chore(perf): optimize model aggregation pipelines and remove dead ones - #41157
Draft
KevLehman wants to merge 5 commits into
Draft
chore(perf): optimize model aggregation pipelines and remove dead ones#41157KevLehman wants to merge 5 commits into
KevLehman wants to merge 5 commits into
Conversation
- LivechatRooms.getAnalyticsBetweenDate/getAnalyticsMetricsBetweenDateWithMessages:
count messages inside the $lookup sub-pipeline instead of $unwind+$group
(conversation-totalizers -39% median; preserves msgs:1 for zero-message rooms)
- Messages.findAllNumberOfTransferredRooms: group by rid before the room $lookup,
one join per distinct room projecting only departmentId (3.18x)
- LivechatInquiry: widen {status} index to {status, department} so
getDistinctQueuedDepartments becomes a covered scan (120k -> 0 docs examined
per queue-worker cycle); allowDiskUse on getCurrentSortedQueueAsync
- Rooms.getSubscribedRoomIdsWithoutE2EKeys: filter subscriptions inside the
$lookup, avoiding full-membership arrays that can overflow the 16MB doc limit
- Sessions: trim session:'$$ROOT' to used fields in hourly engagement pipelines,
add allowDiskUse, fix logintAt index typo
- Users: drop redundant second $group in getTotalOfRegisteredUsersByDate;
replace $unwind+$group regroup with $project+$map in findAgentsWithDepartments
- Remove dead methods: LivechatRooms.findAllNumberOfTransferredRooms,
Rooms.findChannelsByTypesWithNumberOfMessagesBetweenDate (+builder),
Messages.findOneByFederationIdAndUsernameOnReactions,
Sessions.getActiveUsersBetweenDates,
LivechatDepartmentAgents.findAgentsByAgentIdAndBusinessHourId (CE+EE files)
All rewritten pipelines verified to return identical output against seeded data.
Contributor
|
Looks like this PR is not ready to merge, because of the following issues:
Please fix the issues and try again If you have any trouble, please check the PR guidelines |
🦋 Changeset detectedLatest commit: 6630944 The changes in this PR will be included in the next version bump. This PR includes changesets to release 5 packages
Not sure what this means? Click here to learn what changesets are. Click here if you're a maintainer who wants to add another changeset to this PR |
Contributor
WalkthroughThis PR removes unused livechat model APIs and rewrites several MongoDB aggregation pipelines and indexes across livechat, messages, rooms, sessions, and users. It also adds a changeset for patch releases and performance notes. ChangesPerformance refactors and dead code removal
Estimated code review effort: 4 (Complex) | ~60 minutes Suggested labels: Suggested reviewers: 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (1)
packages/models/src/models/LivechatRooms.ts (1)
2076-2076: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick winRemove the new implementation comments.
These comments explain expected legacy behavior, but the project guideline asks to avoid implementation comments; consider encoding this expectation in tests instead. As per coding guidelines, "Avoid code comments in the implementation".
Also applies to: 2133-2133
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@packages/models/src/models/LivechatRooms.ts` at line 2076, Remove the implementation comments from LivechatRooms and related logic, including the note in the rooms aggregation flow and the matching comment near the other referenced location. Keep the behavior unchanged in the relevant methods or query helpers, and if the legacy msgs: 1 expectation needs to be preserved, capture it in tests instead of inline comments. Use the surrounding LivechatRooms aggregation code to locate the commented sections and delete only the comments, not the logic.Source: Coding guidelines
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@packages/models/src/models/LivechatRooms.ts`:
- Line 2076: Remove the implementation comments from LivechatRooms and related
logic, including the note in the rooms aggregation flow and the matching comment
near the other referenced location. Keep the behavior unchanged in the relevant
methods or query helpers, and if the legacy msgs: 1 expectation needs to be
preserved, capture it in tests instead of inline comments. Use the surrounding
LivechatRooms aggregation code to locate the commented sections and delete only
the comments, not the logic.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: 15a2cee0-bdc2-4247-b147-e7ef24c315e7
📒 Files selected for processing (16)
.changeset/model-aggregations-perf.mdapps/meteor/ee/server/models/LivechatDepartmentAgents.tsapps/meteor/ee/server/models/raw/LivechatDepartmentAgents.tsapps/meteor/ee/server/models/startup.tspackages/model-typings/src/models/ILivechatDepartmentAgentsModel.tspackages/model-typings/src/models/ILivechatRoomsModel.tspackages/model-typings/src/models/IMessagesModel.tspackages/model-typings/src/models/IRoomsModel.tspackages/model-typings/src/models/ISessionsModel.tspackages/models/src/models/LivechatDepartmentAgents.tspackages/models/src/models/LivechatInquiry.tspackages/models/src/models/LivechatRooms.tspackages/models/src/models/Messages.tspackages/models/src/models/Rooms.tspackages/models/src/models/Sessions.tspackages/models/src/models/Users.ts
💤 Files with no reviewable changes (9)
- packages/model-typings/src/models/ILivechatRoomsModel.ts
- packages/model-typings/src/models/IMessagesModel.ts
- packages/model-typings/src/models/ILivechatDepartmentAgentsModel.ts
- packages/models/src/models/LivechatDepartmentAgents.ts
- apps/meteor/ee/server/models/LivechatDepartmentAgents.ts
- apps/meteor/ee/server/models/raw/LivechatDepartmentAgents.ts
- packages/model-typings/src/models/ISessionsModel.ts
- packages/model-typings/src/models/IRoomsModel.ts
- apps/meteor/ee/server/models/startup.ts
📜 Review details
⏰ Context from checks skipped due to timeout. (5)
- GitHub Check: 📦 Build Packages
- GitHub Check: cubic · AI code reviewer
- GitHub Check: Hacktron Security Check
- GitHub Check: CodeQL-Build
- GitHub Check: CodeQL-Build
🧰 Additional context used
📓 Path-based instructions (1)
**/*.{ts,tsx,js}
📄 CodeRabbit inference engine (.cursor/rules/playwright.mdc)
**/*.{ts,tsx,js}: Write concise, technical TypeScript/JavaScript with accurate typing in Playwright tests
Avoid code comments in the implementation
Files:
packages/models/src/models/LivechatInquiry.tspackages/models/src/models/Sessions.tspackages/models/src/models/Users.tspackages/models/src/models/Messages.tspackages/models/src/models/Rooms.tspackages/models/src/models/LivechatRooms.ts
🧠 Learnings (4)
📚 Learning: 2026-03-16T21:50:37.589Z
Learnt from: amitb0ra
Repo: RocketChat/Rocket.Chat PR: 39676
File: .changeset/migrate-users-register-openapi.md:3-3
Timestamp: 2026-03-16T21:50:37.589Z
Learning: For changes related to OpenAPI migrations in Rocket.Chat/OpenAPI, when removing endpoint types and validators from rocket.chat/rest-typings (e.g., UserRegisterParamsPOST, /v1/users.register) document this as a minor changeset (not breaking) per RocketChat/Rocket.Chat-Open-API#150 Rule 7. Note that the endpoint type is re-exposed via a module augmentation .d.ts in the consuming package (e.g., packages/web-ui-registration/src/users-register.d.ts). In reviews, ensure the changeset clearly states: this is a non-breaking change, the major version should not be bumped, and the changeset reflects a minor version bump. Do not treat this as a breaking change during OpenAPI migrations.
Applied to files:
.changeset/model-aggregations-perf.md
📚 Learning: 2026-02-26T19:25:44.063Z
Learnt from: gabriellsh
Repo: RocketChat/Rocket.Chat PR: 38778
File: packages/ui-voip/src/providers/useMediaSession.ts:192-192
Timestamp: 2026-02-26T19:25:44.063Z
Learning: In the Rocket.Chat repository, do not reference Biome lint rules in code review feedback. Biome is not used even if biome.json exists; only reference Biome rules if there is explicit, project-wide usage documented. For TypeScript files, review lint implications without Biome guidance unless the project enables Biome rules.
Applied to files:
packages/models/src/models/LivechatInquiry.tspackages/models/src/models/Sessions.tspackages/models/src/models/Users.tspackages/models/src/models/Messages.tspackages/models/src/models/Rooms.tspackages/models/src/models/LivechatRooms.ts
📚 Learning: 2026-02-26T19:25:44.063Z
Learnt from: gabriellsh
Repo: RocketChat/Rocket.Chat PR: 38778
File: packages/ui-voip/src/providers/useMediaSession.ts:192-192
Timestamp: 2026-02-26T19:25:44.063Z
Learning: In this repository (RocketChat/Rocket.Chat), Biome lint rules are not used even if a biome.json exists. When reviewing TypeScript files (e.g., packages/ui-voip/src/providers/useMediaSession.ts), ensure lint suggestions do not reference Biome-specific rules. Rely on general ESLint/TypeScript lint rules and project conventions instead.
Applied to files:
packages/models/src/models/LivechatInquiry.tspackages/models/src/models/Sessions.tspackages/models/src/models/Users.tspackages/models/src/models/Messages.tspackages/models/src/models/Rooms.tspackages/models/src/models/LivechatRooms.ts
📚 Learning: 2026-05-06T12:21:44.083Z
Learnt from: juliajforesti
Repo: RocketChat/Rocket.Chat PR: 40256
File: apps/meteor/client/components/CreateDiscussion/CreateDiscussion.tsx:121-149
Timestamp: 2026-05-06T12:21:44.083Z
Learning: Field wrappers in rocket.chat/fuselage-forms (Field, FieldLabel, FieldRow, FieldError, FieldHint) auto-create htmlFor/id associations, aria-describedby, and role="alert" for errors. Do not manually set htmlFor, id, aria-describedby, or role attributes when using these wrappers. This automatic wiring does not apply to plain rocket.chat/fuselage components, which require explicit ID wiring per the accessibility docs. In code reviews, prefer using fuselage-forms wrappers for form fields and verify there is no unnecessary manual ID/aria wiring in files that use these wrappers. If a component uses plain fuselage components, ensure proper id wiring as per docs.
Applied to files:
packages/models/src/models/LivechatInquiry.tspackages/models/src/models/Sessions.tspackages/models/src/models/Users.tspackages/models/src/models/Messages.tspackages/models/src/models/Rooms.tspackages/models/src/models/LivechatRooms.ts
🔇 Additional comments (8)
packages/models/src/models/LivechatRooms.ts (1)
2045-2045: LGTM!Also applies to: 2063-2075, 2077-2077, 2102-2102, 2119-2132, 2134-2134
packages/models/src/models/LivechatInquiry.ts (1)
66-66: LGTM!Also applies to: 285-290
packages/models/src/models/Messages.ts (2)
195-216: LGTM!Also applies to: 563-564
225-230: 🎯 Functional CorrectnessNo change needed for unscoped counts. The
nulldepartment bucket is intentional when nodepartmentIdis passed, so theroom.0filter should stay inside the department-specific branch.> Likely an incorrect or invalid review comment.packages/models/src/models/Rooms.ts (1)
12-12: LGTM!Also applies to: 1996-2013
packages/models/src/models/Sessions.ts (1)
990-990: LGTM!Also applies to: 1105-1105, 1128-1128, 1196-1203, 1237-1237
packages/models/src/models/Users.ts (1)
275-288: LGTM!Also applies to: 1043-1045
.changeset/model-aggregations-perf.md (1)
1-7: LGTM!
![cubic-dev-ai[bot]](https://avatars.githubusercontent.com/in/1082092?s=60&v=4){kind=small}
Contributor
There was a problem hiding this comment.
1 issue found across 16 files
Prompt for AI agents (unresolved issues)
Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.
<file name="packages/models/src/models/Users.ts">
<violation number="1" location="packages/models/src/models/Users.ts:282">
P2: Agents without department assignments now return a different `departments` shape (`[]` instead of previous `[null]`), which changes the omnichannel users response for that edge case. If output compatibility is required, preserve the empty-lookup case before mapping department IDs.</violation>
</file>
Reply with feedback, questions, or to request a fix.
Re-trigger cubic
Comment on lines
282
to
290
| departments: { | ||
| $map: { | ||
| input: '$departments', | ||
| as: 'department', | ||
| in: '$$department.departmentId', | ||
| }, | ||
| }, | ||
| }, | ||
| }, |
Contributor
There was a problem hiding this comment.
P2: Agents without department assignments now return a different departments shape ([] instead of previous [null]), which changes the omnichannel users response for that edge case. If output compatibility is required, preserve the empty-lookup case before mapping department IDs.
Prompt for AI agents
Check if this issue is valid — if so, understand the root cause and fix it. At packages/models/src/models/Users.ts, line 282:
<comment>Agents without department assignments now return a different `departments` shape (`[]` instead of previous `[null]`), which changes the omnichannel users response for that edge case. If output compatibility is required, preserve the empty-lookup case before mapping department IDs.</comment>
<file context>
@@ -272,21 +272,20 @@ export class UsersRaw extends BaseRaw<IUser, DefaultFields<IUser>> implements IU
+ name: 1,
+ emails: 1,
+ livechat: 1,
+ departments: {
+ $map: {
+ input: '$departments',
</file context>
Suggested change
| departments: { | |
| $map: { | |
| input: '$departments', | |
| as: 'department', | |
| in: '$$department.departmentId', | |
| }, | |
| }, | |
| }, | |
| }, | |
| departments: { | |
| $cond: [ | |
| { $gt: [{ $size: '$departments' }, 0] }, | |
| { | |
| $map: { | |
| input: '$departments', | |
| as: 'department', | |
| in: '$$department.departmentId', | |
| }, | |
| }, | |
| [null], | |
| ], | |
| }, |
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## develop #41157 +/- ##
===========================================
- Coverage 69.26% 69.21% -0.05%
===========================================
Files 3536 3536
Lines 138592 138598 +6
Branches 24681 24720 +39
===========================================
- Hits 95991 95935 -56
- Misses 38592 38651 +59
- Partials 4009 4012 +3
Flags with carried forward coverage won't be shown. Click here to find out more. 🚀 New features to boost your workflow:
|
Rocket.Chat requires MongoDB >=7.0, where allowDiskUseByDefault is true (since 6.0) and memory-exceeding stages spill to disk automatically — the explicit flags added no behavior on supported deployments.
The trimmed projection measured a consistent ~5% regression (128.5 -> 135.9ms at 40k sessions, 756 -> 795ms at 200k; non-overlapping ranges) — building the sub-object per document costs more than the smaller $unwind copies save, and there is no failure mode to protect against. Only the logintAt -> loginAt index typo fix remains.
Contributor
There was a problem hiding this comment.
🧹 Nitpick comments (2)
packages/models/src/models/Sessions.ts (2)
1191-1198: 🚀 Performance & Scalability | 🔵 Trivial | ⚡ Quick winSame
$$ROOTbloat issue — onlyday/month/yearare used downstream.The subsequent
groupsreference onlysession.day,session.month, andsession.year(Lines 1204-1206), yet the entire document is embedded undersessionand then duplicated by$unwindacross the hour range. Narrowing this projection to the actually-used fields would reduce the per-bucket payload size, consistent with this PR's performance goals.♻️ Suggested fix
const rangeProject = { $project: { range: { $range: [{ $hour: '$loginAt' }, { $sum: [{ $ifNull: [{ $hour: '$closedAt' }, 23] }, 1] }], }, - session: '$$ROOT', + session: { day: '$day', month: '$month', year: '$year' }, }, };🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@packages/models/src/models/Sessions.ts` around lines 1191 - 1198, The $project in the rangeProject stage is carrying the full $$ROOT document into session even though downstream only reads session.day, session.month, and session.year in the groups logic. Update the projection in rangeProject to include only those needed session fields instead of $$ROOT, so the $unwind over range does not duplicate unnecessary document data and the pipeline stays lean.
1100-1107: 🚀 Performance & Scalability | 🔵 Trivial | ⚡ Quick winAvoid
$$ROOTbefore$unwind— project only needed fields.Setting
session: '$$ROOT'embeds the entire session document undersessionright before$unwindfans it out per range bucket, multiplying the duplicated payload by the bucket count. Since this PR's goal is performance improvement, consider projecting only the specific fields consumed later (if any beyondrange), rather than the full document.♻️ Suggested direction
const rangeProject = { $project: { range: { $range: [0, 24, groupSize], }, - session: '$$ROOT', + // project only the specific fields needed downstream, e.g.: + // loginAt: 1, }, };🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@packages/models/src/models/Sessions.ts` around lines 1100 - 1107, The `rangeProject` stage in `Sessions` is carrying the full document via `session: '$$ROOT'` before `$unwind`, which duplicates unnecessary payload across every bucket. Update the aggregation to project only the fields actually needed in later stages (or drop `session` entirely if it is unused), keeping the `range` logic intact while reducing fan-out overhead.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@packages/models/src/models/Sessions.ts`:
- Around line 1191-1198: The $project in the rangeProject stage is carrying the
full $$ROOT document into session even though downstream only reads session.day,
session.month, and session.year in the groups logic. Update the projection in
rangeProject to include only those needed session fields instead of $$ROOT, so
the $unwind over range does not duplicate unnecessary document data and the
pipeline stays lean.
- Around line 1100-1107: The `rangeProject` stage in `Sessions` is carrying the
full document via `session: '$$ROOT'` before `$unwind`, which duplicates
unnecessary payload across every bucket. Update the aggregation to project only
the fields actually needed in later stages (or drop `session` entirely if it is
unused), keeping the `range` logic intact while reducing fan-out overhead.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Organization UI
Review profile: CHILL
Plan: Pro
Run ID: fd8aa600-d298-4bc6-b8c9-5d92b71589ba
📒 Files selected for processing (1)
packages/models/src/models/Sessions.ts
📜 Review details
⏰ Context from checks skipped due to timeout. (4)
- GitHub Check: ⚙️ Variables Setup
- GitHub Check: cubic · AI code reviewer
- GitHub Check: CodeQL-Build
- GitHub Check: CodeQL-Build
🧰 Additional context used
📓 Path-based instructions (1)
**/*.{ts,tsx,js}
📄 CodeRabbit inference engine (.cursor/rules/playwright.mdc)
**/*.{ts,tsx,js}: Write concise, technical TypeScript/JavaScript with accurate typing in Playwright tests
Avoid code comments in the implementation
Files:
packages/models/src/models/Sessions.ts
🧠 Learnings (3)
📚 Learning: 2026-02-26T19:25:44.063Z
Learnt from: gabriellsh
Repo: RocketChat/Rocket.Chat PR: 38778
File: packages/ui-voip/src/providers/useMediaSession.ts:192-192
Timestamp: 2026-02-26T19:25:44.063Z
Learning: In the Rocket.Chat repository, do not reference Biome lint rules in code review feedback. Biome is not used even if biome.json exists; only reference Biome rules if there is explicit, project-wide usage documented. For TypeScript files, review lint implications without Biome guidance unless the project enables Biome rules.
Applied to files:
packages/models/src/models/Sessions.ts
📚 Learning: 2026-02-26T19:25:44.063Z
Learnt from: gabriellsh
Repo: RocketChat/Rocket.Chat PR: 38778
File: packages/ui-voip/src/providers/useMediaSession.ts:192-192
Timestamp: 2026-02-26T19:25:44.063Z
Learning: In this repository (RocketChat/Rocket.Chat), Biome lint rules are not used even if a biome.json exists. When reviewing TypeScript files (e.g., packages/ui-voip/src/providers/useMediaSession.ts), ensure lint suggestions do not reference Biome-specific rules. Rely on general ESLint/TypeScript lint rules and project conventions instead.
Applied to files:
packages/models/src/models/Sessions.ts
📚 Learning: 2026-05-06T12:21:44.083Z
Learnt from: juliajforesti
Repo: RocketChat/Rocket.Chat PR: 40256
File: apps/meteor/client/components/CreateDiscussion/CreateDiscussion.tsx:121-149
Timestamp: 2026-05-06T12:21:44.083Z
Learning: Field wrappers in rocket.chat/fuselage-forms (Field, FieldLabel, FieldRow, FieldError, FieldHint) auto-create htmlFor/id associations, aria-describedby, and role="alert" for errors. Do not manually set htmlFor, id, aria-describedby, or role attributes when using these wrappers. This automatic wiring does not apply to plain rocket.chat/fuselage components, which require explicit ID wiring per the accessibility docs. In code reviews, prefer using fuselage-forms wrappers for form fields and verify there is no unnecessary manual ID/aria wiring in files that use these wrappers. If a component uses plain fuselage components, ensure proper id wiring as per docs.
Applied to files:
packages/models/src/models/Sessions.ts
🔇 Additional comments (1)
packages/models/src/models/Sessions.ts (1)
989-994: LGTM!
The redundant second $group in getTotalOfRegisteredUsersByDate operates on ~#days documents and the findAgentsWithDepartments regroup on a small admin listing — no measurable perf or memory gain, so they don't belong in this branch.
KevLehman
changed the title
KevLehman
marked this pull request as draft
July 3, 2026 14:03
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Proposed changes (including videos or screenshots)
Audit of every
.aggregate()inpackages/models+ EE raw models, applying only rewrites that return byte-identical output (verified doc-by-doc against seeded data).Pipeline rewrites
LivechatRooms.getAnalyticsBetweenDate/getAnalyticsMetricsBetweenDateWithMessages: message counting moved into the$lookupsub-pipeline ($count) instead of exploding every joined message through$unwind+$group. The historicalmsgs: 1for zero-message rooms is preserved.Messages.findAllNumberOfTransferredRooms: transfer messages grouped byridbefore the room$lookup— one join per distinct room projecting onlydepartmentId, department filter inside the sub-pipeline.Rooms.getSubscribedRoomIdsWithoutE2EKeys: subscription filter (u._id,E2EKey) inside the$lookup+$limit 1. Availability fix: the old shape materializes the full room membership as one array per room doc, so very large encrypted rooms can overflow the 16MB BSON limit and error.Sessions:logintAt→loginAtindex typo fix. (A$$ROOT-trim of the hourly engagement pipelines was measured as a consistent ~5% regression and reverted — see benchmark notes.)LivechatInquiry:{status}index widened to{status, department}(prefix keeps serving status-only queries).Dead code removed (only model + typing references existed):
LivechatRooms.findAllNumberOfTransferredRooms(runtime uses theMessagesvariant),Rooms.findChannelsByTypesWithNumberOfMessagesBetweenDate+ builder,Messages.findOneByFederationIdAndUsernameOnReactions,Sessions.getActiveUsersBetweenDates,LivechatDepartmentAgents.findAgentsByAgentIdAndBusinessHourId(CE + EE files, class went empty). Net −400 lines.Benchmarks — k6 at 8 req/s for 60s per build (develop vs branch, same seeded data) +
explain('executionStats')on old vs new pipelines. MongoDB 7.0.16. Dataset: 2.2k livechat rooms / 146k messages / 60k queued inquiries / 118k subscriptions. Every rewritten pipeline verified to return identical output before measuring.k6, endpoint level (median / p95):
livechat/analytics/dashboards/conversation-totalizers(30d)livechat/analytics/dashboards/charts/chatsengagement-dashboardsessions charts ×2explain, engine level:
getDistinctQueuedDepartments(60k queued)getAnalyticsBetweenDate(146k msgs)$unwind+$group)getSubscribedRoomIdsWithoutE2EKeys(50 rooms ×2,000 subs)Changes that measured neutral-or-worse were reverted rather than shipped (Sessions
$$ROOTtrim: real ~5% regression; Users pipeline micro-cleanups andallowDiskUseflags: no measurable gain).Issue(s)
Steps to test or reproduce
Further comments
Every rewrite was validated for output-equivalence before benchmarking; regressions found during benchmarking (e.g. sub-pipeline lookups losing the SBE hash-join) are documented and were only kept where they fix a failure mode (the 16MB BSON doc limit on the E2E lookup) rather than latency.
An earlier revision added
allowDiskUse: trueto a few pipelines; those flags were dropped — Rocket.Chat requires MongoDB >=7.0, whereallowDiskUseByDefault: true(since 6.0) spills memory-exceeding stages to disk automatically, making them no-ops on supported deployments.Summary by CodeRabbit
Performance Improvements
Bug Fixes