03:06:58 INFO worker-service pollQueue depth=0 lag=41ms

03:07:02 INFO POST /api/followups/send 200 118ms org=w_4821

03:07:41 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:07:44 ERROR refreshSession failed: invalid_grant (expired access token)

03:07:52 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:03 WARN retry 2/3 followup id=fu_88213 backoff=4s

03:08:15 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:20 INFO trace 8f21c04a captured: 6 spans, 1 error span

03:08:31 ERROR followup batch aborted org=w_4821 sent=0 failed=7

03:09:12 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:09:40 INFO deploy marker: api release 2026-07-19T02:59Z PR #1553

03:10:07 ERROR refreshSession failed: invalid_grant (expired access token)

03:10:26 WARN error rate /api/followups/send 14 in 8m, prior hour: 0

03:11:03 INFO libra: correlating 14 traces against deploy window

03:06:58 INFO worker-service pollQueue depth=0 lag=41ms

03:07:02 INFO POST /api/followups/send 200 118ms org=w_4821

03:07:41 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:07:44 ERROR refreshSession failed: invalid_grant (expired access token)

03:07:52 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:03 WARN retry 2/3 followup id=fu_88213 backoff=4s

03:08:15 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:20 INFO trace 8f21c04a captured: 6 spans, 1 error span

03:08:31 ERROR followup batch aborted org=w_4821 sent=0 failed=7

03:09:12 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:09:40 INFO deploy marker: api release 2026-07-19T02:59Z PR #1553

03:10:07 ERROR refreshSession failed: invalid_grant (expired access token)

03:10:26 WARN error rate /api/followups/send 14 in 8m, prior hour: 0

03:11:03 INFO libra: correlating 14 traces against deploy window

03:08:20 INFO trace 8f21c04a captured: 6 spans, 1 error span

03:08:31 ERROR followup batch aborted org=w_4821 sent=0 failed=7

03:09:12 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:09:40 INFO deploy marker: api release 2026-07-19T02:59Z PR #1553

03:10:07 ERROR refreshSession failed: invalid_grant (expired access token)

03:10:26 WARN error rate /api/followups/send 14 in 8m, prior hour: 0

03:11:03 INFO libra: correlating 14 traces against deploy window

03:06:58 INFO worker-service pollQueue depth=0 lag=41ms

03:07:02 INFO POST /api/followups/send 200 118ms org=w_4821

03:07:41 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:07:44 ERROR refreshSession failed: invalid_grant (expired access token)

03:07:52 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:03 WARN retry 2/3 followup id=fu_88213 backoff=4s

03:08:15 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:20 INFO trace 8f21c04a captured: 6 spans, 1 error span

03:08:31 ERROR followup batch aborted org=w_4821 sent=0 failed=7

03:09:12 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:09:40 INFO deploy marker: api release 2026-07-19T02:59Z PR #1553

03:10:07 ERROR refreshSession failed: invalid_grant (expired access token)

03:10:26 WARN error rate /api/followups/send 14 in 8m, prior hour: 0

03:11:03 INFO libra: correlating 14 traces against deploy window

03:06:58 INFO worker-service pollQueue depth=0 lag=41ms

03:07:02 INFO POST /api/followups/send 200 118ms org=w_4821

03:07:41 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:07:44 ERROR refreshSession failed: invalid_grant (expired access token)

03:07:52 ERROR POST /api/followups/send 500 SalesforceTokenExpired

03:08:03 WARN retry 2/3 followup id=fu_88213 backoff=4s

03:08:15 ERROR POST /api/followups/send 500 SalesforceTokenExpired

deploy · PR #1553 · 02:59

anomaly detected03:07:41

POST /api/followups/send is failing in production.

14

failures

8 min

window

0

prior hour

Nobody paged Libra. It opened this incident on its own.

Libra Issues sign in

watching production nowalert to fix PR, one loopreports in Slack

Production broke at 3:07 AM.You slept. Libra didn't.

By 3:14 it had read the traces, named the deploy that caused it, and opened the fix PR with a regression test. Libra watches your logs, traces, and deploys around the clock. You review and merge in the morning.

01 / 05Detect03:07:41

14 failures in 8 minutes on a route that had zero the hour before. Libra opens the incident itself, before anyone is paged.