                                    `appmaster_memory` int(11) DEFAULT '1024',
-                                    `num_executors` int(11) DEFAULT '1',
-                                    `num_executor_cores` int(11) DEFAULT '1',
-                                    `executor_memory` int(11) DEFAULT '1024',
-                                    `dynamic_initial_executors` int(11) DEFAULT '1',
-                                    `dynamic_min_executors` int(11) DEFAULT '1',
-                                    `dynamic_max_executors` int(11) DEFAULT '1',
                                     `secret` varchar(255) COLLATE latin1_general_cs NOT NULL,
-                                    `log_level` varchar(32) COLLATE latin1_general_cs DEFAULT 'INFO',
-                                    `mode` varchar(32) COLLATE latin1_general_cs NOT NULL,
-                                    `umask` varchar(32) COLLATE latin1_general_cs DEFAULT '022',
                                     `advanced` tinyint(1) DEFAULT '0',
-                                    `archives` varchar(1500) COLLATE latin1_general_cs DEFAULT '',
-                                    `jars` varchar(1500) COLLATE latin1_general_cs DEFAULT '',
-                                    `files` varchar(1500) COLLATE latin1_general_cs DEFAULT '',
-                                    `py_files` varchar(1500) COLLATE latin1_general_cs DEFAULT '',
-                                    `spark_params` varchar(6500) COLLATE latin1_general_cs DEFAULT '',
                                     `shutdown_level` int(11) NOT NULL DEFAULT '6',
+                                    `base_dir` varchar(255) COLLATE latin1_general_cs DEFAULT NULL,
+                                    `job_config` varchar(11000) COLLATE latin1_general_cs DEFAULT NULL,
+                                    `docker_config` varchar(1000) COLLATE latin1_general_cs DEFAULT NULL,
                                     PRIMARY KEY (`project_id`,`team_member`),
                                     KEY `team_member` (`team_member`),
                                     KEY `secret_idx` (`secret`),
-                                    CONSTRAINT `RS_FK_USERS` FOREIGN KEY (`team_member`) REFERENCES `users` (`email`) ON DELETE CASCADE ON UPDATE NO ACTION,
+                                    CONSTRAINT `RS_FK_USERS` FOREIGN KEY (`team_member`) REFERENCES `users` (`email`) ON DELETE CASCADE ON UPDATE NO
+                                        ACTION,
                                     CONSTRAINT `RS_FK_PROJS` FOREIGN KEY (`project_id`) REFERENCES `project` (`id`) ON DELETE CASCADE ON UPDATE NO ACTION
 ) ENGINE=ndbcluster DEFAULT CHARSET=latin1 COLLATE=latin1_general_cs;
 /*!40101 SET character_set_client = @saved_cs_client */;
 
 /*!40101 SET character_set_client = utf8 */;
 /*!40101 SET character_set_client = @saved_cs_client */;
 
+CREATE TABLE `rstudio_environment_build` (
+                                             `id` int NOT NULL AUTO_INCREMENT,
+                                             `build_script` varchar(1000) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                             `user` int NOT NULL,
+                                             `project` int NOT NULL,
+                                             `build_start` bigint DEFAULT NULL,
+                                             `build_finish` bigint DEFAULT NULL,
+                                             `build_result` varchar(128) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                             `secret` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                             `logFile` varchar(1000) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                             `build_name` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                             `description` varchar(1000) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                             PRIMARY KEY (`id`),
+                                             KEY `user_fk` (`user`),
+                                             KEY `rstudio_env_build_project_fk` (`project`),
+                                             CONSTRAINT `rstudio_env_build_project_fk` FOREIGN KEY (`project`) REFERENCES `project` (`id`) ON DELETE CASCADE,
+                                             CONSTRAINT `rstudio_env_build_usr_fkc` FOREIGN KEY (`user`) REFERENCES `users` (`uid`) ON DELETE CASCADE
+) ENGINE=ndbcluster AUTO_INCREMENT=5154 DEFAULT CHARSET=latin1;
+
 --
 -- Table structure for table `serving`
 --

diff --git a/files/default/sql/ddl/updates/3.4.0.sql b/files/default/sql/ddl/updates/3.4.0.sql
@@ -177,9 +177,57 @@ WHERE REGEXP_SUBSTR(`subject`, "^([0-9]+)_([0-9]+)_(.+)_([0-9]+)(_onlinefs|$)");
 UPDATE `hopsworks`.`feature_group` `fg`
     JOIN `hopsworks`.`feature_store` `fs` ON `fg`.`feature_store_id` = `fs`.`id`
 SET `fg`.`topic_name` = CONCAT(fs.project_id, "_", fg.id, "_", fg.name, "_", fg.version, IF(fg.online_enabled , "_onlinefs", ""));
-SET SQL_SAFE_UPDATES = 1;
 
 -- FSTORE-1010: Don't leave orphaned subjects when deleting online enabled fg
 UPDATE `hopsworks`.`project_topics`
 SET `subject_id` = NULL
-    WHERE REGEXP_SUBSTR(`topic_name`, "^([0-9]+)_([0-9]+)_(.+)_([0-9]+)(_onlinefs|$)");
+    WHERE REGEXP_SUBSTR(`topic_name`, "^([0-9]+)_([0-9]+)_(.+)_([0-9]+)(_onlinefs|$)");
+
+SET SQL_SAFE_UPDATES = 1;
+
+
+-- Rstudio updates
+ALTER TABLE `hopsworks`.`rstudio_project` DROP COLUMN `host_ip`,
+    DROP COLUMN `token`;
+
+ALTER TABLE `hopsworks`.`rstudio_project` ADD COLUMN `login_password` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL;
+ALTER TABLE `hopsworks`.`rstudio_project` ADD COLUMN `login_username` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL;
+
+ALTER TABLE `hopsworks`.`rstudio_project` MODIFY COLUMN `pid` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL;
+
+ALTER TABLE `hopsworks`.`rstudio_settings`
+    DROP COLUMN num_tf_ps, DROP COLUMN num_tf_gpus,
+    DROP COLUMN num_mpi_np, DROP COLUMN appmaster_cores,
+    DROP COLUMN appmaster_memory, DROP COLUMN num_executors,
+    DROP COLUMN num_executor_cores, DROP COLUMN executor_memory,
+    DROP COLUMN dynamic_initial_executors, DROP COLUMN dynamic_min_executors,
+    DROP COLUMN dynamic_max_executors, DROP COLUMN log_level,
+    DROP COLUMN mode, DROP COLUMN umask,
+    DROP COLUMN archives, DROP COLUMN jars,
+    DROP COLUMN files, DROP COLUMN py_files,
+    DROP COLUMN spark_params;
+
+ALTER TABLE `hopsworks`.`rstudio_settings` ADD COLUMN  `base_dir` varchar(255) COLLATE latin1_general_cs DEFAULT NULL;
+ALTER TABLE `hopsworks`.`rstudio_settings` ADD COLUMN  `job_config` varchar(11000) COLLATE latin1_general_cs DEFAULT NULL;
+ALTER TABLE `hopsworks`.`rstudio_settings` ADD COLUMN  `docker_config` varchar(1000) COLLATE latin1_general_cs DEFAULT NULL;
+
+ALTER TABLE `hopsworks`.`project` ADD COLUMN `rstudio_docker_image` VARCHAR(255) COLLATE latin1_general_cs DEFAULT NULL;
+
+CREATE TABLE `rstudio_environment_build` (
+                                             `id` int NOT NULL AUTO_INCREMENT,
+                                             `build_script` varchar(1000) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                             `user` int NOT NULL,
+                                             `project` int NOT NULL,
+                                             `build_start` bigint DEFAULT NULL,
+                                             `build_finish` bigint DEFAULT NULL,
+                                             `build_result` varchar(128) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                             `secret` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs NOT NULL,
+                                             `logFile` varchar(1000) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                             `build_name` varchar(255) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                             `description` varchar(1000) CHARACTER SET latin1 COLLATE latin1_general_cs DEFAULT NULL,
+                                             PRIMARY KEY (`id`),
+                                             KEY `user_fk` (`user`),
+                                             KEY `rstudio_env_build_project_fk` (`project`),
+                                             CONSTRAINT `rstudio_env_build_project_fk` FOREIGN KEY (`project`) REFERENCES `project` (`id`) ON DELETE CASCADE,
+                                             CONSTRAINT `rstudio_env_build_usr_fkc` FOREIGN KEY (`user`) REFERENCES `users` (`uid`) ON DELETE CASCADE
+) ENGINE=ndbcluster AUTO_INCREMENT=5154 DEFAULT CHARSET=latin1;
diff --git a/metadata.rb b/metadata.rb
@@ -51,6 +51,7 @@
 recipe  "hopsworks::migrate", "Call expat to migrate between Hopsworks versions"
 
 recipe  "hopsworks::purge", "Deletes glassfish installation."
+recipe  "hopsworks::reindex", "Reindex the featurestore search index"
 #######################################################################################
 # Required Attributes
 #######################################################################################

diff --git a/providers/worker.rb b/providers/worker.rb
@@ -37,8 +37,12 @@
               start_domain_timeout: systemd_start_timeout,
               stop_domain_timeout: systemd_stop_timeout,
               authbind: new_resource.requires_authbind)
-    notifies :start, "service[#{service_name}]", :delayed
   end
+
+  kagent_config service_name do
+    action :systemd_reload
+  end
+
   if node['services']['enabled'].casecmp?("true")
     service service_name do
       supports start: true, restart: true, stop: true, status: true