Gluesync release notes and changes

Welcome to the Gluesync release notes. This page provides detailed information about each release, including new features, improvements, and fixes.

📬 Stay in the Loop!

Looking forward to keep you updated on Gluesync releases? Subscribe to Gluesync Updates to receive our latest news and updates, including new features, improvements, and fixes right into your inbox. We promise to keep it short and not spammy.

Quick Links

Latest Release: 2.2.11.11

Released: October 5, 2026

  • Agents:

  • UI:

    • Fixed annoying scroll of modal when selecting a checkbox;

    • Added missing "On duplicate key" option in group settings;

2.2.11.10

Released: October 1, 2026

  • Agents:

  • UI:

    • Fixed change password form can be locked in some cases;

    • The "Edit group" button is now hidden for users who lack the create-pipeline permission;

    • Fixed the calls to Chronos. The scheduler job and trigger-flow queries and mutations now use a dedicated Chronos endpoint;

2.2.11.9

Released: September 29, 2026

2.2.11.8

Released: September 28, 2026

  • Agents:

  • UI:

    • Fixed Notifications Hub interface;

  • Core Hub:

    • GSSD-1354: An interrupted start no longer leaves the entities it did not reach flagged as syncing with no reader behind them, which made every later start skip them as already syncing;

    • GSSD-1354: An entity connectivity badge is cleared on the next successful poll, so a table that never changes no longer stays in error forever after a single outage;

    • GSSD-1354: A failure is classified from what the driver reports, so an unreachable database is no longer reported as a credentials problem;

    • Chronos: now the scheduled snapshots always reset the migration checkpoint;

2.2.11.7

Released: September 25, 2026

  • Agents:

  • UI:

    • Added log collection help modal;

    • Separated log collection statuses tracking success/failure independently for "collect logs locally" vs "send to FTP";

    • Normalized support ticket IDs;

  • Core Hub:

    • GSSD-1370: An entity created on a view is registered for snapshot only, since no source connector captures the changes of a view;

2.2.11.6

Released: September 24, 2026

2.2.11.5

Released: September 22, 2026

  • Agents:

  • Core Hub:

    • Prometheus metrics now carry the pipeline, entity and agent name beside the id, so the Grafana checkpoint and entity-lag panels no longer show bare uuids;

    • A copy of a column keeps the type it was built with, instead of throwing on the first read of it after a renumbering;

2.2.11.4

Released: September 21, 2026

2.2.11.3

Released: September 16, 2026

  • UI:

    • New settings icon for global settings page allows you to quickly navigate to the settings page;

    • Fixes and improvements;

  • Agents:

  • Core Hub:

    • Added new remote support access request, with its own tunnel manager and REST routes;

    • Duplicate connection names are refused instead of answering 201 for a row that was never stored;

    • An agent that has no pipeline yet reports the real connection error instead of an opaque 500;

    • Persisted JDBC gateway settings are read after the database is initialised, so they are no longer ignored in favour of the default TTLs;

    • Fixes and improvements;

2.2.11.2

Released: September 14, 2026

  • Source agent cache (ArenaCache):

    • The local source-agent cache is now ArenaCache, replacing Chronicle Queue on every agent that persisted changes locally. ArenaCache ran in production on Oracle LogMiner for about six months before this rollout.

    • Per-entity cache files, lower disk and IOPS cost, and restart barriers that isolate a truncated or corrupted file.

    • Chronicle Queue files are not reused after upgrade — bring pipelines to low lag first. CDC reader position for cache-based agents still lives in Core Hub (from 2.2.11).

    • See Source agent caching layer.

  • Agents:

  • Core Hub:

    • Added per-entity Insert Conflict Strategy (UPSERT / SKIP / FAIL) replacing the blind insert-to-upsert fallback that silently truncated values through a CAST;

    • Added ScalarFunctionCall function, supported on every RDBMS target and on every column type;

    • Added possibility to use field functions and UDF together in the same entity;

    • Affected data is grouped by keys with a hash lookup instead of a quadratic scan;

  • Query Studio:

    • Manual transaction support, advertised per agent and forwarded to the agent executor;

    • Bounded distinct value sampling per column;

    • Complete cell update semantics;

    • EXTERNAL_MODULE users can execute Query Studio SQL;

2.2.11.1

Released: September 11, 2026

  • Agents:

  • UI:

    • Added AI agents (MCP) settings tab;

  • Core Hub:

    • Fixed metrics snapshot concurrent access causing errors;

  • Query Studio:

    • The errors that caused nothing but noise have been silenced;

    • Enabled auto commit for some databases;

    • Distinct-value retrieval for a selected column (scopeSql, column, optional parameters, sampleLimit, limit, includeNull; response values, sampleLimit, limit, truncated; Hub clamps sampleLimit to 5,000 and limit to 200; UI browse filters use this with a current-page fallback);

    • Cell update with an explicit valueMode of VALUE, NULL, or DEFAULT;

    • Client-provided queryId so the matching query can be cancelled;

    • Database identifiers quoted according to the selected agent’s dialect;

    • Matching Control Plane chrome (UI !538): one-click browse with SafetyGate forceReadOnly, browse-filter chips, a value viewer, record view, a client-side session statement feed, Copy as Markdown/HTML/XML, and Set NULL / Set DEFAULT cell actions. See Query Studio.

The Query Studio operator documentation has also been synchronized with capabilities already shipped in 2.2.11 but previously under-documented: Chart results, XLSX export, edit-in-place, CSV/XLSX import, manual transactions, and the Dev Mode MCP tools and toolsets workbench.

2.2.11

Released: September 9, 2026

Gluesync 2.2.11: your readers stop forgetting, and your dashboard starts telling the truth

Release highlights

Dear data wranglers,

The upgrade to 2.2.11 may take a few minutes to complete. While it runs, the UI stays paused until the process finishes — that’s expected. Wait for it to complete, then refresh if needed.

Last time we handed you a federated query gateway and a wizard that actually fills itself in. This time we went after the stuff that still made you hold your breath after an agent restart, squint at a settings page that looked like a stats textbook, and stare at a Grafana dashboard that could not tell source lag from Core Hub thinking-time.

We’ve been busy. Here’s what’s new:

Query Studio: now with transactions, spreadsheets, and a phone

Query Studio already saved you from tabbing over to DBeaver. This round it grew the features people actually asked for after the first week:

  • Begin / commit / rollback on a dedicated connection — because "oops" should be undoable

  • Download a result set as .xlsx — for the colleague who still lives in Excel

  • UPDATE a cell in place — no round-trip through a SQL editor for a one-cell fix

  • And yes: Query Studio on mobile. Your pipelines were already in your pocket. Your SQL workbench can come too.

It’s still the workbench inside Core Hub. It just stopped feeling like a first draft.

Grafana + Prometheus — the dashboard finally grew up

Back in 2.2.8 we promised the Grafana dashboard would "soon receive some love." Soon is now.

The official Core Hub metrics dashboard is rebuilt around the KPIs you actually use at 2 a.m.:

  • Rows and bytes delivered, who’s snapshotting, and who’s gone quiet

  • CDC reader freshness (gluesync_read_checkpoint_age_seconds) — that’s source lag, not Core Hub’s internal stopwatch

  • Stalled entities and the last transaction ID a target actually confirmed

  • JDBC connections by purpose (pipeline vs Query Studio vs Query Forge) and HikariCP pool pressure

  • Host CPU, JVM heap, GC pauses, and Core Hub uptime

  • Live smart-alert thresholds drawn on the same charts as the measurements

The /metrics scrape grew up with it: entity-heartbeat configuration, connection totals, JVM runtime, Ktor HTTP request timings, and checkpoint/pool series that show up when those subsystems are actually running. Pipeline name filters delivery panels; pipeline id filters checkpoints — pick both, or leave both on All.

It’s the same Prometheus endpoint. It just stopped making you guess which number was "lag." See Grafana dashboards and Metrics reference.

Smart alerts you pick — sensitivity, not a knob zoo

The 2.2.8 beta watched every pipeline like an over-caffeinated intern. That was…​ a lot. In Settings → Thresholds, Monitored objects is now the allowlist: empty means nobody, and new objects stay off until you add them.

Detection is Low / Normal / High plus a cooldown. Nested statistician knobs stay on the wire (hello, compatibility) but they are ignored. What still actually fires:

  • Source watchdog (hard floor + heartbeat)

  • Target idle stall

  • An optional latency ceiling

Seasonal, throughput EWMA, and latency-pattern detectors no longer wake you at 3 a.m. for a vibe. You pick who gets watched. You pick how jumpy it is. That’s it.

Who’s holding that connection? Now you can actually see

"How many SQL sessions is Core Hub sitting on, and why?" used to be a mystery novel. Not anymore.

Core Hub now tracks every live SQL connection and what it is for — pipeline work, Query Studio, Query Forge. Admins can see personal access tokens and last-used time. The same snapshot is scraped as Prometheus gauges (gluesync_db_connections_active), so Grafana can tell you when Query Studio is competing with replication for the customer’s database.

If something is hogging a connection, you no longer have to guess whose hand is on it.

CoreHub MCP server — your AI assistant finally has a badge

An MCP server now runs inside Core Hub: same JVM, same RBAC, no sidecar, no extra privileges. Point Claude, Cursor, or any MCP client at /mcp or /mcp/http and it gets 55 tools for pipelines, entities, metrics, Query Studio, and the connection catalogue (list_connections, get_connection, get_connection_usages) — plus per-agent SQL tools discovered from the caller’s token.

Remote clients can connect by IP. Discovery even works against a self-signed HTTPS Core Hub, because of course your lab cert is self-signed.

Your agent can talk to the hub. The hub still decides what it’s allowed to do.

Connections catalog — the plumbing, not the showroom (yet)

Database credentials used to live inside each pipeline’s encrypted blob. That made reuse awkward, and some custom properties (for example logminerPassword) could still show up in views meant for operators who only need to look, not to handle secrets.

This release pulls connections out of that blob and into a reusable global catalog internally. Sensitive fields stay out of those read-only views. There is no Control Plane UI for it yet — you will not pick a shared connection from a wallet in this drop. The catalog is the foundation; the showroom comes next.

The password-once, many-pipelines experience is the destination. Today we moved the furniture.

CDC checkpoints that survive a move — no more "where was I?"

Cache-based CDC readers used to keep their position in a local file. Move the agent, restart the box, and congratulations: the reader just forgot everything and started over.

Now Core Hub owns the checkpoint. An agent move or restart no longer resets the reader. Read errors retry instead of jumping the position like nothing happened. AS/400 ships first, with MySQL, MariaDB, Informix, and Oracle LogMiner in the same release — plus REST/WS exposure and a Prometheus scrape, because a checkpoint you can’t see is just a rumor.

Your reader can move house. It still knows what page it was on.

Additional improvements

  • Core Hub:

    • Connections catalog extracted from the pipeline (internal; Control Plane UI follows);

    • CoreHub-managed CDC read checkpoints (AS400, MySQL, MariaDB, Informix, Oracle LogMiner);

    • Pipeline metrics WebSocket throttle relaxed to 1s, plus GET /ui/pipelines-metrics;

    • Support diagnostics bundle now includes a pipelines inventory without credentials;

    • Every log record carries pipeline / entity / actor context;

    • Idle CDC poll logs collapse to a once-a-minute digest;

    • Smart alerts: allowlist plus Low / Normal / High (see highlights);

    • Pipeline-level Smart Alert overrides — per-pipeline threshold/sensitivity/cooldown/anomaly-type selection and monitored-entity allowlists on top of global defaults;

    • Live SQL connection registry, admin PAT last-used visibility, and Prometheus gluesync_db_connections_active gauges (see highlights);

    • Official Core Hub Grafana dashboard and expanded /metrics scrape (see highlights);

    • Entity status WARNING when a smart-alert threshold fires; real exceptions stay ERROR;

    • Pipelines in maintenance or warm-up can now be deleted;

    • Shared connections keep working when some of the pipelines that use them are in maintenance;

    • Snapshot-only entities via a dedicated flag;

    • After creating a user, credentials can be sent by email;

    • Agent certificate upload no longer requires a connection that does not exist yet;

    • Agent credentials configuration no longer falls back to a placeholder connection;

    • Stopped entities no longer raise false heartbeat alerts;

    • Smart-alert samples persist across a CoreHub restart;

    • Embedded Core Hub MCP server (see highlights): remote clients by IP; discovery against self-signed HTTPS; optional GLUESYNC_MCP_IN_PROCESS (defaults false, dev-only); connection-catalogue tools (list_connections, get_connection, get_connection_usages) bringing the built-in catalog to 55;

    • Hub platform events carry a stable source {kind, id};

    • Store-confirmation logs include the transaction id;

    • Prometheus gluesync_entity_dry_seconds is labelled with pipeline and entity names;

  • Chronos:

    • Added Query Studio actions (query_studio) for schedules, chained events, platform events, and webhook-triggered events — run a saved query or custom SQL against a selected pipeline connection;

  • Query Studio:

    • Manual transactions, XLSX export, edit-in-place;

    • Query Studio on mobile;

    • External modules can call Query Studio;

    • Oracle CLOB / NCLOB cells render as text;

    • PostgreSQL import fix;

    • Couchbase querying via SQL++;

  • Agents:

2.2.10

2.2.10.16

Released: September 1, 2026

2.2.10.15

Released: August 28, 2026

  • UI:

    • Made UUID generation work over HTTP;

    • Added form in global setting for legacy TLS connections. See Change Tracking and CDC TLS troubleshooting;

2.2.10.14

Released: August 28, 2026

2.2.10.13

Released: August 21, 2026

2.2.10.12

Released: August 18, 2026

2.2.10.11

Released: August 17, 2026

2.2.10.10

Released: August 13, 2026

2.2.10.9

Released: August 11, 2026

  • Agents:

    • Oracle LogMiner agent: fixes and improvements;

    • Added snapshot column alignment diagnostics: a warning is now logged when the source result set does not match the expected columns, reporting row keys, position and data type of the not nullable columns read as null;

    • Lowered the default value of the Source change retention in hours property from 3 days to 1 day;

2.2.10.8

Released: August 6, 2026

  • Core Hub:

    • Fixed pipeline placeholder not replaced in Notifications;

  • Agents:

  • UI:

    • Improved the Pipelines page UI by adjusting entity status icons size and colours for better visibility and consistency;

    • Fixed errors occurring on the Notifications page;

    • Fixed text truncation issues in the Entity detail breadcrumb;

    • Fixed the Agent setup form background styling issue;

    • Fixed an issue where Chains displayed both source and target entities in the list;

    • Fixed the incorrect layout displayed in the Chain detail page;

2.2.10.7

Released: August 5, 2026

2.2.10.6

Released: August 4, 2026

  • Agents:

  • Core Hub:

    • Removed cdc batch completed notification save;

    • Fixed pipeline not in draft after agent removal;

2.2.10.5

Released: August 3, 2026

2.2.10.4

Released: July 30, 2026

2.2.10.3

Released: July 30, 2026

  • UI:

    • IBM DB2 LUW CDC agent is now visible from the UI setup wizard;

    • Universal file store agent is now visible from the UI setup wizard;

2.2.10.2

Released: July 29, 2026

  • Agents:

  • UI:

    • Added automatic GSSD- prefix to the Ticket ID input field, ensuring consistent ticket formatting;

    • Improved entity group sorting in the Entities table for a more intuitive and predictable ordering;

    • Enhanced Support navigation: clicking Support now opens the corresponding Support section within Settings, providing a smoother user experience;

2.2.10.1

Released: July 28, 2026

  • Core Hub:

    • Pre/Post CDC commands for every RDBMS target databases, plus new webhook type ENTITY_CDC_BATCH_COMPLETED for the same;

2.2.10

Released: July 27, 2026

Gluesync 2.2.10: your data platform just learned to speak SQL (and a few other tricks)

Release highlights

Dear data wranglers,

Remember when replication was just about copying data from A to B and hoping for the best? Those days are officially over. With 2.2.10, Gluesync stops being just a replication engine and becomes a proper data platform — one that lets you query your federated data, orchestrate complex event chains, and do it all from a UI that doesn’t require a PhD in "where did that button go?"

We’ve been busy. Here’s what’s new:

Query Forge: your data, one SQL query away

Ever wanted to run a single SQL query that joins your PostgreSQL sales data with your MySQL customer records — without standing up a data warehouse, writing a custom ETL pipeline, or sacrificing a goat to the integration gods? Now you can.

Query Forge turns Gluesync into a federated query gateway. Connect with our brand-new JDBC driver (jdbc:gluesync://<host>:<port>), point it at your CoreHub, and your BI tools — DataGrip, DBeaver, Tableau, you name it — can query across agents as if they were a single database.

The best part? It’s secure by design:

  • Authentication is Personal API Token only (gsp_…) — no shared credentials, no password login

  • Each token is scoped to specific pipelines and agents, so users only see what they’re allowed to

  • All queries are read-only — your source data is safe

  • Filter and projection pushdown means only the data you actually need travels over the wire

  • Cross-agent joins, typed aggregations, and ANSI-compatible identifiers work out of the box

It’s like having a data warehouse without the warehouse part.

Chronos grows up: chained events, webhooks, and security that actually works

Chronos — our scheduling module — just got a serious upgrade. It’s no longer just "run this job every Tuesday at 3 AM." Now it can:

  • Chain events together: Execute a sequence of actions one after another, in sync mode (wait for each step’s completion callback before firing the next) or async mode (fire-and-forget). Perfect for "snapshot first, then start CDC" workflows.

  • React to external events via webhooks: Chronos can now be triggered by webhooks from CoreHub, so external systems can kick off scheduled flows on demand using secret tokens.

  • Listen to platform-bound events: CoreHub platform events — like pipeline state changes — can now trigger Chronos flows automatically via persistent webhook registrations. No more polling.

  • Load faster: We’ve optimized job loading with batch queries (goodbye N+1), cached timezone resolution, and backfilled next_run values so the scheduler starts up quicker than ever.

  • Secure every endpoint: All Chronos API endpoints are now gated behind JWT authentication, validated against CoreHub’s /auth/me endpoint. Role-based permissions (SUPER_ADMIN, MANAGER, MONITOR, VIEWER) ensure users only see and do what they’re allowed to. SSL/HTTPS support with configurable TLS verification rounds out the package.

Because scheduling your data pipeline shouldn’t be an open door.

Pipeline wizard: now with less "wait, what do I put here?"

We rebuilt the pipeline creation wizard from the ground up. The new experience is:

  • Streamlined: A clean, step-by-step flow that guides you from source to target without overwhelming you with options you don’t need yet.

  • Auto-filling: The wizard now pulls metadata from your backend agents and pre-fills form fields — connection details, schema info, table lists — so you spend less time copy-pasting and more time actually getting things done.

  • Validated in real-time: Form validation powered by react-hook-form and yup catches configuration errors before they become runtime problems.

It’s the pipeline wizard you always wished you had. Now you have it.

New UI for Scheduler and Modules Manager — now with user roles

Two of our most powerful management screens just got the full UI treatment:

Scheduler: A brand-new scheduling interface with weekly and CRON schedule types, a visual chained-events editor, timezone settings, and run-now capabilities — all gated by user role permissions so only the right people can create, edit, or trigger schedules.

Modules Manager: A complete view of every agent, module, and third-party service orchestrated by CoreHub. See status at a glance (healthy, error, updating, update available), filter by family, and perform lifecycle actions (start, restart, stop, remove) — again, only if your role allows it.

Both screens respect the same role hierarchy used across Gluesync: SUPER ADMIN, MANAGER, and VIEWER — each with appropriate permissions for managing schedules, controlling replication, and modifying configuration.

Power where you need it. Guardrails where you want them.

Additional improvements

As per every major update Gluesync may require a few minutes to restart after the update is applied.

  • Core Hub:

    • Query Forge: use Gluesync as a gateway to run secure federated queries via the new JDBC driver (jdbc:gluesync://), powered by Apache Calcite/Avatica;

    • Personal API Token support for JDBC gateway authentication;

    • Ability to disable the SMTP client;

    • New local logs collection with optional database export;

    • New event type for table reorganization;

    • External modules user role support;

  • Chronos (Scheduler Module):

    • Chained events with sync/async execution modes and configurable webhook timeouts;

    • Webhook triggers for external events;

    • Platform-bound events via persistent webhook registration in CoreHub with retry and filter support;

    • Faster loads: optimized job loading with batch queries and cached timezone resolution;

    • All API endpoints are now secured with JWT authentication via CoreHub introspection;

    • Role-based access control (SUPER_ADMIN, MANAGER, VIEWER, EXTERNAL_MODULE);

  • Conductor (v0.10.0):

    • New Conductor module for managing Gluesync services and agents;

    • Authentication via CoreHub token introspection with cookie and bearer token support;

    • Role-based access control (SUPER_ADMIN, MANAGER, VIEWER, EXTERNAL_MODULE) with route-level authorization guards;

    • Sensitive data redaction in API responses for users without configuration-modification permissions;

    • Readonly functionality for VIEWER role;

  • UI:

    • All-new pipeline creation wizard with form autofill and real-time validation;

    • New Scheduler page with weekly/CRON schedule types and chained events editor;

    • New Modules Manager with service status, lifecycle actions, and family filtering;

    • User role permissions integrated across Scheduler and Modules Manager;

    • Source primary keys uniqueness check;

    • Empty schemas are now hidden from the object browser at source database level;

    • Added distinction between table, view, or member for every database;

    • User-customizable notification retention settings;

  • Agents:

2.2.9

2.2.9.13

Released: July 24, 2026

2.2.9.12

Released: July 22, 2026

2.2.9.11

Released: July 20, 2026

2.2.9.10

Released: July 17, 2026

  • Agents:

  • Core Hub:

    • GSSD-1160: Fixed operation column alias can collide if customer table has a column called operation;

    • GSSD-1211: Fixed entity paused automatically during retry;

    • GSSD-1189: Fixed high cpu usage due to map not thread safe;

    • GSSD-1205: Made cache record evictions persistent across restarts;

    • GSSD-1193: Fixed entity status not recovering from error;

    • Improved log management;

2.2.9.9

Released: July 16, 2026

2.2.9.8

Released: July 14, 2026

  • Agents:

  • Core Hub:

    • GSSD-1185: Fixed missing user permissions check on some endpoints;

    • GSSD-1104: Fixed check connection to DB made too often;

    • Fixed post snapshot statement not executed;

  • UI (v0.3.6):

    • GSSD-1188: Fixed primary key auto selection logic;

    • Fixed snapshot metrics and status display issues in Entity Details;

    • Fixed incorrect snapshot labels, partitions value, and Play/Pause button state;

    • Fixed multiple issues in the Pipelines page, including RPS calculation, feature icons, chain selection, colors, group management, navigation, and text truncation;

    • Fixed the bulk selection indicator disappearing unexpectedly;

    • Fixed an issue preventing CDC and Snapshot operations from starting in Gluesync Mobile;

2.2.9.7

Released: July 10, 2026

2.2.9.6

Released: July 9, 2026

  • Agents:

    • For all agents with bulk load support:

      • GSSD-1120: Fixed deduplication logic for primary keys columns with case insensitive collation

    • IBM i series agent: fixes and improvements;

    • NOTE: for the IBM i series agent, if you’re using Members please read the change log for more information;

  • UI:

    • Fixed save entity with unlock schema and custom column;

  • Core Hub:

    • GSSD-1160: Fixed unsupported operation error in the query editor;

    • Aligned jetty idle timeout to 10 minutes as rest api timeout;

    • GSSD-1120: Fixed deduplication logic for primary keys columns with case insensitive collation;

    • GSSD-1134: Added flush on close with related configuration param enabled by default;

2.2.9.5

Released: July 8, 2026

  • UI:

    • GSSD-1143/1163: Fixed an issue that is causing wrong tables and columns information to be saved in the database. Gluesync will auto-align all the information during the reboot;

  • Core Hub:

    • Fixed logs wrong path;

2.2.9.4

Released: July 7, 2026

  • Agents:

  • UI:

    • Fixed NaN shown for date columns in tables partitions;

  • Core Hub:

    • GSSD-1120: Improvements to Primary Keys management to avoid violation of primary keys constraint errors when writing on target tables;

    • GSSD-1139: Improved warm up to retry only when agent exist and always reload entities to make sure to send always the last entities available;

    • Added a new Gluesync image, based on Debian 13, to support X86-64 v2 CPU bite code (for legacy hosts);

2.2.9.3

Released: July 6, 2026

2.2.9.2

Released: July 2, 2026

Multimember Files Support – Because One Member Is Not Enough

If you’ve ever worked with IBM i, you know that physical files don’t always play by the "one file, one member" rule. In fact, some of your most important tables are physical files with multiple members — each one holding a separate slice of data, all sharing the same record format.

Until now, Gluesync could only see the first member of a physical file. The rest? Invisible. Like they didn’t exist. Which, for replication purposes, they kind of didn’t.

  • Discover and replicate data from all members within a multimember physical file

  • Each member is treated as its own entity, with its own pipeline and CDC tracking

  • Journal-based CDC works per-member, so changes in any member are captured independently

  • Snapshots and bulk loads respect member boundaries

Whether you’re partitioning data by date, region, or business unit across members of the same physical file — Gluesync now sees every single one of them. No member left behind. See IBM i series intro for more details.

  • UI:

    • Fixed pipeline list styling;

    • Fixed chain colors;

    • Fixed various UI styling issues;

    • Fixed pipeline navigation;

    • Fixed agent icon click navigation;

    • Fixed logout after restarting from the Danger Zone;

    • GSSD-1117: Added possibility to force creation of primary keys when target table is created through Gluesync;

  • Core Hub:

    • GSSD-1140: Fixed wrong result set values parsing when creating partitions with pivot columns of type Date and Time or Short;

  • Agents:

2.2.9.1

Released: July 1, 2026

  • UI:

    • Query Studio:

      • export create table statements;

      • copy create table statements;

      • run selected query statements within same query editor;

      • improved error handling;

    • Improved Chains support;

    • Improved entity detail;

  • Core Hub:

    • Fixed an issue with the logs retention settings;

    • Fixed an issue with the display of Last transaction message KPI;

2.2.9.0

Released: June 29, 2026

Release Highlights

Gluesync 2.2.9: The One Where the UI Got a Glow-Up and the Data Got a Workshop

Dear brave data wranglers,

Remember last time when we gave Gluesync a personality? Well, turns out personality is great, but what people really wanted was for it to also look the part — and to stop making them write code for things that should honestly just be a simple date conversion.

So we rolled up our sleeves, emptied the junk drawer, and rebuilt the entire Control Plane from the ground up. Then we kept going: new agents, a SQL workbench that lives inside your browser, a little something to tell your data exactly how to behave without writing a single line of Java, and — oh yeah — we put Gluesync in your pocket.

Buckle up. This is a big one.

What’s new (in order of how much they’ll change your daily life):

Control Plane UI: Same Gluesync, New Wardrobe

We took the entire Control Plane, gave it a fresh coat of paint, and then also rebuilt the engine while we were at it. Here’s what’s new:

  • An all-new pipelines list that doesn’t make you scroll until your finger files a complaint

  • A modern-looking entities list that actually respects your time (and your screen real estate)

  • Refined statuses across the board — because "is it working or not?" shouldn’t be a philosophical question

  • An all-new entity’s detail page with smooth status animations so you can watch your pipeline breathe in real-time

  • A serious performance boost, especially in loading times. We aim for real-time, and the UI should match that too.

It’s the same Gluesync you know and love, just wearing a tailored suit now.

Field Functions – Because Not Every Transformation Needs a PhD

You know that moment when you need to convert a date to a string, or trim some whitespace, and suddenly you’re writing a full UDF in Java just to do what is essentially a one-liner? Yeah, we fixed that.

Field Functions give you a library of pre-built, high-performance transformations you can apply directly at the field level — no code, no scripts, no existential crisis.

  • Date and time conversions (with sensible default patterns)

  • Numeric casting between decimals, ints, floats, doubles, and more

  • String trimming, byte array encoding, number-to-string

  • Technical fields: auto-generate timestamps, inject fixed constants

  • All configurable from the Fields Editor with a simple dropdown

It’s like giving your data a quick makeover before it walks out the door. See Field functions for the full catalog.

Pipeline Warm Up Status – Now You Know It’s Thinking, Not Dead

Ever stared at a pipeline that just started and wondered: "Is it broken? Is it loading? Should I panic?" We’ve all been there.

Enter the new warm up status. When a pipeline is first started or resumed, it now clearly tells you it’s warming up — initializing agents, loading configs, establishing connections — so you know things are under control while housekeeping activities are underway.

No more guessing. No more premature support tickets. Just a calm, reassuring "give me a few seconds, I’ve got this."

Universal File Store Agent – Your Data Lake’s New Best Friend

We built a new agent that can replicate data to basically any file storage you can think of. FTP? Sure. SFTP? Obviously. WebDAV, SMB, CIFS, NFS? All of them. We even brought our own snacks.

  • Writes data in Parquet (default), JSON, or CSV — selectable per pipeline

  • Supports FTP, FTPS, SFTP, WebDAV, WebDAVS, SMB/CIFS, and NFS

  • Parquet output is optimized for analytics workloads and BigQuery/Snowflake ingestion

  • Sidecar JSON metadata files for schema preservation and audit trails

  • Configurable concurrent uploads and row counts per file

Whether you’re building a data lake, archiving to a NAS, or feeding a legacy ETL that only speaks CSV — this agent has you covered. See Universal file store agent to get started.

IBM Db2 LUW CDC Agent – Transaction Logs, Without the Headache

IBM Db2 LUW already had a triggers-based agent, but some of you wanted more: richer transaction metadata, better performance, and that sweet, sweet CDC experience without relying on triggers.

So we built a brand new CDC agent that reads and decodes Db2’s transaction logs directly. Under the hood, we wrote a custom C native library (libdb2.so) that interfaces with Db2’s internal log structures to extract change events — this is invoked through a stored procedure (GS_SP_LOGMINER) that feeds the decoded changes into dedicated CDC tables.

  • Captures transaction IDs, timestamps, operation types, and session user names

  • Works with IBM Db2 for LUW 11.5+ (community edition included)

  • Supports both single-instance and partitioned database environments

  • Connection pooling, auto-reconnect, and configurable polling intervals

  • No complex transaction log configuration required

It’s CDC for Db2, the way it should have always been. See IBM Db2 LUW CDC agent for the full setup guide.

Query Studio – Your SQL Workbench, Now Living Inside Gluesync

You know the drill: you’re in Gluesync, something looks off, and you immediately tab over to DBeaver or pgAdmin to run a quick query. Context switching, connection strings, "wait which port was that again?" — it’s a whole thing.

Query Studio puts an end to that. It’s a full-featured SQL workbench embedded directly in the Control Plane, leveraging Gluesync’s existing knowledge of your database credentials so you can just…​ query.

And here’s the kicker: unlike your usual desktop SQL clients that live on your laptop and need a VPN tunnel the size of a garden hose just to reach a database in a private subnet — Query Studio is fully web-based. It runs inside Gluesync’s Core Hub, which is already connected to your databases. That means you can query remote databases that would otherwise be completely inaccessible without a mountain of VPN configs, firewall exceptions, and headaches. Open a browser, pick a connection, run your query. That’s it.

  • Multi-tab SQL editor with dialect-aware syntax highlighting and autocomplete

  • Interactive schema browser with lazy-loading tables and columns

  • Explain plan visualization (text or tree, depending on the database)

  • Real-time query progress streaming via WebSocket

  • Client-side statistics: min, max, mean, null counts, unique values

  • Export results as CSV, JSON, or SQL INSERT scripts

  • Saved queries with personal or team-wide visibility

  • Read-only mode by default to protect production data, with a writable mode toggle for authorized admins (SUPER ADMIN / MANAGER)

  • Query auditing with 90-day retention for compliance

  • Supports 16 database engines out of the box

It’s like DBeaver moved into your Gluesync apartment and started paying rent. See Query Studio for the grand tour.

Gluesync Mobile – Your Pipelines, Now in Your Pocket

You’re at dinner. Or on a train. Or pretending to pay attention in a meeting. And you get that nagging feeling: "Is the pipeline still running? Did that snapshot finish?" So you pull out your laptop, connect to the VPN, log in, and — oh wait, it’s fine. Great. Now close everything and try to enjoy your evening.

Not anymore. Gluesync Mobile is here — the first release of our native mobile companion app for the Control Plane. Everything you need to keep an eye on your replication pipelines, right from your phone.

  • Monitor pipeline statuses and entity health at a glance

  • Quick access to entity details, including current status and last sync information

  • Intuitive, mobile-first interface designed for on-the-go monitoring

  • Secure authentication with your existing Gluesync credentials

  • Available for both iOS and Android

It’s Gluesync, minus the laptop. Because sometimes you just need to know that everything’s fine — without having to go to the office to check it.

Public Roadmap – For the Community, by the Community

Ever wondered what we’re cooking next? Or had a feature idea and thought: "I wish I could just tell them"? Well, now you can — and you should.

We’ve opened up our public roadmap at roadmap.molo17.com. Here’s what you can do:

  • Vote on upcoming features to help us prioritize what matters most to you

  • Submit your own feature requests — if enough people upvote it, it goes on the radar

  • See what’s in progress, what’s planned, and what we’ve already shipped — full transparency, no guessing

This is your chance to shape Gluesync’s future. No gatekeeping, no black box — just a direct line from your idea to our sprint board. Head over to roadmap.molo17.com and make your voice heard.

PS. there’s a big new button to route you there directly from the Gluesync Control Plane!

Additional Improvements

  • Core Hub:

    • Field Functions: easily transform data via pre-built functions direcly at field level;

    • New REST endpoints for one-time-snapshot with "Start from scratch" capability;

    • Query Studio: run SQL queries within Gluesync with a full-web experience;

    • Improved telemetry reliability (which remains optional) also removing a known CVE caused by a legacy dependency;

    • New pipeline warming up status to help you knowing that things are under control while housekeeping activities are underway;

    • Minor bug fixes and improvements;

  • UI:

    • All new Control Plane pipeline root’s UI;

    • All new Control Plane entity’s detail page;

    • Groups within entities list table now persist your preference (collapsed, expanded), also supporting filtering by group;

    • Added ability for the admin and managers to reset users' password;

    • Notifications can now be filtered by area (e.g. filter by pipeline or entity);

    • New Roadmap button under the main top menu for a quick access to our public roadmap at roadmap.molo17.com — vote, request features, and see what’s coming next;

    • Minor bug fixes and improvements;

  • Mobile:

    • NEW Gluesync Mobile: first release of the mobile companion web app, with pipeline monitoring, and entity health at a glance;

  • Security (CVE):

    • Removed legacy dependency for Logback’s built-in Logstash and transitioned to a custom implementation (removing a critical CVE);

    • All third-party dependencies have been updated to mitigate several CVEs (thanks to all vendors who helped throughout the process, e.g. RavenDB);

    • Swapped Gluesync base layer to Fedora to improve security (Debian was not providing enough patches);

  • Agents:

2.2.8

2.2.8.13

Released: June 25, 2026

2.2.8.12

Released: June 24, 2026

  • Core Hub:

    • Fixed entity status not updated when One Shot Snapshot launched;

2.2.8.11

Released: June 23, 2026

2.2.8.10

Released: June 22, 2026

2.2.8.9

Released: June 18, 2026

  • Agents:

  • UI:

    • GSSD-1107: Included SnapshotWithoutTruncateThenCDC in snapshotWriteMethod mapping;

  • Core Hub:

    • GSSD-1110: Improve handling of blank collation names in queries;

    • GSSD-1008: Enhanced error handling for checkpoint file managers;

    • Fixed BigDecimal scale & precision filter logic in AffectedData.applyFilter;

    • Improved entity start/stop commands reliability also adding logs;

2.2.8.8

Released: June 16, 2026

  • Agents:

  • Core Hub:

    • GSSD-1101: Fixed connection exhaustion on internal database due to Mutex lock on metrics writes;

    • GSSD-1079: Fixed missing pipeline name in some notifications/metrics;

    • Added CoreHub reboot API;

    • Scheduler: Added user property to webhook payload and user filter to webhook configuration;

  • UI:

    • Fixed column nullability always false;

    • Fixed tooltip not working correctly;

    • Fixed technical fields columns sorting;

    • Removed update button from Modules Manager;

    • Fixed OTA updates error messages;

    • Restart Gluesync without Conductor (e.g. K8s users);

    • Notifications - Code Editor component loading issue when notification items are roughly above 15k;

    • Notifications - Empty selectable boxes in header;

  • Platform components:

    • Traefik (image molo17/traefik) bumped to version 3.7.4 (Linux only);

    • Prometheus (image molo17/prometheus) bumped to version 3.12.0 (Linux only);

    • Grafana (image molo17/grafana) bumped to version 13.0.2 (Linux only);

2.2.8.7

Released: June 15, 2026

2.2.8.6

Released: June 12, 2026

  • Agents:

  • Core Hub:

    • Fixed Metrics loss;

    • Fixed entity on error not retrying;

    • Added technical fields exclusions during CDC table setup across multiple database types;

    • GSSD-1064: Fixed entities not switching to CDC at snapshot completion;

2.2.8.5

Released: June 11, 2026

  • Agents:

  • Core Hub:

    • ClassLoader: Fixed possible use of an instance that is about to be shutdown;

    • Database connection cache: Fixed possible use of an instance that is about to be shutdown;

2.2.8.4

Released: June 10, 2026

  • Agents:

  • Core Hub:

    • Made data type match case insensitive for all the agents;

    • GSSD-1056: Fixed bulk data mirroring inconsistency for every database that has bulk feature;

    • GSSD-1071: Fixed high usage of ram by Prometheus Metrics and Notifications;

2.2.8.3

Released: June 10, 2026

2.2.8.2

Released: June 9, 2026

  • Core Hub:

    • Revised and enriched stats in Logs;

    • Fixed missing agents error notifications forwarding;

    • Minor bug fixes and improvements;

  • Agents:

  • Automator:

    • Bumped to version 1.4.0 with support for settings backup & restore, including (optionally) secrets.

  • UI:

    • GSSD-1066: Fixed an issue preventing an entity on error to be put into pause;

    • GSSD-1058: Fixed an issue that was causing the UI to corrupt decimal values metadata when editing an entity;

2.2.8.1

Released: June 8, 2026

2.2.8.0

Released: June 4, 2026

Release Highlights

Gluesync Just Got a Personality Upgrade (and Some Serious Superpowers)

Dear brave data wranglers,

We heard you. You wanted your replication tool to be less “quietly reliable in the background” and more “loud, proud, and actually tells you when it’s having a bad day.”

So we said: fine. Let’s give your data the emotional support it deserves.

Today we’re shipping a whole suite of “please don’t ignore me” features that turn Gluesync from a silent ninja into a slightly dramatic but extremely helpful friend.

What’s new (in order of how much they’ll save your weekend):

Smart Alert Thresholds – Because Your Data Has Feelings Too (beta)

We gave the anomaly detectors actual settings. Latency thresholds, row count drift, stall detection with proper warmup periods… basically therapy for your pipelines so they stop ghosting you at 3 a.m.

Notifications That Actually Reach You

Your pipelines can now scream at you via multiple channels. You’re welcome.

  • Email (proper SMTP client with all the bells and whistles)

  • Webhooks (CloudEvents, severity filtering, signing, custom headers — the whole fancy buffet)

  • Entities now reflect their actual status: healthy (the same "green" you’ve been used to), or error, yes, finally they display that status too in the case something bad happens to any of them.

Don’t worry, you can still choose your preferred method of interruption or snooze them entirely. Plus, for those who complained about a nuked mailbox: well, we’ve got you covered: emails (with a whole new style, you’re welcome) are burst-proof too.

Just that? Well, not really…​ There’s a whole new notifications hub waiting for you: clearer, faster, and much powerful.

PS. thanks Jesus!

OTA Updates – Now Actually Enjoyable

We completely rebuilt the Over-The-Air update experience from the ground up. No more staring at a vague spinner wondering if it’s dead or just thinking. You now get:

  • Crystal-clear progress tracking

  • Step-by-step breakdown per component

  • A real changelog for every single item

Basically, we turned software updates into something you don’t want to immediately cancel. As I always say: - "It’s like the Tesla but for data integration".

Global Settings – Now Less Chaotic, More Civilized

We gave the entire Settings tab a proper spring cleaning. Now everything lives in one logical place:

  • Reorganized navigation menu

  • More clarity and room for new upcoming features you’ll love

  • A preview of the new theme selector (Choose Light, Dark, or "whatever my OS is having an existential crisis about today" (System) mode)

Same settings. Way better organized. Much less "where the hell is that toggle again?"

Metrics That Don’t Suck

Enriched Prometheus endpoint with now more detailed metrics that will help you monitor your pipelines more effectively by hooking up on the same data sources where the new smart alert thresholds take data from. We will promise you that the Grafana dashboard will soon receive some love too.

Technical Columns – Long Requested, Finally Delivered

A most requested feature is here: rich technical metadata on all agents. Every replicated record now carries:

  • Transaction ID

  • Transaction Timestamp

  • User who modified the record

  • Operation type (INSERT / UPDATE / DELETE …)

  • And more juicy context

No more playing detective in your target database or worse, "Do I really have to write an UDF for that?!". Now, audit trails, debugging, and compliance just got way easier.

Additional Improvements

2.2.7

2.2.7.6

Released: June 3, 2026

2.2.7.5

Released: June 1, 2026

2.2.7.4

Released: May 29, 2026

2.2.7.3

Released: May 28, 2026

2.2.7.2

Released: May 28, 2026

2.2.7.1

Released: May 27, 2026

  • Core Hub:

    • Fixed an issue that could cause the loss of the Grafana bookmark from the main menu;

2.2.7.0

Released: May 25, 2026

Release Highlights

The Gates of the User Realm have been opened. The ancient prophecy has been fulfilled. OIDC has awakened. No more shall your users wander the forgotten lands of "password reset hell." They have identity. They have tokens. They have purpose.

New Features

User realm

Gluesync Core Hub now supports a multi-user environment with Role-Based Access Control (RBAC). Administrators can create multiple accounts, assign distinct roles (SUPER ADMIN, MANAGER, VIEWER), and manage platform security through an access control based on those roles. See User Management & RBAC for details.

OIDC, OpenID Connect

Core Hub now supports Single Sign-On (SSO) via OpenID Connect (OIDC), allowing users to authenticate through external identity providers such as Keycloak, Auth0, and Azure Entra ID. Role mappings from the identity provider can be synchronized with CoreHub RBAC roles. See OIDC Authentication for setup details.

More on the React UI transition

As we announced earlier this month, we’re pushing forward with the React UI transition. This release includes further UI refactors such us the main navigation menu, the sidebar, and the entire settings tab. We also took the chance to refresh our login page because…​ why not?

Additional Improvements

  • Core Hub:

    • User realm: a complete new feature which covers user management, role-based access control (RBAC), and integration with external identity providers via OpenID Connect (OIDC);

    • Session management and session tokens: improved session management and session token handling with customizable settings;

    • OIDC: setup your identity provider of choice and map external roles to Gluesync roles, quickly and automatically onboard new users in the platform;

    • Minor bug fixes and improvements;

  • UI:

    • New roles have also direct impact on the UI, allowing you to decide whether a user can actually made modifications to your control plane or not (a VIEWER user can only view the UI without the possibility to perform any edit action);

    • New and lightweight UI elements for an improved experience: topbar, sidebar, settings panel have all been rewritten from scratch;

    • Minor bug fixes and improvements;

  • Agents:

  • Modules:

    • Automator & Bootstrapper: have been updated to the new session token management system;

2.2.6

2.2.6.9

Released: May 22, 2026

2.2.6.8

Released: May 21, 2026

2.2.6.7

Released: May 19, 2026

  • Agents:

  • Core Hub:

    • Fixed notification write burst causing loss of notification;

    • Added mandatory resync check on entity columns configuration change;

  • UI:

    • Improved auto data type matching during table creation phase;

2.2.6.6

Released: May 18, 2026

2.2.6.5

Released: May 12, 2026

  • Agents:

  • Core Hub:

    • CSV Files management: Fixed an issue with the serialization of BigDecimal values;

    • GSSD-928: Fixed missing truncate on resync when a selective evition filter is applied;

    • GSSD-904: Fixed NoSQL Data Type migration issues;

    • GSSD-935:

      • Improved connection validation when the database is added to the pool;

      • Changed Hikari timeout values to prevent network high usage and TCP stack exhaustion;

2.2.6.4

Released: May 8, 2026

  • Agents:

  • Core Hub:

    • GSSD-929: Fixed an issue that was causing the Core Hub to not properly certain data types within "Selective evention of target table data" feature;

  • UI:

    • GSSD-925: Fixed an issue with UI cache preventing for a full metadata refresh when field lenght changes within the Field Editor screen;

    • GSSD-918: Fixed max length value in the Field Editor screen being displayed with initial default value of -1 when max was instead set at source;

    • Improved table creation, allowing no primary keys on target tables when creating new tables and no phisical key was actually present at source (all logical keys scenario);

2.2.6.3

Released: May 7, 2026

  • Agents:

  • Core Hub:

    • GSSD-898: Improved unsupported data types management.

    • GSSD-909: Fixed an issue with the _RRN column from AS400 that was causing a malformed Parquet file;

  • UI:

    • Fixed UI not redirecting to notification page properly when clicking over the notification snack bar;

2.2.6.2

Released: May 6, 2026

  • Agents:

  • Core Hub:

    • GSSD-887 made scheduler scope explicit and visible before save;

    • Fix wrong snapshot concurrency coerce

    • Various other scheduler improvements;

    • Fix incorrectly loaded JDBC with wrong ClassLoader;

2.2.6.1

Released: May 5, 2026

  • Core Hub:

    • GSSD-860: Fixed MySQL and MariaDB quoted identifier;

    • GSSD-908: Updated base JDK 21 docker image to solve a problem reading hardware info like RAM;

    • GSSD-909: Fixed parquet file corrupted when inserting decimal with scale zero (bumped ParquetKt to 1.0.8);

    • GSSD-911: Fixed problem with migration of entities.

2.2.6.0

Released: May 4, 2026

Release Highlights 🌟

This release focuses on some under-the-hood features plus the addition of new agents (RedShift, as a target, and SAP HANA as a source) and it brings performance improvements during write operation thanks for a reduced network round trips.

The upgrade process towards this release may take some time to complete. The process will display a "Bad gateway" message while it is loading, this is totally normal as the ingress proxy handles the 443 GETs from your client until Core Hub activates the Web Server back again at the end of the migration process.

New Features

Columns length, precision and scale

This version brings support for a refined discovery of columns length, precision and scale for all source and target databases. Allowing a more fine graned control over the data types configuration and inference, especially useful during operations such as table creation, data type mapping, and schema synchronization (e.g. Parquet files).

SAP HANA agent (source)

SAP HANA is now available as a source agent, bringing support for real-time data integration from SAP HANA databases using our triggers-based auditing structure.

RedShift agent (target)

RedShift is now available as a target agent, with support for Bulk operations via AWS S3 as staging area (required) and batch operations for transactional writes.

Additional Improvements

2.2.5

2.2.5.6

Released: April 30, 2026

2.2.5.5

Released: April 29, 2026

2.2.5.4

Released: April 24, 2026

2.2.5.3

Released: April 23, 2026

2.2.5.2

Released: April 17, 2026

  • UI:

    • Fixed all columns checkbox logics;

    • Fixed columns data type mismatch during create table;

  • Agents:

2.2.5.1

Released: April 16, 2026

2.2.5.0

Released: April 15, 2026

Release Highlights 🌟

This version brings several performance improvements to all RDBMS sources including expanded query pre-fetch size where it matters the most (Snapshots!), configurable query parallelism and reduced round-trip operations. LogMiner agent has been improved to support larger transactions in parallel and better error handling.

Source agents that rely on our Triggers-auditing structure are also receiving some love with this release, starting with Oracle triggers agent first, then all others following suit with features like enhanced indexes and improved overall transaction confirmation flow to lower network footprint.

This release includes also the new SAP HANA target agent (with its source counterpart following suit in one of the next few releases), plus a bunch of improvements and bug fixes.

New Features

RDBMSses Snapshot performances improvements

This version brings several performance improvements to all RDBMS sources including expanded query pre-fetch size where it matters the most (Snapshots!), configurable query parallelism and reduced round-trip operations. Query parallelism is now set to 4 by default, but can be adjusted in the agent’s configuration. Pre-fetch size is now set to 10000 by default (from the original 1000), also adjustable in the agent’s configuration.

SAP HANA agent (target)

SAP HANA is now available as a target agent, with soon to come support as a source.

Additional Improvements

2.2.4

2.2.4.9

Released: April 13, 2026

  • Core Hub:

    • GSSD-778: Fixed only last transaction rows are written in Parquet files;

2.2.4.8

Released: April 13, 2026

2.2.4.7

Released: April 10, 2026

  • Core Hub:

    • Fixes on custom path for CSV/Parquet files;

    • Fixed an issue where the metrics endpoint could send objects ids instead of object names after a Core Hub restart;

    • GSSD-800 & GSSD-778: Fixed an issue that could prevent the read access to Chronicle Queues' metadata on Windows systems;

  • UI:

    • Removed Chains support from databases where Chains are not supported (e.g., IBM i Series);

  • Agents:

2.2.4.6

Released: April 9, 2026

  • Core Hub:

    • UDFs stability improvements;

2.2.4.5

Released: April 9, 2026

  • Core Hub:

    • Performance improvements for UDFs (2-3x faster in typical workloads);

    • GSSD-800 & GSSD-778: Improvements to mitigate an issue that could prevent the read access to Chronicle Queues' metadata on Windows systems;

  • Agents:

2.2.4.4

Released: April 7, 2026

2.2.4.3

Released: April 3, 2026

2.2.4.2

Released: April 2, 2026

2.2.4.1

Released: April 2, 2026

2.2.4.0

Released: April 1, 2026

Release Highlights 🌟

With this release we’re introducing the new unicorns and rainbows mode: as soon as you hit 1M rows synchronized per second, you’ll be granted access to the shiny new unicorns and rainbows mode, isn’t that cool?

We’re kidding! It’s April fools!!! :-D

Now that we’ve got your attention, here’s what’s new in this release:

New Features

IBM i Journal Reading Improvements

For IBM i series users, it’s a big day: the engineering team has been working hard to bring you the best experience possible when it comes for reading the journal from your beloved IBM i systems. With this release, we’re transitioning from the usage of the "classic" DISPLAY JOURNAL command to a new approach that leverages the power of the IBM i API’s to read the journal more efficiently and with less overhead.

Before updating we suggest to review the complete list of permissions required for the new APIs under the IBM i agent’s page.

Snapshot duration report in notifications

Yep, we’ve added a new field to the notifications to show the duration of the snapshot phase. So you can now see how long the snapshot took to complete in the notifications, pretty useful for monitoring the historical performance of your syncs (GSSD-590).

Data lakes additions

Data lakes agents now support a custom root path configuration (GSSD-779), allowing you to specify where the data should be stored. Plus for CSV files, a new technical field has been added as unique row identifier (GSSD-780).

Additional Improvements

2.2.3

2.2.3.2

Released: March 31, 2026

  • Core Hub:

    • Database pool technology has been extended to all database types;

    • It is now possible to configure the minimum connections pool size, default to 5;

    • GSSD-778: Fixed an issue that could cause the internal agent’s cache to fail to acquire a write lock when the underlying disks have insufficient I/O performance;

  • UI:

    • Bug fixes and improvements;

  • Agents:

2.2.3.1

Released: March 30, 2026

  • Core Hub:

    • Applied safeguards on connection pool management to prevent resource exhaustion:

      • MAX CONNECTION count is now set at 25 connections per agent, as a sefeguard measure (from the previous 100);

  • Agents:

2.2.3.0

Released: March 23, 2026

Release Highlights 🌟

This release includes some highly requested features to streamline email alerts configuration and delivery, additional logs settings now configurable from the UI, and A LOT of bug fixes and improvements starting from a reduced CPU footprint and performance enhancements, including important enhancements on multi-pipeline deployments. Here following a comprehensive list of changes:

New Features

Multi-pipeline performance improvements

Improved performance and resource utilization for multi-pipeline deployments: we’ve introduced a new scheduling algorithm that optimizes resource allocation and reduces CPU overhead, resulting in more efficient parallel processing across multiple pipelines. If you have multiple pipelines belonging to a same source/target database with same credentials (even though they might be pointing to different instances), the system will automatically reuse already initialized connections and resources, significantly reducing overhead and improving throughput (this feature currently applies to Oracle, MS SQL Server and Informix database types, more database types will be added in future releases).

Email alerts and logging settings

You can now configure email alerts and advanced logging settings directly from the UI, making it easier to setup an SMTP server and configure logging retention and rotation.

Additional Improvements

2.2.2

2.2.2.3

Released: March 24, 2026

2.2.2.2

Released: March 23, 2026

  • Parquet management improvements:

    • We bumped our Parquet library to the latest version (1.0.4), bringing better performance and compatibility, with a focus on squashing OOM errors experienced under heavy load of concurrent large-scale operations;

  • Agents:

  • UI:

    • Fixed a compatibility issue with some old browsers not displaying the UI correctly;

    • Fixed an issue that could prevent a snapshot task from being created if the entity had a failed CDC attempt enqueued;

    • Bug fixes and improvements.

2.2.2.1

Released: March 20, 2026

  • Agents:

    • Oracle target agents now support Bulk loads;

  • UI:

    • Fixed an issue where the UI would not properly display the status of the entity last metrics if no data is flowing;

2.2.2.0

Released: March 19, 2026

Release Highlights 🌟

CSV files support for Datalake target agents such as AWS S3, Google Cloud Storage, and Azure Blob Storage, UI enhancements for better user experience and the ability to schedule the pipeline maintenance mode are all part of this release.

New Features

Schedule pipeline maintenance mode

With the latest update, you can now schedule pipeline maintenance mode to temporarily pause data synchronization for routine maintenance, updates, or troubleshooting without losing connection settings or historical data.

CSV files support for Datalake targets

CSV files support for Datalake target agents such as AWS S3, Google Cloud Storage, and Azure Blob Storage.

Additional Improvements

  • Core Hub:

    • GSSD-627 & GSSD-759: Fixed snapshot completion not triggering CDC once the snapshot is done;

  • Agents:

  • UI:

    • Major UI improvements and bug fixes;

  • Installers:

    • Windows installers now include more stronger error handling and logging especially for endpoints reachability issues;

    • Linux installers follow the same pattern;

    • Linux installer now include support for Oracle Linux;

    • Offline installers for Linux now available (please contact support separately for download, it is an early access feature);

  • Chronos:

    • You can now schedule maintenance windows for pipelines;

  • Automator:

    • Added support for scheduling pipeline maintenance windows via Chronos APIs;

  • Bootstrapper:

    • Added support for scheduling pipeline maintenance windows via Chronos APIs;

2.2.1

2.2.1.2

Released: March 18, 2026

  • UI:

    • Major UI improvements and bug fixes;

    • Initial splash screen UX improvements;

2.2.1.1

Released: March 18, 2026

2.2.1.0

Released: March 17, 2026

Release Highlights 🌟

This version brings maintenance mode for pipelines, allowing you to temporarily pause data synchronization for routine maintenance, updates, or troubleshooting without having to be constantly notifed about known errors and warning messages for disconnected databases.

New Features

Pipeline maintenance mode

Added maintenance mode for pipelines to temporarily pause data synchronization for routine maintenance, updates, or troubleshooting without losing connection settings or historical data.

Pipeline reset

A pipeline can now be reset to clear checkpoint data and start synchronization from scratch. This is useful when you want to reprocess data or correct synchronization issues. You can access it from within the settings tab under Danger zone.

Minor improvements

2.2.0.0

2.2.0.9

Released: March 16, 2026

  • Core Hub:

    • For agents supporting local cache we made improvements to the Rotation and Eviction mechanisms, especially for Windows users;

  • UI:

    • Improvements to agent deployment (less clicks required);

    • Fixed back behavior;

  • Agents:

2.2.0.8

Released: March 13, 2026

2.2.0.7

Released: March 12, 2026

  • Core Hub:

    • GSSD-707: Fixed an issue that could cause a completed Snapshot to never transition to the "Completed" state;

    • GSSD-641: Fixed an issue that was causing UDP broadcast to flood the subnet in certain Linux distributions with recent Linux kernel updates;

    • Enhanced TLS certs support during upload and management phases;

  • Agents:

2.2.0.6

Released: March 10, 2026

2.2.0.5

Released: March 10, 2026

2.2.0.4

Released: March 10, 2026

2.2.0.3

Released: March 9, 2026

2.2.0.2

Released: March 9, 2026

2.2.0.1

Released: March 5, 2026

  • Core hub:

    • Fix restart entities with resync option enabled;

2.2.0

Released: March 5, 2026

Upgrading from 2.1.x to 2.2.x may require several minutes to complete especially on Windows-based environments, plan this upgrade on a maintenance window. A "fail to fetch" error at the end of the upgrade process is expected and can be safely ignored. Just refresh the browser and you’ll be good to go after a new login.

2.2 is the second major update to Gluesync 2 since its initial release one and a half years ago. It changes the way we connect agents by introducing a new plugin-based system which makes it easier to manage and scale your data integration platform under heavy pipelines scenarios; shifting from a WebSocket-based connection to a plugin-based system for agents allows to create more room for things that matter: performances, reliability, and ease of use. Optimization is paramount in MOLO17 and we’re committed to delivering the best possible experience to our users, that’s why today more than ever, quoting, "best part, is no part".

As per our iterative agile approach to data integration, we’re continuously and relentlessly improving and enhancing the platform with new features and capabilities. 2.2 includes a set of features that will be gradually released in the upcoming releases such as:

  • Pipeline reusable functions (similar to UDFs, but simpler and reusable across multiple entities);

  • Enhanced security features such as user realm management and OIDC client authentication;

  • New frontend for better user experience, based out of React;

  • Third-party APIs for seamless integration with other systems, as targets;

  • Enhanced monitoring and observability with new metrics and dashboards;

and much more to come.

Release Highlights 🌟

Agents are part of the major overhaul packed in this initial 2.2.0 release: plugin-based system for easier management and scaling, shifting from the WebSocket per-agent connection.

The agents deployment UX you all loved won’t change, we’re just making it faster, more reliable, and more scalable. You’ll immediately notice the difference in performance and reliability.

The upgrade process from legacy 2.1 agents to new plugin-based agents will be carried out by our in-platform orchestrator Conductor, which will take care of decommissioning the legacy 2.1 agents and their connections, cleaning up your environment after the upgrade process is complete.

New Features

Performance improvements

The core aspects we focused on in this 2.2.0 release are bringing you a more performant and reliable platform, with less memory usage and faster processing times.

New Windows Server MSI installer

We’ve worked hard to make the installation process as smooth as possible, and we’re excited to introduce a new Windows Server MSI installer that makes it easier than ever to deploy and run Gluesync.

Minor improvements