Add portworx integration

apps/portworx.yaml

@@ -0,0 +1,15 @@
+---
+apiVersion: v1
+kind: App
+name: "Portworx"
+keywords: 
+  - Storage
+  - Available
+availableVersions: 
+  - '2.9'
+shortDescription: "Portworx is an end-to-end storage and data management solution for Kubernetes"
+description: |
+  Portworx provides a fully integrated solution for persistent storage, data protection, disaster recovery, data security, cross-cloud and data migrations, and automated capacity management for applications running on Kubernetes.
+icon: https://raw.githubusercontent.com/sysdiglabs/promcat-resources/master/apps/images/portworx.svg
+website: https://portworx.com/
+available: true

resources/portworx/ALERTS.md

@@ -0,0 +1,61 @@
+# Alerts
+## No Quorum
+Portworx No Quorum.
+
+## Node Status Not OK
+Portworx Node Status Not OK.
+
+## Offline Nodes
+Portworx Offline Nodes.
+
+## Nodes Storage Full or Down
+Portworx Nodes Storage Full or Down.
+
+## Offline Storage Nodes
+Portworx Offline Storage Nodes.
+
+## Unhealthy Node KVDB
+Portworx Unhealthy Node KVDB.
+
+## Cache read hit rate is low
+Portworx Cache read hit rate is low.
+
+## Cache write hit rate is low
+Portworx Cache write hit rate is low.
+
+## High Read Latency In Disk
+Portworx High Read Latency In Disk.
+
+## High Write Latency In Disk
+Portworx High Write Latency In Disk.
+
+## Low Cluster Capacity
+Portworx Low Cluster Capacity.
+
+## Disk Full In 48H
+Portworx Disk Full In 48H.
+
+## Disk Full In 12H
+Portworx Disk Full In 12H.
+
+## Pool Status Not Online
+Portworx Node Status Not Online.
+
+## High Write Latency In Pool
+Portworx High Write Latency In Pool.
+
+## Pool Full In 48H
+Portworx Pool Full In 48H.
+
+## Pool Full In 12H
+Portworx Pool Full In 12H.
+
+## High Write Latency In Volume
+Portworx High Write Latency In Volume.
+
+## High Read Latency In Volume
+Portworx High Read Latency In Volume.
+
+## License Expiry
+Portworx License Expiry.
+

resources/portworx/INSTALL.md

@@ -0,0 +1,2 @@
+# Prerequisites
+Portworx instruments Prometheus metrics and annotates the pods with Prometheus annotations. 

resources/portworx/README.md

@@ -0,0 +1,12 @@
+# Portworx
+Portworx provides a fully integrated solution for persistent storage, data protection, disaster recovery, data security, cross-cloud and data migrations, and automated capacity management for applications running on Kubernetes.
+
+# Prometheus and exporters
+Portworx already has a Prometheus endpoint with all the metrics exposed on the port 9001 (port 17001 if deployed in Openshift). In Kubernetes the pod is already annotated, so with the Sysdig agent you can scrape the endpoint right away.
+
+# Metrics
+- Portworx cluster statistics
+- Portworx volumes statistics
+
+# Attributions
+Configuration files, dashboards and alerts are maintained by [Sysdig team](https://sysdig.com/).

resources/portworx/alerts.yaml

@@ -0,0 +1,164 @@
+apiVersion: v1
+kind: Alert
+app: Portworx
+version: 1.0.0
+appVersion:
+- '2.9'
+descriptionFile: ALERTS.md
+configurations:
+- kind: Prometheus
+  data: |-
+    groups:
+    - name: Portworx
+      rules:
+      - alert: '[Portworx] No Quorum'
+        expr: "px_cluster_status_quorum != 1"
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx No Quorum.
+      - alert: '[Portworx] Node Status Not OK'
+        expr: "px_node_status_node_status != 2"
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx Node Status Not OK.
+      - alert: '[Portworx] Offline Nodes'
+        expr: "px_cluster_status_nodes_offline > 0"
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx Offline Nodes.
+      - alert: '[Portworx] Nodes Storage Full or Down'
+        expr: "px_cluster_status_nodes_storage_down > 0"
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx Nodes Storage Full or Down.
+      - alert: '[Portworx] Offline Storage Nodes'
+        expr: "px_cluster_status_storage_nodes_offline > 0"
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx Offline Storage Nodes.
+      - alert: '[Portworx] Unhealthy Node KVDB'
+        expr: "px_kvdb_health_state_node_view == 2"
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx Unhealthy Node KVDB.
+      - alert: '[Portworx] Cache read hit rate is low'
+        expr: "px_px_cache_read_hits/( px_px_cache_read_hits +  px_px_cache_read_miss)< 0.80"
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Cache read hit rate is low.
+      - alert: '[Portworx] Cache write hit rate is low'
+        expr: "px_px_cache_write_hits/( px_px_cache_write_hits +  px_px_cache_write_miss)< 0.80"
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Cache write hit rate is low.
+      - alert: '[Portworx] High Read Latency In Disk'
+        expr: |
+          px_disk_stats_read_latency_seconds{ disk=~$disk} > 0.100
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx High Read Latency In Disk.
+      - alert: '[Portworx] High Write Latency In Disk'
+        expr: |
+          px_disk_stats_write_latency_seconds{ disk=~$disk} > 0.250
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx High Write Latency In Disk.
+      - alert: '[Portworx] Low Cluster Capacity'
+        expr: |
+          (sum (px_cluster_disk_available_bytes))/(sum (px_cluster_disk_total_bytes))< 0.10
+        for: 5m
+        labels:
+          severity: critical
+        annotations:
+          description: Portworx Low Cluster Capacity.
+      - alert: '[Portworx] Disk Full In 48H'
+        expr: |
+          predict_linear(px_cluster_disk_available_bytes[48h], 48 * 3600) < 0
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Disk Full In 48H.
+      - alert: '[Portworx] Disk Full In 12H'
+        expr: |
+          predict_linear(px_cluster_disk_available_bytes[12h], 12 * 3600) < 0
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Disk Full In 12H.
+      - alert: '[Portworx] Pool Status Not Online'
+        expr: "px_pool_stats_status{ pool=~$pool} != 1"
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Node Status Not Online.
+      - alert: '[Portworx] High Write Latency In Pool'
+        expr: |
+          px_pool_stats_write_latency_seconds{ pool=~$pool} > 0.250
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx High Write Latency In Pool.
+      - alert: '[Portworx] Pool Full In 48H'
+        expr: |
+          predict_linear(px_pool_stats_available_bytes{ pool=~$pool}[48h], 48 * 3600) < 0
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Pool Full In 48H.
+      - alert: '[Portworx] Pool Full In 12H'
+        expr: |
+          predict_linear(px_pool_stats_available_bytes{ pool=~$pool}[12h], 12 * 3600) < 0
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx Pool Full In 12H.
+      - alert: '[Portworx] High Write Latency In Volume'
+        expr: |
+          px_volume_write_latency_seconds{ pvc=~$pvc} > 0.250
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx High Write Latency In Volume.
+      - alert: '[Portworx] High Read Latency In Volume'
+        expr: |
+          px_volume_read_latency_seconds{ pvc=~$pvc} > 0.100
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx High Read Latency In Volume.
+      - alert: '[Portworx] License Expiry'
+        expr: |
+          min(px_node_status_license_expiry) < 30
+        for: 5m
+        labels:
+          severity: warning
+        annotations:
+          description: Portworx License Expiry.

resources/portworx/dashboards.yaml

@@ -0,0 +1,28 @@
+apiVersion: v1
+kind: Dashboard
+app: Portworx
+version: 1.0.0
+appVersion:
+- '2.9'
+configurations:
+- name: Cluster
+  kind: Sysdig
+  image: portworx/images/portworx_cluster_sysdig.png
+  description: |
+    This dashboard offers information on:
+    * Health
+    * Network
+    * Disk Stats
+    * Pool Stats
+    * Cache
+  file: include/Portworx_Cluster.json
+- name: Volumes
+  kind: Sysdig
+  image: portworx/images/portworx_volumes_sysdig.png
+  description: |
+    This dashboard offers information on:
+    * Volume Status
+    * Volume Capacity & Usage
+    * Volume Replication
+    * Volume IOPS
+  file: include/Portworx_Volumes.json

resources/portworx/description.yaml

@@ -0,0 +1,7 @@
+apiVersion: v1
+kind: Description
+app: Portworx
+version: 1.0.0
+appVersion:
+- '2.9'
+descriptionFile: README.md