Shared tables mode experiences significant worker queue contention during
parallel test execution. 120s was insufficient for attribute processing
in Shared V1 mode and occasionally for PostgreSQL under load.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix float/int/bool attribute creation failures caused by type loss
during queue serialization in the Databases worker
- Rework session limit test to retry session creation for cache
propagation in shared mode
- Increase GraphQL attribute polling timeouts from 30s to 60s
- Increase SchemaPolling waitForAllAttributes timeout to 180s
- Increase Realtime WebSocket client timeout from 30s to 45s
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Throw Critical exception when attribute/index/deployment status is
'failed' to fail tests immediately instead of waiting 15-20 minutes.
Increase default schema polling timeouts from 15 to 20 minutes for
CI stability under parallel load.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Increase SchemaPolling default timeouts from 8min to 15min for
waitForAttribute, waitForAttributes, waitForAttributeCount, and
waitForIndex to match waitForAllAttributes/waitForAllIndexes. Under
parallel test load the database worker gets backlogged, causing
attributes to stay in 'processing' state longer than 8 minutes.
- Add isEmpty check in Functions Delete before updating schedules to
prevent 500 errors when the schedule document is missing (consistent
with other modules like Executions/Delete).
- Add retry logic to cleanupFunction in FunctionsBase for transient
MongoDB transaction abort errors during function deletion.
- Increase Realtime attribute/index polling timeouts from 2min to 15min
to handle worker backlog under CI load.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Spatial attribute creation under parallel CI load consistently exceeds
10 minutes. Increased both waitForAllAttributes and waitForAllIndexes
from 10/8 minutes to 15 minutes to accommodate heavy parallel load.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Spatial attribute creation under parallel MariaDB load and function
deployment builds under MongoDB can exceed previous limits. Increased
SchemaPolling timeout from 8 to 10 minutes and GraphQL function build
timeout from 2 to 4 minutes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
MariaDB attribute/index processing consistently times out under parallel
test load in CI with different tests failing each run. Increase all default
polling timeouts from 240s to 480s to handle worst-case MariaDB load.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
MariaDB attribute/index processing consistently times out under parallel
test load in CI. Increase all default polling timeouts from 240s to 360s.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
120s was insufficient for attribute processing under parallel CI load,
causing spurious timeout failures in spatial, transaction, and attribute tests.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Fix testCreateAttributes: use $actorsAttr instead of undefined $actors variable,
and use $this->getSchemaResource() instead of hardcoded 'attributes' for TablesDB compat
- Fix testUpsertDocument: add missing 'data' => $upsertData in PUT request body
- Increase SchemaPolling default timeout from 60s to 120s for CI stability with
spatial attributes under parallel load
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Under parallel CI load with MariaDB, attribute processing frequently
exceeds the previous 5-minute default timeout. Increased all polling
defaults (waitForAttribute, waitForAttributes, waitForAttributeCount,
waitForAllAttributes, waitForIndex, waitForAllIndexes) to 10 minutes.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Under parallel load in CI, attribute/index workers can take longer than
3 minutes to process. Increase all polling timeouts to 5 minutes to
reduce flaky failures from worker queue congestion.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- DatabasesBase: Wait for twoWayKey attribute on library collection in
setupOneToManyRelationship; use server API key headers for document creation
- DatabasesStringTypesTest: Add delays between attribute creation batches to
avoid overwhelming the database worker
- SchemaPolling: Improve waitForAllAttributes error messaging
- GraphQL UsersTest: Use fresh user in testDeleteUserSession to avoid
session conflicts
- Realtime: Increase WebSocket timeout to 120s and index polling to 60s
- Account: Add better error reporting for phone session creation failures
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The root cause of flaky attribute processing tests is that a single
appwrite-worker-databases container cannot keep up with attribute
creation jobs when tests run in parallel via paratest.
This fix:
- Scales the databases worker to 4 instances during E2E service tests
- Reverts timeout increases that masked the underlying issue
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Increase all default timeouts in SchemaPolling trait from 5 minutes to
10 minutes. With tests running in parallel via paratest, the attribute
processing worker can become backlogged under heavy CI load, requiring
longer wait times for attributes to transition from 'processing' to
'available' status.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Increase all default timeouts in SchemaPolling trait from 3 minutes to
5 minutes to handle slower attribute processing when running tests in
parallel under heavy CI load.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Increase all default timeouts in SchemaPolling trait from 2 minutes to
3 minutes to handle slower attribute processing in CI environments.
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
- Increase SchemaPolling default timeout from 60s to 120s
- Replace hardcoded sleep() with assertEventually() for message status polling in Messaging tests
Jake Barnby