I just updated a self-hosted docker-compose install of Artifactory 7.98.12 from 7.104.6. It looks like I didn't notice that the release from 29 January 2025 introduced a new microservice, One Model, and it's dependency on Apollo GraphQL. When One Model is starting, it has a step that starts an Apollo server and waits for it to be happy. Unfortunately, nothing in the rover's (the Apollo server) STDERR shows that it isn't starting. This release is pretty new, so nothing on JFrog even mentions Apollo except the release notes.
Any ideas?
2025-02-07T02:08:30.307Z [jfomr] [INFO ] [297adb1b618817bd] [rover.go:64 ] [main ] [ ] - Running Apollo Rover [rover] [trigger][manual] [trigger_id][initial]
2025-02-07T02:08:30.434Z [jfomr] [INFO ] [297adb1b618817bd] [rover.go:153 ] [main ] [ ] - Rover stdout: <none> [rover] [trigger][manual] [trigger_id][initial]
2025-02-07T02:08:30.434Z [jfomr] [INFO ] [297adb1b618817bd] [rover.go:156 ] [main ] [ ] - Rover stderr:
INFO rover_std::fs: reading /opt/jfrog/artifactory/var/work/onemodel/apollo/machine.txt from disk
at crates/rover-std/src/fs.rs:32
INFO rover::utils::telemetry: Telemetry has been disabled.
at src/utils/telemetry.rs:87
INFO rover_std::fs: reading __rover.yaml from disk
at crates/rover-std/src/fs.rs:32
INFO rover_std::fs: reading jfomr.graphqls from disk
at crates/rover-std/src/fs.rs:32
supergraph config loaded successfully
INFO rover_std::fs: checking existence of parent path in '/opt/jfrog/artifactory/var/work/onemodel/apollo/elv2_license.toml'
at crates/rover-std/src/fs.rs:57
INFO rover_std::fs: writing /var/opt/jfrog/artifactory/work/onemodel/apollo/elv2_license.toml to disk
at crates/rover-std/src/fs.rs:88
composing supergraph with Federation v2.9.3
HINT: [UNUSED_ENUM_TYPE]: Enum type "SortOrderDirection" is defined but unused. It will be included in the supergraph with all the values appearing in any subgraph ("as if" it was only used as an output type).
INFO rover_std::fs: checking existence of parent path in '/var/opt/jfrog/artifactory/work/onemodel/composition-1738894110306-712337169/__supergraph.graphqls'
at crates/rover-std/src/fs.rs:57
INFO rover_std::fs: writing /var/opt/jfrog/artifactory/work/onemodel/composition-1738894110306-712337169/__supergraph.graphqls to disk
at crates/rover-std/src/fs.rs:88
Supergraph Schema was printed to /var/opt/jfrog/artifactory/work/onemodel/composition-1738894110306-712337169/__supergraph.graphqls
[rover] [trigger][manual] [trigger_id][initial]
2025-02-07T02:08:30.436Z [jfomr] [INFO ] [297adb1b618817bd] [worker.go:160 ] [main ] [ ] - Successfully generated supergraph 8b3ac1932aa7cab4cbf055d1295be16180b549d5f6e0c08a0923a14e9f007a40 with Apollo Federation version 2.9.3 [composition] [trigger][manual] [trigger_id][initial]
2025-02-07T02:08:30.447Z [jfomr] [INFO ] [297adb1b618817bd] [worker.go:239 ] [main ] [ ] - Publishing new supergraph [composition] [trigger][manual] [trigger_id][initial]
2025-02-07T02:08:44.611Z [jfomr] [WARN ] [297adb1b618817bd] [health_status.go:105 ] [main ] [ ] - Apollo is now UNHEALTHY because: apollo router: Get "http://localhost:8074/health?ready": dial tcp [::1]:8074: connect: connection refused [system]
2025-02-07T02:08:59.610Z [jfomr] [WARN ] [297adb1b618817bd] [health_status.go:111 ] [main ] [ ] - Apollo is still UNHEALTHY (since 2025-02-07 02:08:44.611445737 +0000 UTC) because: apollo router: Get "http://localhost:8074/health?ready": dial tcp [::1]:8074: connect: connection refused [system]
...
For now you can disable the onemodel service in the system.yaml and do restart of the Artifactory.
onemodel:
enabled: false