Repository navigation
Add Job Queue reliability and scheduling guidance - #148
Conversation
Jesper Schulz-Wedde (JesperSchulz)
left a comment
There was a problem hiding this comment.
The overall set is useful, but four boundaries need correction before these become agent rules.
-
The scheduled-task good sample exposes IsolatedStorage through a public procedure.
store-scheduled-task-id-to-avoid-duplicate-tasks.good.aldeclaresEnsureCleanupTask()withoutlocal/internalwhile it reads and writesIsolatedStorage. That contradictsmicrosoft/knowledge/security/isolatedstorage-access-must-be-local-or-internal.mdand lets dependent extensions invoke the storage-backed scheduling operation. Make itinternal(or put the storage access behind an internal helper if a deliberately public intent API is required). -
The cooperative-cancellation good sample processes one batch before checking an existing stop request. Its
repeat ... until IsStopRequested(...)always executes once. Because the request is persisted, a later recurring run also processes one extra batch before exiting. Check before each unit, including the first (for example, awhile not IsStopRequested(...) doloop), and state who clears the request when scheduling is resumed. -
Category serialization is company-scoped, not a general mutex. Job queue dispatch and the Waiting state coordinate entries in the same company/category. The article should explicitly say this does not serialize same-named categories across companies, environments, or non-job-queue workers; shared external/cross-company resources need another lock. Otherwise an agent can recommend a category that does not actually protect the stated resource.
-
Job queue work is not unconditionally retried.
job-queue-external-effects-must-be-idempotent.mdsays Business Central “retries the queued work,” but automatic rescheduling depends on recurrence and configured attempt limits; a failed entry can remain in Error and later be restarted manually. Phrase this as work that can be retried/rescheduled or rerun after rollback. The idempotency requirement still follows, without promising a retry on every failure.
The no-UI, failure-propagation, idempotency mechanism, On Hold semantics, and stored-task-ID approach are otherwise sound. All repository validators pass.
…bqueue # Conflicts: # microsoft/knowledge/performance/job-queue-category-code-serializes-conflicting-jobs.bad.al # microsoft/knowledge/performance/job-queue-category-code-serializes-conflicting-jobs.good.al # microsoft/knowledge/performance/job-queue-category-code-serializes-conflicting-jobs.md # microsoft/knowledge/performance/job-queue-external-effects-must-be-idempotent.bad.al # microsoft/knowledge/performance/job-queue-external-effects-must-be-idempotent.good.al # microsoft/knowledge/performance/job-queue-external-effects-must-be-idempotent.md # microsoft/knowledge/performance/job-queue-handlers-must-not-require-ui.bad.al # microsoft/knowledge/performance/job-queue-handlers-must-not-require-ui.good.al # microsoft/knowledge/performance/job-queue-handlers-must-not-require-ui.md # microsoft/knowledge/performance/job-queue-handlers-must-propagate-failures.bad.al # microsoft/knowledge/performance/job-queue-handlers-must-propagate-failures.good.al # microsoft/knowledge/performance/job-queue-handlers-must-propagate-failures.md # microsoft/knowledge/performance/job-queue-on-hold-does-not-stop-running-work.bad.al # microsoft/knowledge/performance/job-queue-on-hold-does-not-stop-running-work.good.al # microsoft/knowledge/performance/job-queue-on-hold-does-not-stop-running-work.md # microsoft/knowledge/performance/store-scheduled-task-id-to-avoid-duplicate-tasks.bad.al # microsoft/knowledge/performance/store-scheduled-task-id-to-avoid-duplicate-tasks.good.al # microsoft/knowledge/performance/store-scheduled-task-id-to-avoid-duplicate-tasks.md
Jesper Schulz-Wedde (JesperSchulz)
left a comment
There was a problem hiding this comment.
The four concerns from the previous round are resolved, and all checks pass at this head. One fixture issue remains inline.
There is also an integration blocker outside the changed files: the six new articles are indexed, but microsoft/skills/review/al-performance-review.md has no deterministic Worklist route to them. Its token inventory and targeted cues omit the relevant Job Queue Entry/category, job-handler UI, external-effect, handled-failure, persisted-cancellation, and TaskScheduler.CreateTask/TaskExists signals; the performance fixture override still exercises only use-isempty-for-existence-check. Please update the owning skill with explicit selection cues and appropriate contextual exclusions, and add bad/good routing regressions for these rules. Otherwise all current validators can pass while agents never open or apply the new guidance.
The underlying guidance is valuable, but the executable sample and deterministic review routing need to be complete before these become authoritative agent rules.
Jesper Schulz-Wedde (JesperSchulz)
left a comment
There was a problem hiding this comment.
The previous guidance, clean-company category fixture, and deterministic Worklist-route concerns are resolved. The practical exact-tree validators pass, and I found no new AL or platform-behavior issue.
One regression-contract issue remains inline. It is limited to the category case, but the added evaluation currently expects a finding after removing the only evidence needed by the rule's own contextual guard. Please preserve that case-specific context, then this should be ready for approval.
Jesper Schulz-Wedde (JesperSchulz)
left a comment
There was a problem hiding this comment.
The category regression now preserves explicit shared-exclusive-resource evidence after fixture neutralization, while the good control correctly applies one shared non-empty category. All prior guidance, routing, evaluation, and AL/API concerns remain resolved, and the practical exact-tree validators pass. Approving.
8c26ba4
into
microsoft:main
…inks XmlPort has no RunModal method (static or instance) - Microsoft Learn confirms only Xmlport.Run(Integer [, Boolean RequestWindow] [, Boolean] [, var Record]). Replaces the invented API with the real one throughout the article, worklist cue, and token list, and explains the platform error message's own "XmlPort.RunModal" wording as the same kind of legacy phrasing already noted for Form.RunModal/RequestForm. Also corrects UseRequestPage(false): that's a Report instance method only, not applicable to XMLport, which uses the UseRequestPage = false object property or Run's RequestWindow argument instead. Fixes the two sample references to use the READ-convention markdown link form (Test-KnowledgeIndex.ps1's Knowledge-Retrieval.ps1 check was failing on plain backtick text). Rebased onto upstream/main to resolve conflicts with microsoft#148's job-queue additions to al-performance-review.md - both sets of worklist tokens and cues are retained.
- dimension-management-wiring: SaveDefaultDim's third argument is the shortcut dimension number (1-8), not the field's AL field ID; the fixture passed FieldNo(...) = 10. GetDefaultDimID's InheritFromDimSetID must be 0 when recomputing after the linking record changes, not the document's existing Dimension Set ID (which would retain the previous customer's leftover dimensions). Verified against DimensionManagement.Codeunit.al and BankDepositHeader.Table.al in the BCApps reference clone. - check-post-line-batch-pattern: "Post Line writes exactly one line to the ledger" overclaimed - Gen. Jnl.-Post Line alone calls InsertGLEntry from a dozen call sites (balancing entry, VAT, currency rounding, deferrals) and can write several G/L Entries per journal line. Reworded to "posts exactly one journal line" and softened the "distinct, non-overlapping responsibilities" absolute. - namespace-must-be-verified-from-source.bad.al: dropped the "resolves in a local build, fails in VS Code" comment (taught an inherent compiler/language-server disagreement that isn't real); reframed as stale/cached symbols, matching the prose fix already made. - file-datatype-saas.good.al: replaced the deprecated 5-argument UploadIntoStream overload with the current 2-argument one, and actually staged through TempBlob as the article's own Best Practice instructs (the declared TempBlob variable was previously unused). - test-data-must-be-random-and-complete: no longer treats a short-but-valid value as defective merely for being "underfilled" - AL field lengths are maxima, not minimums. Scoped to missing values or a scenario with an explicit length/format requirement (e.g. a truncation test). Updated the al-testing-review.md routing cue to match. - stored-derived-fields-must-not-be-exposed-directly: stopped mandating source-field exposure as part of the core pattern: the good fixture exposed only one of the derived value's two inputs (Hours Used, not Budgeted Hours), making the claimed "so the consumer can verify it" impossible. Reframed as an optional, all-or-nothing addition and fixed the fixture to expose both inputs. Rebased onto upstream/main to resolve conflicts in al-breaking-changes-review.md, al-data-modeling-review.md, al-performance-review.md, and al-style-review.md against merged PRs microsoft#148 and microsoft#153; all sides' worklist tokens/cues retained.
…aking-changes, performance, testing (#156) * Add 18 community AL/BC patterns across style, data-modeling, web-services, appsource, breaking-changes, performance, and testing Contributed by CURABIS ApS, generalized from patterns observed across real AppSource/PTE development. Each article follows the knowledge file format (frontmatter, Description/Best Practice/Anti Pattern, sibling .good.al/.bad.al samples). * Address Jesper Schulz-Wedde's review on PR #156 - Rename 3 articles so their .good.al/.bad.al companion stems match (do-not-change-primary-key, testfield-required-setup-field, al-identifiers-english), fixing the R14 orphan-sample errors. - do-not-change-primary-key.good.al: include Flow in the new table's own primary key so it actually models the discriminating dimension. - al-build-output-must-not-pollute-project-root.md: drop the unsubstantiated AL0197 causal claim and the non-existent al.outputPath setting; reframe as build-artifact hygiene sourced from ALTool --outfolder / al_build outputPath. - prefer-email-module.md: Email Message is Codeunit 8904, not a table; distinguish it from the underlying Sent/Outbox/Draft storage. - file-datatype-saas.md: File.Open/Create/Read/Write fails to compile against a Cloud-scoped project, it does not compile and silently fail at runtime. - namespace-must-be-verified-from-source.md: narrow to "resolve from the referenced object's source or symbols," since source-file line one is not the only authoritative source (symbol packages, comments before the namespace line). - test-data-must-be-random-and-complete.md: drop "assume an empty database" and "collision-free" absolutes; reframe around independence from unrelated business records and reserving explicit values for scenario-defining inputs. - binary-choice-must-be-boolean.md: scope to genuine true/false semantics, not mechanical two-member-enum-to-boolean conversion. - document-report-word-layout.md: scope down to a sourced Microsoft Learn recommendation instead of an unconditional performance guarantee; cite the three Learn pages. - Wire the new articles into their review skills' candidate-selection signals (file-datatype-saas, prefer-email-module, namespace-must-be-verified-from-source, var-parameters-require-an- addressable-variable) so they can actually enter a worklist. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * Fix dimension-management-wiring.md: ValidateShortcutDimCode and CreateDim do not exist on the current DimensionManagement codeunit Verified against microsoft/BCApps: the real master-table validation procedure is ValidateDimValueCode (or ValidateShortcutDimValues when a DimSetID is also needed), and the real document-side inheritance procedure is GetDefaultDimID, not CreateDim. Caught from Jesper Schulz-Wedde's review thread, which had been partially hidden by GitHub's comment folding. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * Address second round of Jesper Schulz-Wedde's review on PR #156 - dimension-management-wiring.md/.good.al: split into the two distinct models the article was conflating - master data (Default Dimension records via ValidateDimValueCode/SaveDefaultDim) vs. transactional/ document data (a single Dimension Set ID assembled via AddDimSource + GetDefaultDimID, verified against BCApps' ExchRateAdjmtProcess.Codeunit.al). Added a compiling document-table example alongside the existing master table one. - Deleted api-page-flowfields-must-be-calcfields (.md/.good.al/.bad.al): Microsoft's own FlowFields documentation states a FlowField used as a control's direct source expression is automatically calculated on any page - no API-page exception is documented, and none could be reproduced. - prefer-email-module.bad.al/.md: Codeunit Mail has no Send/GetErrorDesc members; fixed to the real current 7-argument CreateMessage signature, and corrected the claim that the legacy path "still runs" - its base implementation no longer sends anything, only raises integration events. - check-post-line-batch-pattern.md/.good.al: reframed from a universal invariant to the standard shape, naming the real Gen./Item/CA/Res./Job/ Insurance/Mfg. Item/FA Jnl.-Check Line/-Post Line/-Post Batch codeunits it's based on. Added the missing Check Line companion codeunit so the good fixture is internally complete. - test-data-must-be-random-and-complete.good.al: removed leftover "collision-free" wording contradicting the already-corrected article text. - fixed-choice-set-must-use-enum-not-integer.md: removed the reintroduced state-count heuristic ("the line is the state count"), aligned with binary-choice-must-be-boolean.md's semantics-based distinction. - namespace-must-be-verified-from-source.md: removed the false claim that the compiler and AL Language Server use different namespace-resolution rules. - intrinsic-al-functions-must-use-modern-casing.md: removed the unverified claim that PascalCase is the VS Code formatter's default output. Worklist completeness: added cues for the 8 rules in data-modeling, testing, performance, and web-services that had none (Jesper's explicit ask), plus the same gap in all 7 style rules from this PR (not explicitly named this round, but the identical systemic issue) - 15 cues total across al-data-modeling-review.md, al-testing-review.md, al-performance-review.md, al-web-services-review.md, and al-style-review.md. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> * Fix remaining correctness issues from Jesper's 2026-09-15 re-review - dimension-management-wiring: SaveDefaultDim's third argument is the shortcut dimension number (1-8), not the field's AL field ID; the fixture passed FieldNo(...) = 10. GetDefaultDimID's InheritFromDimSetID must be 0 when recomputing after the linking record changes, not the document's existing Dimension Set ID (which would retain the previous customer's leftover dimensions). Verified against DimensionManagement.Codeunit.al and BankDepositHeader.Table.al in the BCApps reference clone. - check-post-line-batch-pattern: "Post Line writes exactly one line to the ledger" overclaimed - Gen. Jnl.-Post Line alone calls InsertGLEntry from a dozen call sites (balancing entry, VAT, currency rounding, deferrals) and can write several G/L Entries per journal line. Reworded to "posts exactly one journal line" and softened the "distinct, non-overlapping responsibilities" absolute. - namespace-must-be-verified-from-source.bad.al: dropped the "resolves in a local build, fails in VS Code" comment (taught an inherent compiler/language-server disagreement that isn't real); reframed as stale/cached symbols, matching the prose fix already made. - file-datatype-saas.good.al: replaced the deprecated 5-argument UploadIntoStream overload with the current 2-argument one, and actually staged through TempBlob as the article's own Best Practice instructs (the declared TempBlob variable was previously unused). - test-data-must-be-random-and-complete: no longer treats a short-but-valid value as defective merely for being "underfilled" - AL field lengths are maxima, not minimums. Scoped to missing values or a scenario with an explicit length/format requirement (e.g. a truncation test). Updated the al-testing-review.md routing cue to match. - stored-derived-fields-must-not-be-exposed-directly: stopped mandating source-field exposure as part of the core pattern: the good fixture exposed only one of the derived value's two inputs (Hours Used, not Budgeted Hours), making the claimed "so the consumer can verify it" impossible. Reframed as an optional, all-or-nothing addition and fixed the fixture to expose both inputs. Rebased onto upstream/main to resolve conflicts in al-breaking-changes-review.md, al-data-modeling-review.md, al-performance-review.md, and al-style-review.md against merged PRs #148 and #153; all sides' worklist tokens/cues retained. * Fix remaining READ-convention sample links across this PR's 18 articles The same plain-backtick "See sample: \`x.good.al\`." form fixed on al-methods-limited-during-write-transactions (PR #161) turned up repo-wide on 15 more of this PR's articles - Knowledge-Retrieval.ps1 requires the markdown-link form to associate a sample with its article. All 16 fixed; the four local validators (frontmatter, knowledge-index, knowledge-retrieval, review-fixtures, skill-index) pass. * Fix two merge-critical correctness issues from Jesper's 2026-09-22 review - api-page-key-fields-must-be-editable-on-insert.good.al and stored-derived-fields-must-not-be-exposed-directly.good.al: both were writable API pages missing DelayedInsert = true, contradicting this repo's own api-page-delayedinsert-true rule - the canonical "good" samples were teaching code BCQuality itself flags. - dimension-management-wiring.good.al: UpdateDimensionSetID exited early when Customer.Get failed, leaving the previous customer's shortcut dimension and Dimension Set ID in place - the same staleness bug the InheritFromDimSetID = 0 fix (from the prior review round) was meant to prevent, just triggered by a failed lookup instead of a successful one. Now clears the shortcut field and recomputes with an empty source list on a failed lookup too, so GetDefaultDimID correctly returns an empty Dimension Set ID instead of never running. --------- Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
…#161) * Add runmodal-is-not-allowed-inside-write-transactions.md; drop "modal page" from the prompts article A modal page does not behave like Confirm/StrMenu inside a write transaction: the platform refuses Page.RunModal (and Report/XmlPort .RunModal with a request page, and Codeunit.Run with its return value used) with a runtime error instead of holding the lock. The new article documents that guard - verified against Microsoft Learn (Codeunit.Run transaction semantics), microsoft/AL#5452, Microsoft's own Base Application (Commit(); Page.RunModal pattern), and a live reproduction on Business Central 26 quoted verbatim. avoid-user-prompts-inside- transactions.md keeps its scope to the prompts the platform does allow; "modal page" is removed from its list because that case is refused, not stalled. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * Do not list Message among the lock-holding prompts Message runs asynchronously - it is queued and shown when the calling method ends or another method requests input - so it never pauses the transaction and holds no lock. Only Confirm and StrMenu wait for the user. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * State the four write-transaction conditions as one explicit line per method Mirrors the structure of the platform's own error message so the Report.RunModal and XmlPort.RunModal request-page exceptions are visible at a glance instead of buried in prose. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * Title the article after the platform error it explains "AL methods limited during write transactions: commit before RunModal and Codeunit.Run" - so a developer or agent searching for the runtime error text lands on the one article that covers all four restricted methods. Slug and sample stems renamed to match; keywords gain the error's own phrase and the legacy Form.RunModal name it still uses. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * Trim keywords to 10 per validator guidance * Wire the new article into al-performance-review.md Adds Page.RunModal / Report.RunModal / XmlPort.RunModal / UseRequestPage to the extracted-token list and one deterministic worklist cue with exclusions, so the article is selected from the RunModal call itself rather than only via a co-located Commit/Modify token. Codeunit.Run in that position is routed to its existing owner article. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> * Correct XmlPort.RunModal to Xmlport.Run; fix READ-convention sample links XmlPort has no RunModal method (static or instance) - Microsoft Learn confirms only Xmlport.Run(Integer [, Boolean RequestWindow] [, Boolean] [, var Record]). Replaces the invented API with the real one throughout the article, worklist cue, and token list, and explains the platform error message's own "XmlPort.RunModal" wording as the same kind of legacy phrasing already noted for Form.RunModal/RequestForm. Also corrects UseRequestPage(false): that's a Report instance method only, not applicable to XMLport, which uses the UseRequestPage = false object property or Run's RequestWindow argument instead. Fixes the two sample references to use the READ-convention markdown link form (Test-KnowledgeIndex.ps1's Knowledge-Retrieval.ps1 check was failing on plain backtick text). Rebased onto upstream/main to resolve conflicts with #148's job-queue additions to al-performance-review.md - both sets of worklist tokens and cues are retained. * Add Report.Run alongside Report.RunModal to the write-transaction guard routing al-performance-review.md's tokens/cues only recognized Report.RunModal, so a failing Modify(); Report.Run(..., true, ...) path was never worklisted even though Report.Run shares the exact same RequestWindow- blocking-dialog mechanism as Report.RunModal (they differ only in whether the report instance is cleared afterward). Added Report.Run to the token list and the targeted cue, with the same request-page- suppressed exclusion, and extended the knowledge article's Description bullet to name both methods explicitly instead of only RunModal. --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> Co-authored-by: Jesper Schulz-Wedde <jesper.schulzwedde@microsoft.com> Co-authored-by: Copilot App <223556219+Copilot@users.noreply.github.com>
Summary
Add six community performance rules that document reliability and scheduling behaviors for Business Central Job Queue and Task Scheduler workloads. Each rule includes focused good and bad AL examples so reviewers can distinguish unsafe code from intentional behavior.
What changed
SystemIdasidempotencyKeyin a JSON POST body and documents that the external service must atomically create only when the key is absent, returning the existing record on retries.TaskScheduler.TaskExistsbefore creating replacements.Why
These behaviors depend on platform contracts that are easy to misread from otherwise valid AL code. Making the retry, background-session, concurrency, cancellation, and task-identity requirements explicit helps reviews identify real operational risks without flagging safe implementations.
Validation
Build-KnowledgeIndex.ps1: 278 articles indexedTest-ReviewFixtures.ps1: 32 cases across 16 leaf domains passedgit diff --check: passed