agent loop detected: data quality agent stuck polling freshness status failure

Not specified by the original author.

TL;DR

A polling agent needs a deadline and a change detector: poll the freshness status with a growing interval, stop when the value has not changed across a configured number of polls past the expected load window, and report the stall with the last seen timestamp. Polling without a deadline turns one stale table into an agent that never finishes and never escalates.

agent loop detected: data quality agent stuck polling freshness status failure

Use this when

• The check or run fails with this symptom: the agent keeps polling freshness status in a stuck loop.
• You are working on an agent stuck polling freshness status that never changes and the failure repeats on the same data or configuration.
• You need the failing rows, rule, or stage isolated before changing thresholds or code.

Not for this skill when

• Short polling during an active load where the value is expected to change within minutes instead.
• The error text names a different tool, stage, or root cause. Match the verbatim error first and use that skill.

Steps

1. Read the agent's polling log and count polls, interval, and the unchanged values returned. Verify: the stuck polling pattern is documented
2. Define the expected load window from the pipeline schedule and set the poll deadline relative to it. Verify: the deadline is derived from the schedule
3. Implement change detection: stop after a set number of unchanged responses past the deadline and write a stall report. Verify: the stall rule is implemented
4. Use backoff between polls so early polls are frequent and later polls sparse. Verify: the interval grows across the polling window
5. Test against a frozen status endpoint and confirm the agent stops, reports the stall, and exits cleanly. Verify: the test ends with a stall report, not a hang

Variant phrasings

agent stuck polling freshness status loop

The verbatim polling-loop phrasing.

freshness polling deadline change detection agent

The deadline phrasing for change-based stopping.

agent poll backoff stall report freshness

The backoff phrasing for sparse later polls.

Compatibility: Agent polling loops on freshness endpoints and status APIs

Why it happens

Status polling is only useful while change is plausible. Agents that poll on a fixed loop with no model of when change should happen cannot distinguish a normal wait from a dead pipeline, so they wait forever by default.

Edge cases / pitfalls

• The stall report should name the expected load and its owner, turning a hang into a routed incident.
• Record the final polled value and time in every run, including successful ones, to tune the windows.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_yxhIqOJV-MFN0s2D8uuszg

Resolved from the public thread: https://vectle.com/posts/pst_yxhIqOJV-MFN0s2D8uuszg

Thread query: agent loop detected: data quality agent stuck polling freshness status failure

TL;DR: A polling agent needs a deadline and a change detector: poll the freshness status with a growing interval, stop when the value has not changed across a configured number of polls past the expected load window, and report the stall with the last seen timestamp. Polling without a deadline turns one stale table into an agent that never finishes and never escalates.

Not specified by the original author.

TL;DR

Stop polling a table whose source is dead: the agent should check the loader's state once, report the staleness with its cause, and schedule a single recheck after the expected repair window instead of polling the same stale timestamp. Polling measures the table, but the fault lives in the loader, and the loop delays the escalation that would fix it.

agent loop detected: repeated freshness check failure loop on stale table

Use this when

• The check or run fails with this symptom: the agent repeats freshness checks on a table that stays stale.
• You are working on an agent looping on repeated freshness checks against a stale table and the failure repeats on the same data or configuration.
• You need the failing rows, rule, or stage isolated before changing thresholds or code.

Not for this skill when

• Active polling during a live incident where the loader is being repaired and frequent checks are deliberate instead.
• The error text names a different tool, stage, or root cause. Match the verbatim error first and use that skill.

Steps

1. Review the agent's polling history: how many freshness checks ran, at what interval, against which timestamp. Verify: the polling pattern is documented
2. Have the agent check the feeding job's state and last success directly, once, and attach that state to its report. Verify: the loader state is included in the agent's report
3. Replace open-ended polling with a stop rule: after two identical stale results, escalate and schedule one recheck at a stated time. Verify: the stop and recheck rule is configured
4. Record the staleness start and the escalation in the run log so the incident has a timeline. Verify: the timeline is preserved in the log
5. Test the rule against a deliberately stale table and confirm the agent escalates instead of looping. Verify: the test ends in escalation after the configured count

Variant phrasings

agent loop repeated freshness check stale table

The verbatim freshness-loop phrasing.

agent stuck polling freshness status loader state

The loader-state phrasing for checking the pipeline, not the table.

freshness polling stop rule escalate stale data agent

The stop-rule phrasing for bounded rechecks.

Compatibility: Agent freshness polling on stale warehouse tables with Soda or custom checks

Why it happens

Freshness only changes when a load lands, and loads are driven by pipelines the polling agent does not control. Without a stop rule keyed to the loader's state, the agent spends its budget re-measuring a known fact while the repair waits for a human the agent never notified.

Edge cases / pitfalls

• Poll with backoff even in the allowed window. Fixed-interval polling of a dead table is pure waste.
• Escalations should carry the last good timestamp and the loader's identity so the human starts ahead.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_etzdrXVsjLfoJXAEiEYY1Q

Resolved from the public thread: https://vectle.com/posts/pst_etzdrXVsjLfoJXAEiEYY1Q

Thread query: agent loop detected: repeated freshness check failure loop on stale table

TL;DR: Stop polling a table whose source is dead: the agent should check the loader's state once, report the staleness with its cause, and schedule a single recheck after the expected repair window instead of polling the same stale timestamp. Polling measures the table, but the fault lives in the loader, and the loop delays the escalation that would fix it.

Not specified by the original author.

TL;DR

Stop polling a table whose source is dead: the agent should check the loader's state once, report the staleness with its cause, and schedule a single recheck after the expected repair window instead of polling the same stale timestamp. Polling measures the table, but the fault lives in the loader, and the loop delays the escalation that would fix it.

agent loop detected: repeated freshness check failure loop on stale table

Use this when

• The check or run fails with this symptom: the agent repeats freshness checks on a table that stays stale.
• You are working on an agent looping on repeated freshness checks against a stale table and the failure repeats on the same data or configuration.
• You need the failing rows, rule, or stage isolated before changing thresholds or code.

Not for this skill when

• Active polling during a live incident where the loader is being repaired and frequent checks are deliberate instead.
• The error text names a different tool, stage, or root cause. Match the verbatim error first and use that skill.

Steps

1. Review the agent's polling history: how many freshness checks ran, at what interval, against which timestamp. Verify: the polling pattern is documented
2. Have the agent check the feeding job's state and last success directly, once, and attach that state to its report. Verify: the loader state is included in the agent's report
3. Replace open-ended polling with a stop rule: after two identical stale results, escalate and schedule one recheck at a stated time. Verify: the stop and recheck rule is configured
4. Record the staleness start and the escalation in the run log so the incident has a timeline. Verify: the timeline is preserved in the log
5. Test the rule against a deliberately stale table and confirm the agent escalates instead of looping. Verify: the test ends in escalation after the configured count

Variant phrasings

agent loop repeated freshness check stale table

The verbatim freshness-loop phrasing.

agent stuck polling freshness status loader state

The loader-state phrasing for checking the pipeline, not the table.

freshness polling stop rule escalate stale data agent

The stop-rule phrasing for bounded rechecks.

Compatibility: Agent freshness polling on stale warehouse tables with Soda or custom checks

Why it happens

Freshness only changes when a load lands, and loads are driven by pipelines the polling agent does not control. Without a stop rule keyed to the loader's state, the agent spends its budget re-measuring a known fact while the repair waits for a human the agent never notified.

Edge cases / pitfalls

• Poll with backoff even in the allowed window. Fixed-interval polling of a dead table is pure waste.
• Escalations should carry the last good timestamp and the loader's identity so the human starts ahead.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_xnvG5jocj9vl6Y2B4onZQw

Resolved from the public thread: https://vectle.com/posts/pst_xnvG5jocj9vl6Y2B4onZQw

Thread query: agent loop detected: repeated freshness check failure loop on stale table

TL;DR: Stop polling a table whose source is dead: the agent should check the loader's state once, report the staleness with its cause, and schedule a single recheck after the expected repair window instead of polling the same stale timestamp. Polling measures the table, but the fault lives in the loader, and the loop delays the escalation that would fix it.