You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Some log messages have an incorrect level, unclear message, or missing. Due to the huge amount of logs, we should keep in mind that many node operators have logging level set to warn or even error.
ISSUE TYPE
Feature Idea
COMPONENT NAME
runtime
worker/registration
ADDITIONAL INFORMATION
If anything is wrong with the runtime/enclave, it should be an have "level":"error". Currently:
Failed node registrations should be logged with "level":"error". Currently we only log when registration is scheduled and when it succeeds, but not when it fails:
{"caller":"worker.go:373","epoch_height":26221881,"level":"info","module":"worker/registration","msg":"per-epoch re-registration scheduled","target_height":26221896,"ts":"2025-04-19T17:10:57.260326573Z"}
{"caller":"worker.go:835","epoch":43656,"level":"info","module":"worker/registration","msg":"performing node (re-)registration","node_id":"...","ts":"2025-04-18T11:11:12.356493166Z"}
{"caller":"worker.go:951","level":"info","module":"worker/registration","msg":"node registered with the registry","ts":"2025-04-18T11:09:51.232704473Z"}
It is unclear as what the node registered as on success. Which mode/role (validator/compute/observer), for which runtime, and for which entity? Maybe even what it failed to register as?
{"caller":"worker.go:951","level":"info","module":"worker/registration","msg":"node registered with the registry","ts":"2025-04-18T11:09:51.232704473Z"}
The registration process is full of unclear log messages (for a typical node operator). It mentions "nil hook" and "not registering" as if something is wrong, but then either logs a successful registration (or quietly retries in case of a failed node registration).
{"caller":"worker.go:1127","level":"info","module":"worker/registration","msg":"starting node registration service","ts":"2025-04-18T11:19:46.767946795Z"}
{"caller":"worker.go:230","level":"debug","module":"worker/registration","msg":"waiting for consensus sync","ts":"2025-04-18T11:19:46.767967779Z"}
{"caller":"worker.go:951","level":"info","module":"worker/registration","msg":"node registered with the registry","ts":"2025-04-18T11:19:57.851673391Z"}
{"caller":"worker.go:408","level":"debug","module":"worker/registration","msg":"enumerating role provider hooks","ts":"2025-04-18T11:21:10.739051727Z"}
{"caller":"worker.go:426","level":"debug","module":"worker/registration","msg":"nil hook for role","role":"compute","ts":"2025-04-18T11:21:10.739148041Z","ver":3}
{"caller":"worker.go:443","level":"debug","module":"worker/registration","msg":"not registering, no role provider hooks","ts":"2025-04-18T11:21:10.739172249Z"}
{"caller":"worker.go:408","level":"debug","module":"worker/registration","msg":"enumerating role provider hooks","ts":"2025-04-18T11:21:18.206749094Z"}
{"caller":"worker.go:426","level":"debug","module":"worker/registration","msg":"nil hook for role","role":"compute","ts":"2025-04-18T11:21:18.206771224Z","ver":4}
{"caller":"worker.go:443","level":"debug","module":"worker/registration","msg":"not registering, no role provider hooks","ts":"2025-04-18T11:21:18.206776875Z"}
{"caller":"worker.go:408","level":"debug","module":"worker/registration","msg":"enumerating role provider hooks","ts":"2025-04-18T11:21:18.207684376Z"}
{"caller":"worker.go:835","epoch":43656,"level":"info","module":"worker/registration","msg":"performing node (re-)registration","node_id":"...","ts":"2025-04-18T11:21:18.207833678Z"}
The text was updated successfully, but these errors were encountered:
SUMMARY
Some log messages have an incorrect level, unclear message, or missing. Due to the huge amount of logs, we should keep in mind that many node operators have logging level set to
warn
or evenerror
.ISSUE TYPE
COMPONENT NAME
runtime
worker/registration
ADDITIONAL INFORMATION
"level":"error"
. Currently:"level":"error"
. Currently we only log when registration is scheduled and when it succeeds, but not when it fails:validator
/compute
/observer
), for which runtime, and for which entity? Maybe even what it failed to register as?"nil hook"
and"not registering"
as if something is wrong, but then either logs a successful registration (or quietly retries in case of a failed node registration).The text was updated successfully, but these errors were encountered: