ドキュメント変更履歴 2026-08-12

--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/about+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/about@@ -39,5 +39,5 @@ ## Licensing models and critical constraints -The %oracleofc% supports two distinct licensing models. You must select the correct+The %oracleofc% supports three distinct licensing models. You must select the correct model before installation. Failure to select the correct model might result in deployment failure or unintended financial commitments.@@ -53,10 +53,12 @@ terms. -### 1. Embedded License (Snowflake-provided)+### 1. Embedded license for 36-month commitment (Snowflake-provided)++In Openflow, this license option is displayed as **Oracle Embedded License**.  Snowflake provides the Oracle XStream license to you directly for a fee. This model allows you to consume XStream replication without a direct contract with Oracle. For more information, see-[Embedded license details](#label-oracle-embedded-license-details) and the+[Embedded license details for 36-month commitment](#label-oracle-embedded-license-details-36) and the [Openflow Connector for Oracle Addendum](https://www.snowflake.com/en/legal/optional-offerings/offering-specific-terms/openflow-oracle-terms/). @@ -78,5 +80,5 @@       <td> -License and Support & Maintenance (S&M) fees are drawn from your Snowflake+Monthly license fees and Support & Maintenance (S&M) fees are drawn from your Snowflake Capacity. @@ -92,6 +94,6 @@  - **Post-term (36+ months)**: After the initial 36-month term, the license fee-  drops to $0, but the S&M fee continues annually.-- **Lock-out risk**: If you opt-out of S&M renewal, the connector will be+  drops to $0, but the S&M fee auto-renews in 12-month increments, billed monthly.+- **Lock-out risk**: If you opt out of S&M renewal, the connector will be   permanently locked when S&M coverage ends. Unlocking the connector requires   purchasing a new Embedded License, which triggers a new 36-month commitment@@ -117,5 +119,4 @@ The following customers are ineligible: -- Public sector entities. - Customers purchasing Snowflake through the GCP Marketplace. - Customers contracted with Snowflake through a third-party reseller.@@ -127,9 +128,13 @@ </div> -### 2. Independent license (Bring Your Own License - BYOL)--You provide your own Oracle license that includes XStream entitlements (for example, Oracle-GoldenGate license). For more information, see-[Independent license (BYOL) details](#label-oracle-byol-license-details).+### 2. Embedded license for 12-month commitment (Snowflake-provided)++In Openflow, this license option is displayed as **Oracle Embedded License (Public Sector)**.++Snowflake provides the Oracle XStream license to you directly for a fee. This model+allows you to consume XStream replication without a direct contract with Oracle.+For more information, see+[Embedded license details for 12-month commitment](#label-oracle-embedded-license-details-12) and the+[Openflow Connector for Oracle Addendum](https://www.snowflake.com/en/legal/optional-offerings/offering-specific-terms/openflow-oracle-terms/).  <div className="colwidths-given">@@ -148,4 +153,79 @@     <tr>       <td>Billing</td>+      <td>++One-time license fee and annual Support & Maintenance (S&M) fees are drawn from+your Snowflake Capacity.++</td>+    </tr>+    <tr>+      <td>Commitment</td>+      <td>++Activation initiates a non-cancelable 12-month term (after the 60-day trial), with+license fees paid upfront in full and S&M fees billed annually.++</td>+    </tr>+    <tr>+      <td>Lifecycle</td>+      <td>++- **Post-term (12+ months)**: After the initial 12-month term, the license fee+  drops to $0, but the S&M fee auto-renews in 12-month increments, billed annually.+- **Lock-out risk**: If you opt out of S&M renewal, the connector will be+  permanently locked when S&M coverage ends. Unlocking the connector requires+  purchasing a new Embedded License, which triggers a new 12-month commitment+  at full price.++</td>+    </tr>+    <tr>+      <td>Management UI</td>+      <td>++All license actions (Start/Cancel Trial, Monitor Usage, Opt-out) are performed+by the ORGADMIN in %sf-web-interface% under+**Admin** %raa% **Terms** %raa% **Openflow for Oracle**.+For step-by-step instructions, see [](/user-guide/data-integration/openflow/connectors/oracle/manage-commercial-terms).++</td>+    </tr>+    <tr>+      <td>Restrictions</td>+      <td>++The following customers are ineligible:++- Customers purchasing Snowflake through the GCP Marketplace.++</td>+    </tr>+  </tbody>++</div>++### 3. Independent license (Bring Your Own License - BYOL)++You provide your own Oracle license that includes XStream entitlements (for example, Oracle+GoldenGate license). For more information, see+[Independent license (BYOL) details](#label-oracle-byol-license-details).++<div className="colwidths-given">++  <colgroup>+    <col style={{width: "20.0%"}} />+    <col style={{width: "80.0%"}} />+  </colgroup>+  <thead>+    <tr>+      <th>Term</th>+      <th>Details</th>+    </tr>+  </thead>+  <tbody>+    <tr>+      <td>Billing</td>       <td>No additional licensing fees from Snowflake. Standard storage and compute costs (for example, Openflow Compute) will apply.</td>     </tr>@@ -156,5 +236,5 @@     <tr>       <td>Usage</td>-      <td>Mandatory for public sector, GCP Marketplace, and reseller customers.</td>+      <td>Mandatory for GCP Marketplace customers.</td>     </tr>   </tbody>@@ -166,8 +246,9 @@ ## Choosing an Oracle XStream licensing model -The %oracleofc% requires a paid license for Oracle XStream services. Two licensing+The %oracleofc% requires a paid license for Oracle XStream services. Three licensing models are available: -- Embedded Oracle License+- Oracle Embedded License (36-month commitment)+- Oracle Embedded License (12-month commitment) - Independent Oracle License (Bring Your Own License - BYOL) @@ -177,12 +258,14 @@    <colgroup>-    <col style={{width: "20.0%"}} />-    <col style={{width: "40.0%"}} />-    <col style={{width: "40.0%"}} />+    <col style={{width: "18.0%"}} />+    <col style={{width: "27.0%"}} />+    <col style={{width: "27.0%"}} />+    <col style={{width: "28.0%"}} />   </colgroup>   <thead>     <tr>       <th>Consideration</th>-      <th>Embedded License</th>+      <th>Oracle Embedded License (36-month commitment)</th>+      <th>Oracle Embedded License (12-month commitment)</th>       <th>Independent License (BYOL)</th>     </tr>@@ -193,6 +276,13 @@       <td> -Customers who need to license Oracle XStream technology and want to purchase it-directly through their Snowflake agreement.+Customers who need to license Oracle XStream technology directly through their+Snowflake agreement and prefer a monthly payment schedule over 36 months.++</td>+      <td>++Customers who need to license Oracle XStream technology through their Snowflake+agreement but can't agree to a multi-year commitment or prefer an upfront+payment option instead of monthly payments.  </td>@@ -208,7 +298,17 @@       <td> -Billed through Snowflake based on the number of processor cores on your source-Oracle DB. Involves a non-cancelable 36-month commitment. Also billed for support-and maintenance services.+Billed monthly through Snowflake based on the number of processor cores on your+source Oracle DB. Involves a non-cancelable 36-month commitment. Also billed+monthly for support and maintenance services.++Additionally, standard storage and compute costs (for example, Openflow Compute) will apply.++</td>+      <td>++Billed upfront through Snowflake based on the number of processor cores on your+source Oracle DB. Involves a non-cancelable 12-month commitment where all+license fees are paid upfront in full. Also billed ... (truncated) 

差分が長いため、途中まで表示しています。

--- ahttps://docs.snowflake.com/en/migrations/snowconvert-docs/general/release-notes/release-notes/README+++ bhttps://docs.snowflake.com/en/migrations/snowconvert-docs/general/release-notes/release-notes/README@@ -2,4 +2,213 @@  The SnowConvert AI tool is subject to the [Conversion Software Terms of Use](https://www.snowflake.com/en/legal/technical-services-and-education/conversion-software-terms/).++## Version 2.40.0 (Aug 11, 2026)++### CLI++#### New Features++- Added `scai code convert --list-settings` command and `recommend-settings` plugin sub-skill.+- Added `scai data list` command showing connection and database information with affinity filtering.+- Added `scai test etl-validate --pipeline` regex filter for selective ETL pipeline validation.+- Added `scai assessment report` command for migration assessment report generation.+- Added `scai project info` command that reports local config without requiring a Snowflake connection.+- Added worker primitives with local worker `--no-server` mode and SPCS `--service-name` support.+- Added parallelism support for Code Deployment.++#### Improvements++- Shortened DMO and DAE virtual environment paths on Windows to avoid `MAX_PATH` limitations.+- Improved natural-key lookup joins to value-align joins that have no foreign key edge.++#### Bug Fixes++- Fixed an issue where Spectre markup characters in terminal output were not properly escaped.+- Fixed a mismatch between the doctor's schema validator and DEW config format.++### Conversion Engine++#### New Features++##### Oracle++- Added emulation of cursor `%ISOPEN` as a manual `BOOLEAN` flag instead of emitting SSC-EWI-OR0128.+- Added translation of single-column `BULK COLLECT INTO` a dense collection to `SELECT ARRAY_AGG(col) INTO :arr`.+- Added translation of provably-safe `JSON_TABLE` subset to Snowflake `FLATTEN`.+- Added translation of local associative arrays using the `OBJECT` type.+- Added resolution of unqualified cross-schema object references through unique synonym fallback.+- Added combined OverallProcedure and OverallFunction Assessment CSV output family.++##### SQL Server++- Added routing of `ALTER TABLE` masking policies to `masking_policies.sql`.++##### Teradata++- Added parsing of positional bind parameters (`:1`, `:2`).++##### Druid++- Added registration of Druid query code units in the Code Unit Registry.++##### DataStage++- Added DataStage `.pjb` (ISX PJB) parser frontend for processing DataStage jobs.++##### General++- Added registration of masking policy objects in the Code Unit Registry.++#### Bug Fixes++##### Oracle++- Fixed parsing of schema-qualified sequence `CURRVAL`/`NEXTVAL` so the EWI correctly fires.+- Fixed parsing and marking of `CREATE TABLE` partitioning clauses and DB links.+- Fixed `UNPIVOT IN`-list labels to carry across as identifier aliases.+- Fixed emitting a supported exception code for `RAISE_APPLICATION_ERROR`.+- Fixed routing of record/`%ROWTYPE` field `:= NULL` through `OBJECT_INSERT`.+- Fixed translation of structured UDT assignments through the `OBJECT` pipeline.+- Fixed a `MultiReplace` crash on labeled `OPEN FOR` OUT cursor.+- Fixed parsing of `bool`, `file`, `specific`, `uid`, `empty`, `global`, and `clone` as unquoted identifiers.+- Fixed suppression of SSC-EWI-0094 false positive for loop labels.++##### SQL Server++- Fixed conversion of `FOR XML PATH('') + STUFF` idiom to `LISTAGG`.+- Fixed renaming of scalar SQL UDF parameters that collide with body columns.+- Fixed table-valued parameter and `RETURNS TABLE` identifiers.+- Fixed reformatting of packed date literals compared against date columns.++##### Teradata++- Fixed recognition of SQL Assistant `?NAME` client-side named parameters.+- Fixed removal of `NULL CALL` clause on SQL UDFs with SSC-FDM-TD0059 advisory.++##### Druid++- Fixed emitting the aggregate expression in a native-query `groupBy` `HAVING` clause.++##### SSIS++- Fixed `SanitizeEmptyElements` incorrectly deleting data-carrying elements, which caused packages to be silently dropped.++##### Redshift++- Fixed various conversion issues identified through systematic verification.++##### BigQuery++- Fixed various conversion issues identified through systematic verification.++##### Sybase++- Fixed various conversion issues identified through systematic verification.++##### General++- Improved PK/FK column reference binding to their definition name through the dialect key factory.+- Fixed symmetric PK-to-PK inferred foreign keys to emit at most one canonical direction.+- Fixed various cross-platform conversion issues identified through systematic verification.++### Data Validation++#### New Features++- Added "Cloud-Direct" extraction strategy which writes data files directly to cloud storage services instead of writing to disk first.+- Added Iceberg v2-compatible L1 and L3 profile support for data validation.+- Added Iceberg DEA Parquet packaging with codec support, typed timestamps, and UTC homogenization.+- Added support for geographic and geometry data types for BigQuery data migration and validation.+- Added support for geographic and geometry data types for Redshift data migration and validation.+- Added in-warehouse L3 set-difference gate with multiplicity and `MINUS` support.+- Added Snowflake key inference with inline resolve for in-warehouse data validation.+- Added orchestrator-side L1 schema comparison for the in-warehouse lane.+- Added orchestrator-side L2 metrics comparison for the in-warehouse lane.+- Added in-warehouse validation routing.+- Added support for BigQuery `ARRAY<INT64>` and parameterized `STRUCT` in DMO partition strategy.++#### Improvements++- Improved initial partition sync to fix watermark issues.+- Improved handling of `initialWaterMark` when incremental strategy is switched.+- Improved early failure reporting when `trackModifications=true` is set with Snowpipe as the ingestion mechanism.+- Adjusted parameterization of data types for cases in which tables are created from scratch.+- Improved Teradata incremental source isolation and baseline diagnostics.+- Improved extraction of `OdbcConnectionStringBuilder` across all ODBC dialect configs.+- Clustered `CELL_VALIDATION_RESULTS` and `ROW_VALIDATION_RESULTS` tables for improved query performance.++#### Bug Fixes++- Fixed preservation of zero-valued `ordinal_position` and `scale` in `read_schema_from_stage`.+- Fixed target-only columns now being flagged as L1 schema failures.+- Fixed `NaT` sentinel value replacement with regular `None` before converting to Polars DataFrame.+- Fixed a race condition related to the task chain "Extract schema" to "Determine Partition Strategy".+- Fixed `Decimal` zero being rendered as scientific notation in the data types harness.+- Fixed "Clear partition" tasks failing with case-insensitive partition keys.+- Fixed cell early-stop combined with AT and Snowpipe DV configuration.+- Fixed a race condition in the `COMPLETE_TASK` definition for complex task chains.++### Others++#### Testing Framework++##### New Features++- Added coverage collector with history fetch for test execution tracking.+- Added support for SQL Server `DATETIMEOFFSET` (ODBC type -155) in baseline capture.+- Added auto-sorting of result sets for order-independent TVF validation.++##### Improvements++- Made `model-config.toml` the source of truth for lane models in `scai test seed`.+- Annotated empty results with `vacuous_match` for clearer test diagnostics.+- Replaced generic error with a descriptive message when validate selects zero tests.+- Improved target database, schema, and task name resolution from the CUR.+- Improved SSIS package folder, project, and store resolution from the source server.+- Improved SQL Server identifier handling by unquoting at the point of use instead of in the seed.+- Improved reporting of unusable row-comparison index key... (truncated) 

差分が長いため、途中まで表示しています。

--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/manage-commercial-terms+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/manage-commercial-terms@@ -56,10 +56,11 @@ Complete the steps for the license model you selected during configuration: -- [Option A: Embedded license (Snowflake-provided)](#label-oracle-embedded-license-setup)-- [Option B: Independent license / BYOL](#label-oracle-byol-license-setup)--<a id="label-oracle-embedded-license-setup"></a>--### Option A: Embedded license (Snowflake-provided)+- [Option A: Embedded license for 36-month commitment (Snowflake-provided)](#label-oracle-embedded-license-setup-36)+- [Option B: Embedded license for 12-month commitment (Snowflake-provided)](#label-oracle-embedded-license-setup-12)+- [Option C: Independent license / BYOL](#label-oracle-byol-license-setup)++<a id="label-oracle-embedded-license-setup-36"></a>++### Option A: Embedded license for 36-month commitment (Snowflake-provided)  For this licensing model, you must activate the trial to enable the connector.@@ -133,6 +134,6 @@     <tr>       <td>Post-term S&M renewal (after month 36)</td>-      <td>The license fee drops to $0. The annual Support & Maintenance (S&M) fee continues. You may opt out of S&M renewal in the **Openflow for Oracle** dashboard.</td>-      <td>If you opt out and S&M coverage expires, the connector is permanently locked. To resume, you must purchase a new embedded license, which resets the 36-month commitment.</td>+      <td>The license fee drops to $0. The Support & Maintenance (S&M) fee auto-renews in 12-month increments, billed monthly. You may opt out of S&M renewal in the **Openflow for Oracle** dashboard.</td>+      <td>If you opt out and S&M coverage expires, the connector is permanently locked. To resume, you must purchase a new Embedded License, which resets the 36-month commitment.</td>     </tr>   </tbody>@@ -140,7 +141,88 @@ </div> +<a id="label-oracle-embedded-license-setup-12"></a>++### Option B: Embedded license for 12-month commitment (Snowflake-provided)++For this licensing model, you must activate the trial to enable the connector.++Even if you install the connector, data replication doesn't start until this step is complete.++#### Step 1: Start the trial (prerequisite)++To start the trial:++1. Sign in to %sf-web-interface-link%.+2. In the navigation menu, select **Admin** %raa% **Terms**.+3. Select **Openflow for Oracle** tab.+4. Locate the **Trial Status** card (status: "Ready to Activate").+5. Select **Start Trial**.+6. Accept the terms to start the 60-day trial period.++This action enables the captureChangeOracle processor, allowing it to connect to+your database.++#### Step 2: Configure connector++After starting the trial, install and configure the connector. For more information,+see [Configure the connector](/user-guide/data-integration/openflow/connectors/oracle/setup-connector).++After the connector successfully connects to the source database, a subscription is+automatically created and displayed in the **Openflow for Oracle** dashboard.++#### Step 3: Verify inventory++1. Sign in to %sf-web-interface-link%.+2. In the navigation menu, select **Admin** %raa% **Terms**.+3. Select the **Openflow for Oracle** tab.+4. Review the **Subscription Inventory** section.+5. Verify that the CPU core count matches your physical source database hardware.+6. If the core count is incorrect, update the runtime configuration.++#### Step 4: Lifecycle management++For more information about the licensing models and terms, see+[Licensing models and critical constraints](#label-oracle-licensing-models).++The following table describes the actions available at each stage of the embedded+license lifecycle.++<div className="colwidths-given">++  <colgroup>+    <col style={{width: "20.0%"}} />+    <col style={{width: "30.0%"}} />+    <col style={{width: "50.0%"}} />+  </colgroup>+  <thead>+    <tr>+      <th>Stage</th>+      <th>Action</th>+      <th>Result</th>+    </tr>+  </thead>+  <tbody>+    <tr>+      <td>Trial period (Day 1 to 60)</td>+      <td>Select **Cancel Trial** in the **Openflow for Oracle** dashboard before Day 60.</td>+      <td>Oracle XStream services stop. No charges are incurred.</td>+    </tr>+    <tr>+      <td>12-month commitment (Day 61+)</td>+      <td>No action required. If the trial isn't canceled, the non-cancelable 12-month term begins automatically on Day 61, with license fees paid upfront in full.</td>+      <td>The license can't be canceled during this period. If your Snowflake agreement is terminated, the full remaining balance is due immediately.</td>+    </tr>+    <tr>+      <td>Post-term S&M renewal (after month 12)</td>+      <td>The license fee drops to $0. The Support & Maintenance (S&M) fee auto-renews in 12-month increments, billed annually. You may opt out of S&M renewal in the **Openflow for Oracle** dashboard.</td>+      <td>If you opt out and S&M coverage expires, the connector is permanently locked. To resume, you must purchase a new Embedded License, which resets the 12-month commitment.</td>+    </tr>+  </tbody>++</div>+ <a id="label-oracle-byol-license-setup"></a> -### Option B: Independent license / BYOL+### Option C: Independent license / BYOL  If you are using the independent license (Bring Your Own License), no prior trial activation@@ -162,4 +244,4 @@  The **Start Trial** button doesn't appear for this license model, and the-36-month lifecycle rules don't apply. You are responsible for maintaining a valid-Oracle license that includes XStream entitlements.+Embedded License commitment-period rules don't apply. You are responsible for+maintaining a valid Oracle license that includes XStream entitlements. 

--- ahttps://docs.snowflake.com/en/sql-reference/account-usage/block_storage_history+++ bhttps://docs.snowflake.com/en/sql-reference/account-usage/block_storage_history@@ -9,4 +9,6 @@  # BLOCK_STORAGE_HISTORY view++%spcs% is available to accounts in [AWS, Microsoft Azure, and Google Cloud Platform commercial regions](#label-na-general-regions), with some exceptions. For more information, see [](#label-snowpark-containers-overview-available-regions).  Use the BLOCK_STORAGE_HISTORY view in the ACCOUNT_USAGE schema to query the average daily block storage and snapshot usage for an account within the last 365 days.@@ -45,4 +47,9 @@     </tr>     <tr>+      <td>COMPUTE_POOL_ID</td>+      <td>NUMBER</td>+      <td>Internal, Snowflake-generated identifier of the compute pool associated with this storage usage. For the `SNAPSHOT` storage type, this field is NULL.</td>+    </tr>+    <tr>       <td>BYTES</td>       <td>NUMBER</td>@@ -67,5 +74,5 @@ - Latency for the view can be up to 180 minutes (3 hours). - The view provides daily block storage and snapshot usage within the last 365 days (1 year) for an account.-- Snapshots are not associated with compute pools; therefore, for snapshots the view has the NULL value in the COMPUTE_POOL_NAME column.+- Snapshots are not associated with compute pools; therefore, for snapshots the view has the NULL value in the COMPUTE_POOL_NAME and COMPUTE_POOL_ID columns. - The BYTES column shows average usage of block storage volumes for a specific day, for a specific storage type, and for a specific compute pool (where appropriate) in the Snowflake account. For example, consider the following: @@ -85,11 +92,11 @@    ```text-  +-------------------------------+--------------------+-------------------------+----------------+-----------------------+-----------------------------+-  | USAGE_DATE                    | STORAGE_TYPE       | COMPUTE_POOL_NAME       |       BYTES    |       ADDITIONAL_IOPS |       ADDITIONAL_THROUGHPUT |-  |-------------------------------+--------------------+-------------------------+----------------|-----------------------|-----------------------------|-  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_1                  | 2,684,354,560  | 250.000000000         | 25.000000000                |-  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_2                  | 5,368,709,120  | 0.50000000            | 0.500000000                 |-  | 2025-02-01 00:00:00.000 -0700 | SNAPSHOT           | NULL                    | 21,474,836,480 | 0.000000000           | 0.000000000                 |-  +-------------------------------+--------------------+-------------------------+----------------+-----------------------+-----------------------------++  +-------------------------------+--------------------+-------------------------+-----------------+----------------+-----------------------+-----------------------------++  | USAGE_DATE                    | STORAGE_TYPE       | COMPUTE_POOL_NAME       | COMPUTE_POOL_ID |       BYTES    |       ADDITIONAL_IOPS |       ADDITIONAL_THROUGHPUT |+  |-------------------------------+--------------------+-------------------------+-----------------+----------------|-----------------------|-----------------------------|+  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_1                  | 1               | 2,684,354,560  | 250.000000000         | 25.000000000                |+  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_2                  | 2               | 5,368,709,120  | 0.50000000            | 0.500000000                 |+  | 2025-02-01 00:00:00.000 -0700 | SNAPSHOT           | NULL                    | NULL            | 21,474,836,480 | 0.000000000           | 0.000000000                 |+  +-------------------------------+--------------------+-------------------------+-----------------+----------------+-----------------------+-----------------------------+   ``` @@ -119,11 +126,11 @@    ```text-  +-------------------------------+--------------------+-------------------------+----------------+-----------------------+-----------------------------+-  | USAGE_DATE                    | STORAGE_TYPE       | COMPUTE_POOL_NAME       |       BYTES    |       ADDITIONAL_IOPS |       ADDITIONAL_THROUGHPUT |-  |-------------------------------+--------------------+-------------------------+----------------|-----------------------|-----------------------------|-  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_1                  | 2,684,354,560  | 250.000000000         | 25.000000000                |-  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_2                  | 5,368,709,120  | 0.50000000            | 0.500000000                 |-  | 2025-02-01 00:00:00.000 -0700 | SNAPSHOT           | NULL                    | 21,474,836,480 | 0.000000000           | 0.000000000                 |-  +-------------------------------+--------------------+-------------------------+----------------+-----------------------+-----------------------------++  +-------------------------------+--------------------+-------------------------+-----------------+----------------+-----------------------+-----------------------------++  | USAGE_DATE                    | STORAGE_TYPE       | COMPUTE_POOL_NAME       | COMPUTE_POOL_ID |       BYTES    |       ADDITIONAL_IOPS |       ADDITIONAL_THROUGHPUT |+  |-------------------------------+--------------------+-------------------------+-----------------+----------------|-----------------------|-----------------------------|+  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_1                  | 1               | 2,684,354,560  | 250.000000000         | 25.000000000                |+  | 2025-02-01 00:00:00.000 -0700 | BLOCK_STORAGE      | POOL_2                  | 2               | 5,368,709,120  | 0.50000000            | 0.500000000                 |+  | 2025-02-01 00:00:00.000 -0700 | SNAPSHOT           | NULL                    | NULL            | 21,474,836,480 | 0.000000000           | 0.000000000                 |+  +-------------------------------+--------------------+-------------------------+-----------------+----------------+-----------------------+-----------------------------+   ```  

--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/version-history+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/version-history@@ -254,4 +254,33 @@ </div> +## August 11, 2026++<a id="label-runtime-server-2026-8-11-2"></a>++### Runtime Server 2026.8.11.2++- Security patches and dependency upgrades.+- Moved from Java 21 to Java 25.+- Upgraded the Runtime UI to version 0.85.0.++<a id="label-runtime-extensions-2026-8-11-10"></a>++### Runtime Extensions 2026.8.11.10++- CDC SQL Server: Added `Max Batch Size = -1` to disable Change Tracking pagination, reducing repeated `CHANGETABLE` evaluation for large backlogs. Enabling `Use Snapshot Isolation` is recommended when using this mode.+- CDC SQL Server: Prevented brief source outages during Azure SQL Managed Instance maintenance or similar failover windows from removing all tables from replication; empty database discovery now retries.+- CDC Oracle: Added multi-PDB replication for users granted `SET CONTAINER`, while preserving existing single-PDB behavior otherwise.+- Salesforce: Formula views now return typed NULL values for formulas referencing unreplicated relationship lookup fields instead of failing view creation.+- CDC SQL Server: Fixed a concurrency race that could cause Change Tracking to silently skip changes for idle tables.++<a id="label-connectors-2026-8-11-1"></a>++### Connectors 2026.8.11.1++- Dataverse 0.32.0: Added an `Include Formatted Values` parameter so ingestion can include display-friendly formatted values while preserving the original raw values.+- Dataverse 0.32.0: Fixed incorrect `_SNOWFLAKE_ID` values by copying the resolved primary-key value instead of the primary-key field name.+- Salesforce 1.13.0: Special objects now use `queryAll` to ingest archived Task and Event records and soft-deleted records. Existing objects require a connector state reset and full reload to retrieve historical records.+- Salesforce 1.13.0: Retained failed Bulk API query jobs so they remain available for troubleshooting.+ ## August 6, 2026  

--- ahttps://docs.snowflake.com/en/sql-reference/organization-usage+++ bhttps://docs.snowflake.com/en/sql-reference/organization-usage@@ -91,4 +91,10 @@     </tr>     <tr>+      <td>[BLOCK_STORAGE_HISTORY](/sql-reference/organization-usage/block_storage_history)</td>+      <td>Historical</td>+      <td>24 hours</td>+      <td>[Premium view](/user-guide/organization-accounts-premium-views) (only available in organization account).</td>+    </tr>+    <tr>       <td>[CLASSES](/sql-reference/organization-usage/classes)</td>       <td>Object</td>@@ -111,4 +117,10 @@       <td>[COMPLETE_TASK_GRAPHS](/sql-reference/organization-usage/complete_task_graphs)</td>       <td>Historical</td>+      <td>24 hours</td>+      <td>[Premium view](/user-guide/organization-accounts-premium-views) (only available in organization account).</td>+    </tr>+    <tr>+      <td>[COMPUTE_POOLS](/sql-reference/organization-usage/compute_pools)</td>+      <td>Object</td>       <td>24 hours</td>       <td>[Premium view](/user-guide/organization-accounts-premium-views) (only available in organization account).</td>@@ -517,4 +529,10 @@     </tr>     <tr>+      <td>[SERVICES](/sql-reference/organization-usage/services)</td>+      <td>Object</td>+      <td>24 hours</td>+      <td>[Premium view](/user-guide/organization-accounts-premium-views) (only available in organization account).</td>+    </tr>+    <tr>       <td>[SESSION_POLICIES](/sql-reference/organization-usage/session_policies)</td>       <td>Object</td>@@ -557,4 +575,10 @@       <td>6 hours</td>       <td>This view is deprecated. Use the [BACKUPS](/sql-reference/account-usage/backups) view instead.</td>+    </tr>+    <tr>+      <td>[SNOWPARK_CONTAINER_SERVICES_HISTORY](/sql-reference/organization-usage/snowpark_container_services_history)</td>+      <td>Historical</td>+      <td>24 hours</td>+      <td>[Premium view](/user-guide/organization-accounts-premium-views) (only available in organization account).</td>     </tr>     <tr> 

--- ahttps://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents-analytical-search+++ bhttps://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents-analytical-search@@ -30,4 +30,8 @@ Cortex Search narrows the full corpus down to a relevant candidate set by finding documents about a specific topic, isolating records containing a particular clause, or filtering to a date range. This happens without scanning every document with a large model. +The agent runs these searches as SQL queries that call the+[SEARCH_PREVIEW](/sql-reference/functions/search_preview-snowflake-cortex) function. The search results are materialized+in an intermediate table so that subsequent SQL and AI function calls can analyze the result set.+ **Adaptive depth** controls how far to search. Rather than using a fixed top-k limit, the agent dynamically adjusts the search depth based on the relevance of results: extending when the tail of results is still relevant to the question,@@ -42,4 +46,9 @@ - **AI_AGG**: Summarizes and aggregates textual evidence at scale. - **SQL**: Groups, counts, joins, ranks, trends, and calculates deltas over the extracted data.++The agent generates and runs the SQL on a Snowflake virtual warehouse. In the Cortex Agents Run API response,+this implicit SQL execution is represented by+[`system_execute_sql`](/user-guide/snowflake-cortex/cortex-agents-run#label-snowflake-agents-run-non-streaming-response)+tool-use and tool-result blocks. You don't add this tool to the agent specification.  ### Planning and auto-routing@@ -78,10 +87,10 @@  1. Classify the query as analytical and generate an execution plan.-2. Run structured search queries (for example, `search(date="2025-05", query="issues")`) with adaptive depth to capture-   the full relevant set of results.-3. Apply AI_FILTER to keep only rows related to customer issues.-4. Apply AI_EXTRACT to pull issue type and customer name from each document.-5. Use AI_AGG and SQL to aggregate results by issue type.-6. Return a precise, data-backed answer.+2. Run structured search queries through `SEARCH_PREVIEW` with adaptive depth to capture the full relevant set of results.+3. Materialize the search results in an intermediate table.+4. Apply AI_FILTER to keep only rows related to customer issues.+5. Apply AI_EXTRACT to pull issue type and customer name from each document.+6. Use AI_AGG and SQL to aggregate results by issue type.+7. Return a precise, data-backed answer.  Example output:@@ -160,5 +169,5 @@ ## Cost considerations -Analytical search incurs costs associated with agent orchestration and the AI functions it uses+Analytical search incurs costs associated with agent orchestration, warehouse compute for SQL execution, and the AI functions it uses ([](/sql-reference/functions/ai_extract), [](/sql-reference/functions/ai_filter), and [](/sql-reference/functions/ai_agg)). Adaptive depth limits unnecessary AI function calls by stopping the search 

--- ahttps://docs.snowflake.com/en/sql-reference/account-usage/compute_pools+++ bhttps://docs.snowflake.com/en/sql-reference/account-usage/compute_pools@@ -9,4 +9,6 @@  # COMPUTE_POOLS view++%spcs% is available to accounts in [AWS, Microsoft Azure, and Google Cloud Platform commercial regions](#label-na-general-regions), with some exceptions. For more information, see [](#label-snowpark-containers-overview-available-regions).  Use this view to get a historical view of compute pools (creation, deletion) in your account for the last 365 days.@@ -33,4 +35,9 @@       <td>VARCHAR</td>       <td>Compute pool name.</td>+    </tr>+    <tr>+      <td>ID</td>+      <td>NUMBER</td>+      <td>Internal, Snowflake-generated identifier for the compute pool.</td>     </tr>     <tr>@@ -66,20 +73,20 @@     <tr>       <td>CREATED</td>-      <td>TIMESTAMP</td>+      <td>TIMESTAMP_LTZ</td>       <td>Date and time when the compute pool was created.</td>     </tr>     <tr>       <td>LAST_RESUMED</td>-      <td>TIMESTAMP</td>+      <td>TIMESTAMP_LTZ</td>       <td>Date and time when the suspended compute pool was last resumed.</td>     </tr>     <tr>       <td>LAST_ALTERED</td>-      <td>TIMESTAMP</td>+      <td>TIMESTAMP_LTZ</td>       <td>Date and time when the compute pool was last updated.</td>     </tr>     <tr>       <td>DELETED</td>-      <td>TIMESTAMP</td>+      <td>TIMESTAMP_LTZ</td>       <td>Date and time when the compute pool was deleted.</td>     </tr>@@ -106,5 +113,5 @@     <tr>       <td>APPLICATION_ID</td>-      <td>VARCHAR</td>+      <td>NUMBER</td>       <td>Application ID for which the compute pool was created. Null if the compute pool was not created for an application.</td>     </tr> 

--- ahttps://docs.snowflake.com/en/sql-reference/account-usage/services+++ bhttps://docs.snowflake.com/en/sql-reference/account-usage/services@@ -128,4 +128,14 @@     </tr>     <tr>+      <td>LAST_SUSPENDED</td>+      <td>TIMESTAMP_LTZ</td>+      <td>Last suspended time of the service.</td>+    </tr>+    <tr>+      <td>AUTO_SUSPEND_SECS</td>+      <td>NUMBER</td>+      <td>Number of seconds of inactivity after which the service is automatically suspended.</td>+    </tr>+    <tr>       <td>DELETED</td>       <td>TIMESTAMP_LTZ</td>@@ -142,4 +152,9 @@       <td>`true` if the service is a job service; `false` otherwise.</td>     </tr>+    <tr>+      <td>IS_ASYNC_JOB</td>+      <td>BOOLEAN</td>+      <td>`true` if the service is an asynchronous job service; `false` otherwise.</td>+    </tr>   </tbody>  

--- ahttps://docs.snowflake.com/en/user-guide/agent-identity+++ bhttps://docs.snowflake.com/en/user-guide/agent-identity@@ -122,18 +122,4 @@ - [Join policies](/user-guide/join-policies#label-join-policy-agent-active) -## Control session privileges--By default, an agent that acts on behalf of a user inherits the privileges available in that user's-session. Agent workloads can issue many queries in a short time, so that inheritance is often broader-than the task requires.--Restricted Session Scopes (RSS) are privilege ceilings that limit what an agent session can do. The-effective permission is always the intersection of what the user's roles allow and what the RSS-permits. An RSS never grants privileges the user doesn't already have.--You can apply an RSS as an account-wide ceiling, attach it for specific users through a session-policy, or inject it during token exchange for partner-orchestrated agents. Restricted Session-Scopes for agents are in private preview. Contact your Snowflake account team for access.- ## Related topics  

--- ahttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/setup-oracledb+++ bhttps://docs.snowflake.com/en/user-guide/data-integration/openflow/connectors/oracle/setup-oracledb@@ -282,4 +282,10 @@   - V$DATABASE +  If you're replicating data from multiple PDBs, also grant SELECT access to:+  - V$CONTAINERS+  - V$PARAMETER+  - CDB_TABLES+  - CDB_USERS+   `ALL_INDEXES` and `ALL_IND_COLUMNS` are required so the connector can detect   unique constraints and unique indexes as replication keys when a table has no@@ -319,4 +325,8 @@ GRANT SET CONTAINER TO c##connectuser CONTAINER=ALL; ```++If you granted SELECT on individual data dictionary views instead of+SELECT_CATALOG_ROLE or SELECT ANY DICTIONARY, also grant SELECT on+`V$CONTAINERS`, `V$PARAMETER`, `CDB_TABLES`, and `CDB_USERS`.  <a id="label-create-xstream-outbound-server"></a> 

--- ahttps://docs.snowflake.com/en/user-guide/tables-clustering-keys+++ bhttps://docs.snowflake.com/en/user-guide/tables-clustering-keys@@ -285,6 +285,10 @@ #### Important Usage Notes -- **Standard tables:** For each VARCHAR column in the clustering key, the current implementation of clustering uses only the first-  5 bytes.+- **Clustering key length:** The amount of each clustering key column that Snowflake uses depends on the Automatic Clustering version:++  - **Optima Clustering:** The clustering key can use up to 1 KB total across all clustering key columns.+  - **Clustering Classic:** For each column in the clustering key, Clustering Classic uses only the first 5 bytes.++  For more information about Optima Clustering and Clustering Classic, see [](/user-guide/tables-auto-reclustering#label-optima-clustering).    If the first N characters are the same for every row, or do not provide sufficient cardinality, then consider clustering on a 

--- ahttps://docs.snowflake.com/en/release-notes/preview-features+++ bhttps://docs.snowflake.com/en/release-notes/preview-features@@ -86,4 +86,11 @@     </tr>     <tr>+      <td>Intent-Driven Governance</td>+      <td>Open</td>+      <td>August 2026</td>+      <td>[](/user-guide/intent-driven-governance)</td>+      <td>Use Snowflake CoCo to assess governance controls, classify sensitive data, review proposed tags and protections, and approve generated SQL before changes are applied.</td>+    </tr>+    <tr>       <td>Automations in %sf-intelligence%</td>       <td>Open</td> 

--- ahttps://docs.snowflake.com/en/sql-reference/sql/create-iceberg-table-rest+++ bhttps://docs.snowflake.com/en/sql-reference/sql/create-iceberg-table-rest@@ -704,6 +704,9 @@   Failure to include the quotes might result in an `Object does not exist` error or   a similar type of error.-- The OR REPLACE option performs a non-atomic operation, which in this case is a DROP operation followed by CREATE,-  in your external Iceberg catalog. By default, the implicit drop does not signal the catalog to delete the replaced+- In a [catalog-linked database](/user-guide/tables-iceberg-catalog-linked-database), the `OR REPLACE` option performs a non-atomic operation,+  which in this case is a DROP operation followed by CREATE, in your external Iceberg catalog. The new table has a new `table-uuid`.+  In a standard Snowflake database (not linked to a catalog), `OR REPLACE` drops and recreates only the Snowflake table object;+  the table in the remote catalog is unchanged.+- By default, the implicit drop does not signal the catalog to delete the replaced   table's data files. Specify `PURGE` (or set `PURGE_ON_DROP_TABLE` on the table, schema, or database) to forward   the purge intent to the catalog on the implicit drop. Snowflake never deletes external data files directly — the 

--- ahttps://docs.snowflake.com/en/sql-reference/account-usage/snowpark_container_services_history+++ bhttps://docs.snowflake.com/en/sql-reference/account-usage/snowpark_container_services_history@@ -48,4 +48,9 @@     </tr>     <tr>+      <td>COMPUTE_POOL_ID</td>+      <td>NUMBER</td>+      <td>Internal, Snowflake-generated identifier of the compute pool which incurred the credit usage.</td>+    </tr>+    <tr>       <td>IS_EXCLUSIVE</td>       <td>BOOLEAN</td>@@ -59,5 +64,5 @@     <tr>       <td>APPLICATION_ID</td>-      <td>VARCHAR</td>+      <td>NUMBER</td>       <td>The ID of the application for which the compute pool was created; otherwise NULL.</td>     </tr> 

--- ahttps://docs.snowflake.com/en/user-guide/tables-iceberg+++ bhttps://docs.snowflake.com/en/user-guide/tables-iceberg@@ -634,4 +634,7 @@   table in your remote catalog. After that, you can use the [CREATE ICEBERG TABLE (Iceberg REST catalog)](/sql-reference/sql/create-iceberg-table-rest) syntax to create   an Iceberg table in Snowflake and write to it.+- When you use `CREATE OR REPLACE` to create an Iceberg table, behavior depends on the database type:+  - **Catalog-linked database:** Snowflake drops and recreates the table in the remote catalog. The new table has a new `table-uuid`.+  - **Standard Snowflake database (not linked to a catalog):** Snowflake drops and recreates only the Snowflake table object. The table in the remote catalog is unchanged. - For the AWS Glue Data Catalog: Dropping an externally managed table through Snowflake doesn't delete   the underlying table files. This behavior is specific to the AWS Glue Data Catalog implementation.@@ -646,5 +649,4 @@   - Conversion to Snowflake-managed tables.   - External Iceberg catalogs that don't conform to the Iceberg REST protocol.-  - Using the OR REPLACE option when creating a table.   - Using the CREATE ICEBERG TABLE (catalog-linked database) ... AS SELECT syntax if you use one of the following catalogs as your remote catalog:     - AWS Glue 

--- ahttps://docs.snowflake.com/en/user-guide/tables-iceberg-externally-managed-writes+++ bhttps://docs.snowflake.com/en/user-guide/tables-iceberg-externally-managed-writes@@ -547,4 +547,7 @@   table in your remote catalog. After that, you can use the [CREATE ICEBERG TABLE (Iceberg REST catalog)](/sql-reference/sql/create-iceberg-table-rest) syntax to create   an Iceberg table in Snowflake and write to it.+- When you use `CREATE OR REPLACE` to create an Iceberg table, behavior depends on the database type:+  - **Catalog-linked database:** Snowflake drops and recreates the table in the remote catalog. The new table has a new `table-uuid`.+  - **Standard Snowflake database (not linked to a catalog):** Snowflake drops and recreates only the Snowflake table object. The table in the remote catalog is unchanged. - For the AWS Glue Data Catalog: Dropping an externally managed table through Snowflake doesn't delete   the underlying table files. This behavior is specific to the AWS Glue Data Catalog implementation.@@ -559,5 +562,4 @@   - Conversion to Snowflake-managed tables.   - External Iceberg catalogs that don't conform to the Iceberg REST protocol.-  - Using the OR REPLACE option when creating a table.   - Using the CREATE ICEBERG TABLE (catalog-linked database) ... AS SELECT syntax if you use one of the following catalogs as your remote catalog:     - AWS Glue