Resolve L2SS-1821 "Support new pcon"

.gitlab-ci.yml

@@ -126,6 +126,9 @@ docker_build_image:
   script:
     #    Do not remove 'bash' or statement will be ignored by primitive docker shell
     - bash $CI_PROJECT_DIR/sbin/tag_and_push_docker_image.sh $IMAGE $tag
+  artifacts:
+    paths:
+      - docker/snmp-exporter/snmp.yml
 
 docker_build_image_device_base:
   extends: .base_docker_images

README.md

@@ -161,6 +161,7 @@ Next change the version in the following places:
 
 # Release Notes
 
+* 0.33.3 Add support for new PCON devices (ACX)
 * 0.33.2 Fix for XSTs in Observations: Also write FPGA_xst_nof_crosslets_RW
 * 0.33.1 SDPFirmware: replace FPGA_ucp monitoring points with new TR_ucp ones
 * 0.33.0 Run containers with dedicated ethernet devices (ovs/macvlan)

docker/snmp-exporter/README.md

+SNMP monitoring
+======================
+
+The `snmp-exporter` module allows SNMP devices to be periodically queried to obtain Prometheus metrics describing their state.
+
+This setup relies on the following parts:
+
+* `docker/snmp-exporter/`, a daemon derived from `prom/snmp-exporter` that allows on-demand querying of SNMP devices and is extended with custom LOFAR MIBs,
+* `infra/jobs/station/snmp-exporter.levant.nomad`, the Nomad job file to run `snmp-exporter`,
+* `infra/consul/consul.hcl.j2`, which configures settings for SNMP "services" present,
+* `infra/jobs/station/monitoring.levant.nomad`, the Nomad job file that has Prometheus query `snmp-exporter` for each SNMP "service".

docker/snmp-exporter/generator.yml

@@ -4,6 +4,12 @@ auths:
     version: 1
   public_v2:
     version: 2
+  pcon_acc:
+    community: accread
+    version: 1
+  pcon_acx:
+    community: accread
+    version: 2
 
 modules:
   # Default IF-MIB interfaces table with ifIndex.
@@ -34,6 +40,76 @@ modules:
   readynas:
     walk:
       - 1.3.6.1.4.1.4526               # Raid/Disks status
+
+
+# ACX Power Systems / Unipower
+#
+# https://www.unipowerco.com/MIB_files/ACX-MIB.mib
+  powecMIB:
+    allow_nonincreasing_oids: true
+    walk:
+      - systemInfo
+      - 1.3.6.1.4.1.5961.5.2 # powerSystem (this name clashes with the eltek MIBs)
+      - uFlag
+      - autoBoostFlag
+      - autoTestFlag
+      - numbOfAlarms
+      - alarmHistoryCount
+      - rectifierInfo
+      - powecACXTrap
+    overrides:
+      systemVoltage:
+        scale: 0.01
+      loadCurrent:
+        scale: 0.1
+      batteryCurrent:
+        ignore: true
+      rectifierCurrent:
+        scale: 0.1
+      dcdcCurrent:
+        scale: 0.1
+      meanCurrent:
+        scale: 0.1
+      outputCurrent:
+        scale: 0.1
+      outputVoltage:
+        scale: 0.1
+      batteryDescript:
+        ignore: true
+      batteryCapacity:
+        ignore: true
+      remainBatteryTime:
+        ignore: true
+
+  # gosnmp does not visit these in GETBULK, even though snmpwalk does
+  # see https://github.com/gosnmp/gosnmp/issues/467
+  powecMIB_rectifier3:
+    allow_nonincreasing_oids: true
+    walk:
+      # node: rectifierInfo.moduleTable
+      - 1.3.6.1.4.1.5961.5.6.2.1.1.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.2.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.3.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.4.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.5.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.6.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.7.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.8.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.9.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.10.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.11.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.12.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.13.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.14.3
+      - 1.3.6.1.4.1.5961.5.6.2.1.15.3
+    overrides:
+      meanCurrent:
+        scale: 0.1
+      outputCurrent:
+        scale: 0.1
+      outputVoltage:
+        scale: 0.1
+
 #
 # White Rabbit
 #

infra/jobs/station/monitoring.levant.nomad

@@ -290,6 +290,8 @@ postgres.service.consul:5432:grafana:postgres:password
                 target_label: __param_module
               - source_labels: [__meta_consul_service_address]
                 target_label: __param_target
+              - source_labels: [__meta_consul_service_metadata_auth]
+                target_label: __param_auth
               - target_label: __address__
                 replacement: "snmp-exporter.service.consul:9116"
               - source_labels: [__meta_consul_service]

infra/jobs/station/snmp-exporter.levant.nomad

@@ -28,6 +28,7 @@ job "snmp-exporter" {
       config {
         image   = "[[.registry.astron.url]]/snmp-exporter:[[.image_tag]]"
         ports   = ["http"]
+        args    = ["--config.file=/etc/snmp_exporter/snmp.yml", "--snmp.module-concurrency=3", "--snmp.wrap-large-counters"]
       }
       resources {
         cpu    = 100

infra/station/consul/consul.hcl.j2

@@ -59,23 +59,25 @@ services {
 }
 
 services {
-  name = "pcon1-snmp"
-  id = "pcon1-snmp"
+  name = "pcon0-snmp"
+  id = "pcon0-snmp"
   address = "10.99.250.70"
   port = 161
   tags = ["snmp-scrape"]
   meta = {
-    mibs = "eltek"
+    mibs = "powecMIB,powecMIB_rectifier3"
+    auth = "pcon_acx"
   }
 }
 
 services {
-  name = "pcon2-snmp"
-  id = "pcon2-snmp"
+  name = "pcon1-snmp"
+  id = "pcon1-snmp"
   address = "10.99.250.71"
   port = 161
   tags = ["snmp-scrape"]
   meta = {
-    mibs = "eltek"
+    mibs = "powecMIB,powecMIB_rectifier3"
+    auth = "pcon_acx"
   }
 }

tangostationcontrol/VERSION

-0.33.2
+0.33.3