Hello,
I am currently attempting to deploy ORS on Google Cloud Run but am encountering issues with graph generation (guess). Currently, I am utilizing the example-heidelberg.osm.gz for testing purposes. Below is an outline of my setup:
I have configured a separate Cloud Storage bucket for each volume. These appear to be functioning correctly (but no graphs are being constructed in home/ors/graphs). I have set four environment variables to support this configuration.
Additionally, I here’s the YAML configurations that were generated by the service.:
YAML FILE
apiVersion: serving.knative.dev/v1
kind: Service
metadata:
name: ors
namespace: '00000000000000000000000000'
selfLink: /apis/serving.knative.dev/v1/namespaces/00000000000000000000000000/services/ors
uid: 00000000000000000000000000
resourceVersion: AAYW8fdvmr8
generation: 24
creationTimestamp: '2024-04-25T18:21:20.179090Z'
labels:
run.googleapis.com/satisfiesPzs: 'true'
cloud.googleapis.com/location: europe-west9
annotations:
run.googleapis.com/client-name: cloud-console
serving.knative.dev/creator: 00000000000000000000000000@gmail.com
serving.knative.dev/lastModifier: 00000000000000000000000000@gmail.com
run.googleapis.com/launch-stage: BETA
run.googleapis.com/operation-id: 00000000000000000000000000
run.googleapis.com/ingress: all
run.googleapis.com/ingress-status: all
run.googleapis.com/minScale: '0'
run.googleapis.com/max-surge: '0'
spec:
template:
metadata:
labels:
client.knative.dev/nonce: 00000000000000000000000000
run.googleapis.com/startupProbeType: Default
annotations:
run.googleapis.com/client-name: cloud-console
autoscaling.knative.dev/minScale: '1'
run.googleapis.com/execution-environment: gen2
autoscaling.knative.dev/maxScale: '1'
run.googleapis.com/startup-cpu-boost: 'true'
spec:
containerConcurrency: 80
timeoutSeconds: 300
serviceAccountName: 00000000000000000000000000-compute@developer.gserviceaccount.com
containers:
- name: openrouteservice-1
image: openrouteservice/openrouteservice
ports:
- name: http1
containerPort: 8082
env:
- name: CONTAINER_LOG_LEVEL
value: DEBUG
- name: XMS
value: 2g
- name: XMX
value: 4g
- name: REBUILD_GRAPHS
value: 'true'
resources:
limits:
cpu: 2000m
memory: 4Gi
volumeMounts:
- name: graphs
mountPath: /home/ors/graphs
- name: elevation_cache
mountPath: /home/ors/elevation_cache
- name: config
mountPath: /home/ors/config
- name: logs
mountPath: /home/ors/logs
- name: files
mountPath: /home/ors/files
startupProbe:
timeoutSeconds: 240
periodSeconds: 240
failureThreshold: 1
tcpSocket:
port: 8082
volumes:
- name: graphs
csi:
driver: gcsfuse.run.googleapis.com
volumeAttributes:
bucketName: graphs-ors
- name: elevation_cache
csi:
driver: gcsfuse.run.googleapis.com
volumeAttributes:
bucketName: elevation_cache
- name: config
csi:
driver: gcsfuse.run.googleapis.com
volumeAttributes:
bucketName: config-ors
- name: logs
csi:
driver: gcsfuse.run.googleapis.com
volumeAttributes:
bucketName: logs-ors
- name: files
csi:
driver: gcsfuse.run.googleapis.com
volumeAttributes:
bucketName: files-ors
traffic:
- percent: 100
latestRevision: true
And the log from /home/ors/logs
LOG
2024-04-25 20:46:02 INFO [o.h.o.a.s.l.ORSInitContextListener ] - Initializing ORS...
2024-04-25 20:46:02 INFO [o.h.o.r.RoutingProfileManager ] - Total - 1024 MB, Free - 965.46 MB, Max: 2 GB, Used - 59.04 MB
2024-04-25 20:46:02 INFO [o.h.o.r.RoutingProfileManager ] - ====> Initializing profiles from '/home/ors/files/example-heidelberg.osm.gz' (1 threads) ...
2024-04-25 20:46:02 INFO [o.h.o.r.RoutingProfileManager ] - 1 profile configurations submitted as tasks.
2024-04-25 20:46:02 INFO [o.h.o.r.RoutingProfile ] - [1] Profiles: 'driving-car', location: '/home/ors/./graphs/car'.
2024-04-25 20:46:02 INFO [o.h.o.r.g.e.c.CoreLMPreparationHandler ] - Loaded landmark splitting collection from
2024-04-25 20:46:03 INFO [o.h.o.r.g.e.ORSGraphHopper ] - version v4.9.1|2024-01-17T09:08:46Z (7,20,5,4,5,7)
2024-04-25 20:46:03 INFO [o.h.o.r.g.e.ORSGraphHopper ] - graph CH|car_ors|RAM_STORE|3D|turn_cost|,,,,, details:edges:0(0MB), nodes:0(0MB), name:(0MB), geo:0(0MB), bounds:1.7976931348623157E308,-1.7976931348623157E308,1.7976931348623157E308,-1.7976931348623157E308,1.7976931348623157E308,-1.7976931348623157E308, shortcuts:0 (0MB), nodesCH:-1 (0MB), shortcuts:0 (0MB), nodesCH:-1 (0MB), shortcuts:0 (0MB), nodesCH:-1 (0MB)
2024-04-25 20:46:03 INFO [o.h.o.r.g.e.ORSGraphHopper ] - No custom areas are used, custom_areas.directory not given
2024-04-25 20:46:03 INFO [o.h.o.r.g.e.ORSGraphHopper ] - start creating graph from /home/ors/files/example-heidelberg.osm.gz
2024-04-25 20:46:03 INFO [o.h.o.r.g.e.ORSGraphHopper ] - using CH|car_ors|RAM_STORE|3D|turn_cost|,,,,, memory:totalMB:1024, usedMB:294
2024-04-25 20:46:03 INFO [o.h.o.a.Application ] - Started Application in 5.915 seconds (process running for 7.848)
2024-04-25 20:46:03 INFO [o.h.o.a.Application ] - openrouteservice {"build_date":"2024-03-21T13:55:54Z","version":"8.0.0"}
2024-04-25 20:51:00 INFO [o.h.o.a.s.l.ORSInitContextListener ] - Shutting down openrouteservice {"build_date":"2024-03-21T13:55:54Z","version":"8.0.0"} and releasing resources.
2024-04-25 20:51:27 INFO [o.h.o.a.Application ] - Starting Application v8.0.0 using Java 21.0.2 with PID 1 (/ors.jar started by root in /home/ors)
2024-04-25 20:51:27 INFO [o.h.o.a.Application ] - The following 1 profile is active: "default"
2024-04-25 20:51:27 INFO [o.h.o.a.ORSEnvironmentPostProcessor ] -
2024-04-25 20:51:27 INFO [o.h.o.a.ORSEnvironmentPostProcessor ] - Configuration lookup started.
2024-04-25 20:51:27 INFO [o.h.o.a.ORSEnvironmentPostProcessor ] - Configuration file set by environment variable.
2024-04-25 20:51:27 INFO [o.h.o.a.ORSEnvironmentPostProcessor ] - Loaded file '/home/ors/config/ors-config.yml'
2024-04-25 20:51:27 INFO [o.h.o.a.ORSEnvironmentPostProcessor ] - Configuration lookup finished.
Of course the health status is “not ready”.
I’m aware that using Docker Compose with a virtual machine might be simpler, but I’m trying to avoid this configuration since the demands of my use case are not very high.
Or maybe I might have misunderstood some fundamental aspects, but I’m not sure what.
Thanks you for your help