Address comments

Author: anlowee

yscope-k8s/.gitignore

@@ -0,0 +1 @@
+demo-assets/clp-config.yml

yscope-k8s/.helmignore

@@ -16,9 +16,12 @@
 *.tmp
 *.orig
 *~
+
 # Various IDEs
 .project
 .idea/
 *.tmproj
 .vscode/
-clp/
+
+# Demo assets
+demo-assets/

yscope-k8s/README.md

@@ -30,44 +30,7 @@ Follow the guide here: [helm]
 
 3. Replace the content of `etc/clp-config.yml` with the following (also replace the IP address `${REPLACE_IP}` with the actual IP address of the host that you are running the clp-package):
 ```yaml
-package:
-  storage_engine: "clp-s"
-database:
-  type: "mariadb"
-  host: "${REPLACE_IP}"
-  port: 6001
-  name: "clp-db"
-query_scheduler:
-  host: "${REPLACE_IP}"
-  port: 6002
-  jobs_poll_delay: 0.1
-  num_archives_to_search_per_sub_job: 16
-  logging_level: "INFO"
-queue:
-  host: "${REPLACE_IP}"
-  port: 6003
-redis:
-  host: "${REPLACE_IP}"
-  port: 6004
-  query_backend_database: 0
-  compression_backend_database: 1
-reducer:
-  host: "${REPLACE_IP}"
-  base_port: 6100
-  logging_level: "INFO"
-  upsert_interval: 100
-results_cache:
-  host: "${REPLACE_IP}"
-  port: 6005
-  db_name: "clp-query-results"
-  stream_collection_name: "stream-files"
-webui:
-  host: "localhost"
-  port: 6000
-  logging_level: "INFO"
-log_viewer_webui:
-  host: "localhost"
-  port: 6006
+
 ```
 
 4. Launch:
@@ -82,7 +45,13 @@ sbin/start-clp.sh
 sbin/compress.sh --timestamp-key 't.dollar_sign_date' datasets/mongod-256MB-processed.log
 ```
 
-6. Use a JetBrain IDE to connect the database source. The database is `clp-db`, the user is `clp-user` and the password is in `etc/credential.yml`. Then modify the `archive_storage_directory` field in `clp_datasets` table to `/var/data/archives/default`, and submit the change.
+6. Use the following command to update the CLP metadata database so that the worker can find the archives in right place:
+```bash
+# Install mysql-client if necessary
+sudo apt update && sudo apt install -y mysql-client
+# Find the user and password in clp-json-package/etc/credential.yml
+mysql -h ${REPLACE_IP} -P 6001 -u ${REPLACE_USER} -p'${REPLACE_PASSWORD}' clp-db -e "UPDATE clp_datasets SET archive_storage_directory = '/var/data/archives/default';"
+```
 
 # Create k8s Cluster
 Create a local k8s cluster with port forwarding

yscope-k8s/demo-assets/clp-config.yml.bak

@@ -0,0 +1,38 @@
+package:
+  storage_engine: "clp-s"
+database:
+  type: "mariadb"
+  host: "${REPLACE_IP}"
+  port: 6001
+  name: "clp-db"
+query_scheduler:
+  host: "${REPLACE_IP}"
+  port: 6002
+  jobs_poll_delay: 0.1
+  num_archives_to_search_per_sub_job: 16
+  logging_level: "INFO"
+queue:
+  host: "${REPLACE_IP}"
+  port: 6003
+redis:
+  host: "${REPLACE_IP}"
+  port: 6004
+  query_backend_database: 0
+  compression_backend_database: 1
+reducer:
+  host: "${REPLACE_IP}"
+  base_port: 6100
+  logging_level: "INFO"
+  upsert_interval: 100
+results_cache:
+  host: "${REPLACE_IP}"
+  port: 6005
+  db_name: "clp-query-results"
+  stream_collection_name: "stream-files"
+webui:
+  host: "localhost"
+  port: 6000
+  logging_level: "INFO"
+log_viewer_webui:
+  host: "localhost"
+  port: 6006

yscope-k8s/demo-assets/init.sh

@@ -0,0 +1,17 @@
+#!/usr/bin/env bash
+
+if [ "$#" -ne 1 ]; then
+    echo "Usage: $0 <ip-address>"
+    exit 1
+fi
+
+SCRIPT_PATH="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
+
+IP="$1"
+FILE="${SCRIPT_PATH}/clp-config.yml"
+
+cp "${FILE}.bak" "$FILE"
+
+sed -i "s|\${REPLACE_IP}|$IP|g" "$FILE"
+
+echo "Replaced \${REPLACE_IP} with $IP in $FILE"

yscope-k8s/mongod-256MB-presto-clp.log.tar.gz renamed to yscope-k8s/demo-assets/mongod-256MB-presto-clp.log.tar.gz

@@ -1,4 +1,3 @@
-apiVersion: "batch/v1"
 kind: "Job"
 metadata:
   name: "bucket-creation"

yscope-k8s/presto-cli-0.293-executable.jar renamed to yscope-k8s/demo-assets/presto-cli-0.293-executable.jar

@@ -1,4 +1,3 @@
-apiVersion: "v1"
 kind: "Pod"
 metadata:
   labels:

yscope-k8s/templates/object-store/bucket-creation.yaml

@@ -6,9 +6,6 @@ metadata:
   name: "{{ .Values.presto.coordinator.serviceName }}"
 spec:
   initContainers:
-          #    - name: "start-clp-package"
-          #image: "busybox"
-          #command: "/scripts/start-clp-package.sh"
     - name: "generate-configs"
       image: "busybox"
       command: ["/scripts/generate-configs.sh"]
@@ -37,7 +34,6 @@ spec:
     - name: "presto-coordinator-config-volume"
       configMap:
         name: "presto-coordinator-config"
-    # TODO: add CLP data directory volume host path
     - name: "clp-archives-volume"
       hostPath:
         path: "/var/data/archives"