Landing request tracking

group_vars/galaxy_db_servers/vars.yaml

@@ -1,5 +1,17 @@
 ---
 
+galaxy_db_servers_group_templates:
+  - src: templates/telegraf/workflow_landing_requests.sh.j2
+    dest: "/usr/local/bin/workflow_landing_requests.sh"
+    owner: root
+    group: root
+    mode: '0755'
+  - src: templates/telegraf/tool_landing_requests.sh.j2
+    dest: "/usr/local/bin/tool_landing_requests.sh"
+    owner: root
+    group: root
+    mode: '0755'
+
 # Always restart PostgreSQL on exit
 galaxy_db_servers_group_systemd_overrides:
   - name: "postgresql-{{ __postgresql_version_dotless }}"
@@ -169,6 +181,24 @@ galaxy_db_servers_group_telegraf_plugins_extra:
       - interval = "1m"
       - '[inputs.exec.tags]'
       - '  influxdb_database = "{{ galaxy_instance_codename }}_sql"'
+  galaxy_workflow_landing_requests:
+    plugin: "exec"
+    config:
+      - commands = ["/usr/local/bin/workflow_landing_requests.sh"]
+      - timeout = "30s"
+      - data_format = "influx"
+      - interval = "5m"
+      - '[inputs.exec.tags]'
+      - '  influxdb_database = "{{ galaxy_instance_codename }}_sql"'
+  galaxy_tool_landing_requests:
+    plugin: "exec"
+    config:
+      - commands = ["/usr/local/bin/tool_landing_requests.sh"]
+      - timeout = "30s"
+      - data_format = "influx"
+      - interval = "5m"
+      - '[inputs.exec.tags]'
+      - '  influxdb_database = "{{ galaxy_instance_codename }}_sql"'
   galaxy_disk_usage:
     plugin: "exec"
     config:

inventory/tacc.yaml

@@ -230,9 +230,7 @@ all:
         galaxy-main4.tacc.utexas.edu:
           ansible_become: true
         galaxy-db.tacc.utexas.edu:
-          # annoyingly, this overrides become on plays
-          #ansible_become: false
-          ansible_user: root
+          ansible_become: true
         galaxy-vgp.tacc.utexas.edu:
           #ansible_become: false
           ansible_user: root

templates/telegraf/tool_landing_requests.sh.j2

@@ -0,0 +1,72 @@
+#!/bin/bash
+set -euo pipefail
+
+# Query tool_landing_request table for metrics
+# Output in InfluxDB line protocol format for Telegraf
+
+PGDATABASE="galaxy_{{ galaxy_instance_codename }}"
+
+# Total tool landing requests
+total=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM tool_landing_request;")
+echo "tool_landing_requests_total value=${total}i"
+
+# Requests by day (last 30 days)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r day count; do
+SELECT
+    DATE(create_time) as day,
+    COUNT(*) as count
+FROM tool_landing_request
+WHERE create_time >= NOW() - INTERVAL '30 days'
+GROUP BY DATE(create_time)
+ORDER BY day;
+EOF
+    echo "tool_landing_requests_daily,day=${day} count=${count}i"
+done
+
+# Requests by tool_id (top 10)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r tool_id count; do
+SELECT
+    COALESCE(tool_id, 'null') as tool_id,
+    COUNT(*) as count
+FROM tool_landing_request
+WHERE create_time >= NOW() - INTERVAL '7 days'
+GROUP BY tool_id
+ORDER BY count DESC
+LIMIT 10;
+EOF
+    tool_id_escaped=$(echo "$tool_id" | sed 's/[,= ]/\\&/g')
+    echo "tool_landing_requests_by_tool,tool_id=${tool_id_escaped} count=${count}i"
+done
+
+# Public vs Private requests (last 24 hours)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r visibility count; do
+SELECT
+    CASE WHEN public THEN 'public' ELSE 'private' END as visibility,
+    COUNT(*) as count
+FROM tool_landing_request
+WHERE create_time >= NOW() - INTERVAL '24 hours'
+GROUP BY public;
+EOF
+    echo "tool_landing_requests_visibility,visibility=${visibility} count=${count}i"
+done
+
+# Requests with user vs anonymous (last 24 hours)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r user_type count; do
+SELECT
+    CASE WHEN user_id IS NULL THEN 'anonymous' ELSE 'registered' END as user_type,
+    COUNT(*) as count
+FROM tool_landing_request
+WHERE create_time >= NOW() - INTERVAL '24 hours'
+GROUP BY (user_id IS NULL);
+EOF
+    echo "tool_landing_requests_user_type,user_type=${user_type} count=${count}i"
+done
+
+# Recent activity (last 1 hour, 24 hours, 7 days)
+recent_1h=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM tool_landing_request WHERE create_time >= NOW() - INTERVAL '1 hour';")
+recent_24h=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM tool_landing_request WHERE create_time >= NOW() - INTERVAL '24 hours';")
+recent_7d=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM tool_landing_request WHERE create_time >= NOW() - INTERVAL '7 days';")
+
+echo "tool_landing_requests_recent,period=1h value=${recent_1h}i"
+echo "tool_landing_requests_recent,period=24h value=${recent_24h}i"
+echo "tool_landing_requests_recent,period=7d value=${recent_7d}i"

templates/telegraf/workflow_landing_requests.sh.j2

@@ -0,0 +1,72 @@
+#!/bin/bash
+set -euo pipefail
+
+# Query workflow_landing_request table for metrics
+# Output in InfluxDB line protocol format for Telegraf
+
+PGDATABASE="galaxy_{{ galaxy_instance_codename }}"
+
+# Total workflow landing requests
+total=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM workflow_landing_request;")
+echo "workflow_landing_requests_total value=${total}i"
+
+# Requests by day (last 30 days)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r day count; do
+SELECT
+    DATE(create_time) as day,
+    COUNT(*) as count
+FROM workflow_landing_request
+WHERE create_time >= NOW() - INTERVAL '30 days'
+GROUP BY DATE(create_time)
+ORDER BY day;
+EOF
+    echo "workflow_landing_requests_daily,day=${day} count=${count}i"
+done
+
+# Requests by workflow_id (top 10)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r workflow_id count; do
+SELECT
+    COALESCE(workflow_id::text, 'null') as workflow_id,
+    COUNT(*) as count
+FROM workflow_landing_request
+WHERE create_time >= NOW() - INTERVAL '7 days'
+GROUP BY workflow_id
+ORDER BY count DESC
+LIMIT 10;
+EOF
+    workflow_id_escaped=$(echo "$workflow_id" | sed 's/[,= ]/\\&/g')
+    echo "workflow_landing_requests_by_workflow,workflow_id=${workflow_id_escaped} count=${count}i"
+done
+
+# Public vs Private requests (last 24 hours)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r visibility count; do
+SELECT
+    CASE WHEN public THEN 'public' ELSE 'private' END as visibility,
+    COUNT(*) as count
+FROM workflow_landing_request
+WHERE create_time >= NOW() - INTERVAL '24 hours'
+GROUP BY public;
+EOF
+    echo "workflow_landing_requests_visibility,visibility=${visibility} count=${count}i"
+done
+
+# Requests with user vs anonymous (last 24 hours)
+psql -d "$PGDATABASE" -t -A -F',' <<EOF | while IFS=',' read -r user_type count; do
+SELECT
+    CASE WHEN user_id IS NULL THEN 'anonymous' ELSE 'registered' END as user_type,
+    COUNT(*) as count
+FROM workflow_landing_request
+WHERE create_time >= NOW() - INTERVAL '24 hours'
+GROUP BY (user_id IS NULL);
+EOF
+    echo "workflow_landing_requests_user_type,user_type=${user_type} count=${count}i"
+done
+
+# Recent activity (last 1 hour, 24 hours, 7 days)
+recent_1h=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM workflow_landing_request WHERE create_time >= NOW() - INTERVAL '1 hour';")
+recent_24h=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM workflow_landing_request WHERE create_time >= NOW() - INTERVAL '24 hours';")
+recent_7d=$(psql -d "$PGDATABASE" -t -A -c "SELECT COUNT(*) FROM workflow_landing_request WHERE create_time >= NOW() - INTERVAL '7 days';")
+
+echo "workflow_landing_requests_recent,period=1h value=${recent_1h}i"
+echo "workflow_landing_requests_recent,period=24h value=${recent_24h}i"
+echo "workflow_landing_requests_recent,period=7d value=${recent_7d}i"