Data and migrations release incidents

Data and migrations: 7 published incident types backed by 23 public reports.

The root-cause stage named most often in this group is release-process, in 4 of 7 incident types.

7 incident types published · 2 withheld · 23 public reports

Incident types in this group

Data and migrations


Encoding or timezone shift	Likelihood: not published (1 public report in 5 years)	Impact: high (2 of 2 reports: data)	2
Irreversible migration fails midway	Likelihood: low (2 public reports in 5 years)	Impact: medium (2 of 4 reports: availability)	4
Long-running migration locks table	Likelihood: low (2 public reports in 5 years)	Impact: medium (3 of 3 reports: availability)	3
Missing index after migration	Likelihood: medium (3 public reports in 5 years)	Impact: medium (4 of 5 reports: availability)	5
ORM upgrade changes query semantics	Likelihood: low (2 public reports in 5 years)	Impact: high (3 of 3 reports: data)	3
Schema drift between services	Likelihood: medium (4 public reports in 5 years)	Impact: medium (2 of 4 reports: availability)	4
Soft-delete flag ignored	Likelihood: low (2 public reports in 5 years)	Impact: high (1 of 2 reports: data)	2

Withheld in this group: Data backfill overwrites records, Nullable column breaks consumer. Reasons per route

Risk grid

Risk grid: count of incident types per likelihood and impact cell, from data/risk.json
Likelihood \ Impact	Impact: high	Impact: medium	Impact: low
Likelihood: high	0	0	0
Likelihood: medium	0	2	0
Likelihood: low	2	2	0
Impact unrated or likelihood not graded	1

Table equivalent: incidents per cell

Incident	Likelihood	Impact
Encoding or timezone shift	not-published	high
Irreversible migration fails midway	low	medium
Long-running migration locks table	low	medium
Missing index after migration	medium	medium
ORM upgrade changes query semantics	low	high
Schema drift between services	medium	medium
Soft-delete flag ignored	low	high

Root-cause stages

Stage	Incident types naming it
release-process	4
migration	3
code	2
third-party	2
config	1

Median time to detection across the group: 8 minutes from 6 reports that state both the change time and the detection time (values above the site-wide 95th percentile excluded).

Only on this page

Irreversible migration fails midway is the incident type in this group whose root causes do not include the release-process stage; its causes are introduced at migration.

Group checklist

The first pre-release item of each published incident type in the group.

GR-01Encoding or timezone shift: Assert in every environment that the application timezone configuration variable is set to a non-NULL value before the serving change goes live. config
GR-02Irreversible migration fails midway: Run each migration against a production-sized database copy and list every step that executes outside a transaction. migration
GR-03Long-running migration locks table: Run the migration against a copy of the production table at full row count and record whether the engine performs the change in place or copies the table. migration
GR-04Missing index after migration: List every DROP INDEX in the migration and confirm a valid replacement index exists before the drop runs. migration
GR-05ORM upgrade changes query semantics: Capture the SQL generated for every scoped association query before and after the ORM upgrade and diff the WHERE clauses. integration
GR-06Schema drift between services: Run consumer contract tests for each changed internal API against the provider version currently in production, not only the one in the same commit. contract
GR-07Soft-delete flag ignored: Seed one soft-deleted row in a test database and confirm that the relationship query, getQuery() and toBase() paths all exclude it. integration

Questions

What causes Data and migrations incidents?

Published incident types in this group name these root-cause stages: release-process (4), migration (3), code (2), third-party (2), config (1).

Which pipeline stage introduces most Data and migrations incidents?

The release-process stage, named in 4 of 7 published incident types in the group.

Which test classes catch Data and migrations incidents?

Test classes listed on the group's incident pages: integration tests (3), migration tests (3), load and performance tests (3), configuration and environment tests (2), canary and progressive delivery checks (2), synthetic monitoring (1), chaos and resilience tests (1), contract tests (1), unit tests (1).

How quickly are Data and migrations incidents detected in public reports?

The median time to detection is 8 minutes across 6 reports that state both the change time and the detection time.

How likely are Data and migrations incidents compared with each other?

Encoding or timezone shift: not-published likelihood, high impact; Irreversible migration fails midway: low likelihood, medium impact; Long-running migration locks table: low likelihood, medium impact; Missing index after migration: medium likelihood, medium impact; ORM upgrade changes query semantics: low likelihood, high impact; Schema drift between services: medium likelihood, medium impact; Soft-delete flag ignored: low likelihood, high impact.

What should a release checklist for Data and migrations include?

The group checklist collects the first pre-release item of each of the 7 published incident types; each item names the test class that checks it.