Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Appearance settings

create minimal SYS.V$SESSION view#1185

Open
pierreforstmann wants to merge 9 commits intoIvorySQL:masterIvorySQL/IvorySQL:masterfrom
pierreforstmann:patch_dyn_viewspierreforstmann/IvorySQL:patch_dyn_viewsCopy head branch name to clipboard
Open

create minimal SYS.V$SESSION view#1185
pierreforstmann wants to merge 9 commits intoIvorySQL:masterIvorySQL/IvorySQL:masterfrom
pierreforstmann:patch_dyn_viewspierreforstmann/IvorySQL:patch_dyn_viewsCopy head branch name to clipboard

Conversation

@pierreforstmann
Copy link
Contributor

@pierreforstmann pierreforstmann commented Feb 7, 2026

Hello,

I propose this patch to create a minimal view for Oracle SYS.V$SESSION.

Could you please confirm that this is right way to create a dynamic Oracle view.

Thanks and regards

Summary by CodeRabbit

  • New Features

    • Added three public dynamic views—V$SESSION, V$PROCESS, and V$PARAMETER—to surface session, process, and configuration parameter details (status, process info, and parameter name/value metadata).
  • Permissions

    • Granted read (SELECT) access on V$SESSION, V$PROCESS, and V$PARAMETER to all users for broad visibility.

@coderabbitai
Copy link
Contributor

coderabbitai bot commented Feb 7, 2026

Note

Reviews paused

It looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the reviews.auto_review.auto_pause_after_reviewed_commits setting.

Use the following commands to manage reviews:

  • @coderabbitai resume to resume automatic reviews.
  • @coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

  • ▶️ Resume reviews
  • 🔍 Trigger review
📝 Walkthrough

Walkthrough

Adds three new public SYS dynamic views: V$SESSION, V$PROCESS, and V$PARAMETER, defined from pg_stat_activity and pg_settings; each view includes a GRANT SELECT ON SYS.<view> TO PUBLIC. (+66/-0)

Changes

Cohort / File(s) Summary
Database View Schema
contrib/ivorysql_ora/src/sysview/sysview--1.0.sql
Creates three SYS views: V$SESSION (maps SID, USERNAME, STATUS, MACHINE, PROGRAM, TYPE, LOGON_TIME, SQL_TEXT, EVENT, TRANSACTION_START, PORT from pg_stat_activity), V$PROCESS (exposes SPID, USERNAME, PROGRAM, MACHINE, PNAME, BACKGROUND from backend activity), and V$PARAMETER (maps NAME, TYPE, VALUE, DEFAULT_VALUE, ISDEFAULT, ISSES_MODIFIABLE, ISSYS_MODIFIABLE, DESCRIPTION from pg_settings). Adds GRANT SELECT ON SYS.<view> TO PUBLIC for each view.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Poem

🐇 I nibble rows and hop with glee,
New V$ windows open for me to see,
Sessions, procs, and params all in line,
I tidy columns with a carrot-tined sign,
Hooray — a rabbit's hop, compact and fine!

🚥 Pre-merge checks | ✅ 2 | ❌ 1
❌ Failed checks (1 warning)
Check name Status Explanation Resolution
Title check ⚠️ Warning The title 'create minimal SYS.V$SESSION view' mentions only one of three views being added (V$SESSION, V$PROCESS, V$PARAMETER), making it partially related to the changeset rather than fully capturing the main change. Update the title to reflect all three views being created, such as 'create SYS.V$SESSION, SYS.V$PROCESS, and SYS.V$PARAMETER views' or 'add Oracle dynamic views for session, process, and parameter information'.
✅ Passed checks (2 passed)
Check name Status Explanation
Description Check ✅ Passed Check skipped - CodeRabbit’s high-level summary is enabled.
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.

✨ Finishing touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Post copyable unit tests in a comment

Tip

Issue Planner is now in beta. Read the docs and try it out! Share your feedback on Discord.


Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands and usage tips.

Copy link
Contributor

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 2

🤖 Fix all issues with AI agents
In `@contrib/ivorysql_ora/src/sysview/sysview--1.0.sql`:
- Around line 1201-1214: The SYS.V$SESSION view is missing a post-create GRANT
and uses raw PG types; update the view definition for SYS.V$SESSION to cast
columns to Oracle-compatible types (e.g., USENAME::VARCHAR2(128) as USERNAME,
CLIENT_ADDR::VARCHAR2(64) or coalesce(hostname, CLIENT_ADDR::text)::VARCHAR2(64)
as MACHINE, CLIENT_PORT::NUMBER as PORT, QUERY::VARCHAR2(4000) as SQL_TEXT,
STATE/WAIT_EVENT/XACT_START/ BACKEND_START cast to appropriate
VARCHAR2/DATE/NUMBER types) and then add a GRANT SELECT ON SYS.V$SESSION TO
PUBLIC immediately after the CREATE OR REPLACE VIEW statement so non-superusers
can query it.
- Around line 1194-1195: Remove the duplicate grant statement: locate the
redundant SQL line "GRANT SELECT ON SYS.USER_VIEWS TO PUBLIC;" (the second
occurrence around the end of the file) and delete it so only the single original
grant remains; ensure no other identical GRANT statements are introduced
elsewhere in the file.
🧹 Nitpick comments (1)
contrib/ivorysql_ora/src/sysview/sysview--1.0.sql (1)

1205-1208: Column semantics diverge from Oracle V$SESSION.

A few mappings worth noting for Oracle compatibility:

  • STATUS (STATE): PostgreSQL returns values like 'active', 'idle', 'idle in transaction', whereas Oracle V$SESSION.STATUS uses 'ACTIVE', 'INACTIVE', 'KILLED', etc. Consider using UPPER() at minimum, or a CASE expression to map PG states to Oracle equivalents.
  • TYPE (BACKEND_TYPE): Oracle uses 'USER' / 'BACKGROUND'. PostgreSQL returns 'client backend', 'autovacuum launcher', etc. A CASE mapping (e.g., 'client backend''USER', else 'BACKGROUND') would improve compatibility.
  • MACHINE (CLIENT_ADDR): Oracle provides the client hostname; PG provides the IP. This may be acceptable as a "minimal" view, but worth documenting the difference.

contrib/ivorysql_ora/src/sysview/sysview--1.0.sql Outdated Show resolved Hide resolved
contrib/ivorysql_ora/src/sysview/sysview--1.0.sql Outdated Show resolved Hide resolved
@yuanyl630
Copy link
Collaborator

Thank you again for your contribution to IvorySQL. We willl review your codes soon.

@yuanyl630 yuanyl630 self-requested a review February 9, 2026 00:51
Copy link
Contributor

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 3

🤖 Fix all issues with AI agents
In `@contrib/ivorysql_ora/src/sysview/sysview--1.0.sql`:
- Around line 1240-1241: Change the DEFAULT_VALUE column mapping from RESET_VAL
to BOOT_VAL: replace the reference mapping RESET_VAL::VARCHAR2(255) AS
DEFAULT_VALUE with BOOT_VAL::VARCHAR2(255) AS DEFAULT_VALUE so DEFAULT_VALUE
reflects the compiled-in bootstrap default (pg_settings.boot_val) rather than
the reset value; update any related aliases or comments that reference RESET_VAL
to avoid confusion with RESET_VAL/boot_val semantics.
- Around line 1224-1226: The BACKGROUND CASE expression returns an integer -1
and uses redundant WHENs; change it to produce a VARCHAR(1) matching Oracle
V$PROCESS.BACKGROUND semantics: return '1' for background processes and NULL
otherwise. Update the CASE that references BACKEND_TYPE (the current CASE
producing BACKGROUND) to use CASE WHEN BACKEND_TYPE = 'client backend' THEN NULL
ELSE '1' END AS BACKGROUND so the type and values match Oracle's
V$PROCESS.BACKGROUND and remove the redundant WHEN BACKEND_TYPE <> 'client
backend' branch.
- Around line 1247-1251: The CASE currently producing IMMEDIATE/DEFERRED/FALSE
is being assigned to AS ISSES_MODIFIABLE but those values actually correspond to
ISSYS_MODIFIABLE; update the SELECT so that AS ISSES_MODIFIABLE returns TRUE or
FALSE (e.g., CONTEXT IN ('user','superuser') or CONTEXT = 'postmaster' =>
appropriate TRUE/FALSE) and add a separate expression AS ISSYS_MODIFIABLE that
preserves the current IMMEDIATE/DEFERRED/FALSE mapping (or alternatively rename
the existing AS ISSES_MODIFIABLE to AS ISSYS_MODIFIABLE if you prefer a single
column); adjust the CASE(s) around the same CASE ... END expression to use the
exact identifiers ISSES_MODIFIABLE and ISSYS_MODIFIABLE.
🧹 Nitpick comments (1)
contrib/ivorysql_ora/src/sysview/sysview--1.0.sql (1)

1235-1239: TYPE column missing ::NUMBER cast.

All other numeric columns in the file use ::NUMBER. Add a cast for consistency.

Proposed fix
     CASE
         WHEN VARTYPE = 'bool'    THEN 1
         WHEN VARTYPE = 'integer' THEN 3
         ELSE 2
-    END AS TYPE,
+    END::NUMBER AS TYPE,

contrib/ivorysql_ora/src/sysview/sysview--1.0.sql Outdated Show resolved Hide resolved
contrib/ivorysql_ora/src/sysview/sysview--1.0.sql Outdated Show resolved Hide resolved
Comment on lines 1247 to 1251
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟠 Major

ISSES_MODIFIABLE returns ISSYS_MODIFIABLE values (IMMEDIATE/DEFERRED/FALSE).

Oracle's ISSES_MODIFIABLE is TRUE/FALSE (whether ALTER SESSION can change it). The values IMMEDIATE/DEFERRED/FALSE correspond to Oracle's ISSYS_MODIFIABLE column instead. Either rename the column to ISSYS_MODIFIABLE (and optionally add a separate ISSES_MODIFIABLE), or change the values to TRUE/FALSE.

Proposed fix — provide both columns
+    CASE
+        WHEN CONTEXT IN ('user', 'superuser') THEN 'TRUE'
+        ELSE 'FALSE'
+    END::VARCHAR2(5) AS ISSES_MODIFIABLE,
     CASE
         WHEN CONTEXT IN ('user', 'superuser') THEN 'IMMEDIATE'
         WHEN CONTEXT = 'postmaster' THEN 'FALSE'
         ELSE 'DEFERRED'
-    END AS ISSES_MODIFIABLE,
+    END::VARCHAR2(9) AS ISSYS_MODIFIABLE,
📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change
CASE
WHEN CONTEXT IN ('user', 'superuser') THEN 'IMMEDIATE'
WHEN CONTEXT = 'postmaster' THEN 'FALSE'
ELSE 'DEFERRED'
END AS ISSES_MODIFIABLE,
CASE
WHEN CONTEXT IN ('user', 'superuser') THEN 'TRUE'
ELSE 'FALSE'
END::VARCHAR2(5) AS ISSES_MODIFIABLE,
CASE
WHEN CONTEXT IN ('user', 'superuser') THEN 'IMMEDIATE'
WHEN CONTEXT = 'postmaster' THEN 'FALSE'
ELSE 'DEFERRED'
END::VARCHAR2(9) AS ISSYS_MODIFIABLE,
🤖 Prompt for AI Agents
In `@contrib/ivorysql_ora/src/sysview/sysview--1.0.sql` around lines 1247 - 1251,
The CASE currently producing IMMEDIATE/DEFERRED/FALSE is being assigned to AS
ISSES_MODIFIABLE but those values actually correspond to ISSYS_MODIFIABLE;
update the SELECT so that AS ISSES_MODIFIABLE returns TRUE or FALSE (e.g.,
CONTEXT IN ('user','superuser') or CONTEXT = 'postmaster' => appropriate
TRUE/FALSE) and add a separate expression AS ISSYS_MODIFIABLE that preserves the
current IMMEDIATE/DEFERRED/FALSE mapping (or alternatively rename the existing
AS ISSES_MODIFIABLE to AS ISSYS_MODIFIABLE if you prefer a single column);
adjust the CASE(s) around the same CASE ... END expression to use the exact
identifiers ISSES_MODIFIABLE and ISSYS_MODIFIABLE.

Copy link
Contributor

@coderabbitai coderabbitai bot left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Actionable comments posted: 2

🤖 Fix all issues with AI agents
In `@contrib/ivorysql_ora/src/sysview/sysview--1.0.sql`:
- Around line 1235-1246: The CASE expressions producing the TYPE and ISDEFAULT
columns return raw PostgreSQL types; update them to include explicit
Oracle-compatible casts: cast the numeric result of the TYPE CASE to an Oracle
NUMBER (e.g., NUMBER(1)) and cast the string result of the ISDEFAULT CASE to
VARCHAR2(5) (or appropriate VARCHAR2 length). Locate the CASE that checks
VARTYPE (the alias TYPE) and the CASE that compares setting = boot_val (the
alias ISDEFAULT) and append the corresponding ::NUMBER(...) and ::VARCHAR2(...)
style casts (or use TO_NUMBER/TO_CHAR as your style across the file) so they
match the explicit Oracle type casting used for other columns like
SETTING::VARCHAR2(80) and BOOT_VAL::VARCHAR2(255).
- Around line 1247-1251: The CREATE VIEW contains a duplicate, orphaned "END AS
ISSES_MODIFIABLE" after the completed CASE expression; remove the stray line so
the CASE ... END::VARCHAR2(5) AS ISSES_MODIFIABLE stays as the single, valid
expression (look for the CASE that sets ISSES_MODIFIABLE and delete the extra
trailing "END AS ISSES_MODIFIABLE").

Comment on lines 1235 to 1246
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🛠️ Refactor suggestion | 🟠 Major

Missing Oracle-compatible type casts on TYPE and ISDEFAULT columns.

Every other column in this file is explicitly cast to an Oracle type. These two CASE expressions return raw PostgreSQL types (integer and text), breaking consistency.

Proposed fix
     CASE
         WHEN VARTYPE = 'bool'    THEN 1
         WHEN VARTYPE = 'integer' THEN 3
         ELSE 2
-    END AS TYPE,
+    END::NUMBER AS TYPE,
     SETTING::VARCHAR2(80) AS VALUE,
     BOOT_VAL::VARCHAR2(255) AS DEFAULT_VALUE,
     -- Mapping update levels
     CASE
         WHEN setting = boot_val THEN 'TRUE'
         ELSE 'FALSE'
-    END AS ISDEFAULT,
+    END::VARCHAR2(5) AS ISDEFAULT,
📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change
CASE
WHEN VARTYPE = 'bool' THEN 1
WHEN VARTYPE = 'integer' THEN 3
ELSE 2
END AS TYPE,
SETTING::VARCHAR2(80) AS VALUE,
BOOT_VAL::VARCHAR2(255) AS DEFAULT_VALUE,
-- Mapping update levels
CASE
WHEN setting = boot_val THEN 'TRUE'
ELSE 'FALSE'
END AS ISDEFAULT,
CASE
WHEN VARTYPE = 'bool' THEN 1
WHEN VARTYPE = 'integer' THEN 3
ELSE 2
END::NUMBER AS TYPE,
SETTING::VARCHAR2(80) AS VALUE,
BOOT_VAL::VARCHAR2(255) AS DEFAULT_VALUE,
-- Mapping update levels
CASE
WHEN setting = boot_val THEN 'TRUE'
ELSE 'FALSE'
END::VARCHAR2(5) AS ISDEFAULT,
🤖 Prompt for AI Agents
In `@contrib/ivorysql_ora/src/sysview/sysview--1.0.sql` around lines 1235 - 1246,
The CASE expressions producing the TYPE and ISDEFAULT columns return raw
PostgreSQL types; update them to include explicit Oracle-compatible casts: cast
the numeric result of the TYPE CASE to an Oracle NUMBER (e.g., NUMBER(1)) and
cast the string result of the ISDEFAULT CASE to VARCHAR2(5) (or appropriate
VARCHAR2 length). Locate the CASE that checks VARTYPE (the alias TYPE) and the
CASE that compares setting = boot_val (the alias ISDEFAULT) and append the
corresponding ::NUMBER(...) and ::VARCHAR2(...) style casts (or use
TO_NUMBER/TO_CHAR as your style across the file) so they match the explicit
Oracle type casting used for other columns like SETTING::VARCHAR2(80) and
BOOT_VAL::VARCHAR2(255).

contrib/ivorysql_ora/src/sysview/sysview--1.0.sql Outdated Show resolved Hide resolved
@yuanyl630
Copy link
Collaborator

Hi @pierreforstmann,
Thanks for working on this! It's great to see someone tackling Oracle compatibility for the V$ views - these are really useful for DBAs migrating from Oracle.
I took a look at the implementation and while the basic mapping is on the right track, there are some important issues we need to fix. Let me walk through what I found:

  1. Missing SERIAL# in V$SESSION
    This is actually a pretty critical issue. In Oracle, you need both SID and SERIAL# together to uniquely identify a session. The problem is when a session disconnects, PostgreSQL can reuse that PID for a new backend. Without SERIAL#, we could end up killing the wrong session if someone runs ALTER SYSTEM KILL SESSION.
    I know PostgreSQL doesn't track this natively, so we'll need to figure out a way to maintain session sequence numbers - maybe through a sequence or some shared memory tracking. This is definitely worth discussing how we want to implement it.
  2. No lock/blocking information
    One of the main reasons DBAs query V$SESSION is to diagnose locks and find out who's blocking whom. Right now we're missing:
    BLOCKING_SESSION - which session is blocking this one
    WAIT_CLASS - type of wait event
    STATE - whether actively waiting or not
    The good news is we can get most of this from pg_locks.
  3. STATUS values don't match Oracle
    You're mapping PostgreSQL's state directly, but the values are different. Oracle uses ACTIVE/INACTIVE/KILLED, while we have active/idle/idle in transaction/etc.
  4. Important stuff that's missing
    SQL tracking fields - We have SQL_TEXT but missing SQL_ID (really common in Oracle monitoring tools), SQL_EXEC_START, and PREV_SQL_ID
    Client Info

@pierreforstmann
Copy link
Contributor Author

pierreforstmann commented Feb 10, 2026

Hi @yuanyl630

  1. Missing SERIAL#: yes it is missing but I disagree that this is a critical issue. On Linux systems, pg_stat_activity.pid is the Linux process identifier of the backend and in theory that could happen. But in practice I think this is very unlikely given the way GNU/Linux distributions manage processes identifiers. Can you reproduce such an issue and on which distribution ? In Oracle V$SESSION.SID is different from OS process identifier and is managed by Oracle and may behave differently.

2/3/4: I have added BLOCKING_SESSION, WAIT_CLASS, SQL_ID, SQL_EXEC_START and tried to fix STATE and STATUS.
I did not find an easy way to add PREV_SQL_ID.

@yuanyl630
Copy link
Collaborator

yuanyl630 commented Feb 11, 2026

Hi @pierreforstmann,
Thanks for the follow-up commit — adding BLOCKING_SESSION, WAIT_CLASS, SQL_ID and SQL_EXEC_START, and improving the STATUS mapping is a solid step forward. I also saw your reply about SERIAL# in the PR discussion. I get the point that PID reuse may feel unlikely on many systems, but I wouldn’t treat it as “won’t happen in practice”.
On older / long-lived distros (e.g. CentOS 7 / older kernels) and especially on dev/CI machines with high process churn (e.g. frequent go test ./..., parallel builds, lots of short-lived processes), PID reuse absolutely does happen — it’s a normal consequence of PID being a finite resource. pidfd helps avoid PID-as-handle races, but it doesn’t mean PID numbers stop being reused. So if we keep SID = backend PID and don’t have SERIAL#, we should at least document the limitation clearly (and be careful not to build any “kill by SID” semantics on top of it without a generation counter or a second factor like backend_start).

That said, I think we still have a couple of must-fix items before we can merge, mainly around the blocking logic:

  1. BLOCKING_SESSION correctness
    The current BLOCKS CTE matches pg_locks rows by resource identity, but it doesn’t ensure the “blocking” side actually holds a granted lock (missing BLOCKING.GRANTED) and it doesn’t capture queue ordering / soft blocking semantics. This can produce false blockers or miss real blockers.
    PostgreSQL already provides pg_blocking_pids(pid) which encodes the correct blocker semantics (including parallel lock groups). I’d strongly prefer we use that rather than re-implementing the lock matching in SQL.
  2. Row duplication risk
    Because the view LEFT JOINs to BLOCKS, one session can appear multiple times if there are multiple matching blockers. V$SESSION should be 1 row per session — so we need to ensure we return a single blocker PID (or NULL) per session.
  3. Visibility / security
    We’re currently granting SELECT on these SYS views to PUBLIC. Please double-check that this doesn’t unintentionally expose other users’ session details / query text / lock info to non-privileged users. Oracle typically restricts V$ access, and PostgreSQL has pg_read_all_stats/pg_monitor-style separation; we should align with that.

Nice-to-haves (not blocking this PR, but would improve compatibility):
Map TYPE closer to Oracle (USER vs BACKGROUND) instead of exposing backend_type strings.
Avoid returning NULL for STATUS (use a sensible default like INACTIVE).
Consider client_hostname / host(client_addr) for MACHINE.
Document that SQL_ID is based on query_id and may be NULL unless query-id tracking is enabled.

Let me know if you want me to point to a specific snippet for the pg_blocking_pids() approach — it should simplify the view a lot and avoid the tricky edge cases.
Thanks again!

@yuanyl630
Copy link
Collaborator

Hi @pierreforstmann,
Thanks for the new update — this is a clear improvement over the previous version.
I think we’re close, but I still have a few items I’d like to see addressed before merge:

  1. Avoid calling pg_blocking_pids(pid) twice per row
    Right now the view calls it in CARDINALITY(...) and again to index [1]. It’s not huge, but this view is likely to be queried frequently. Please compute it once (e.g. via a LATERAL subquery) or rely on the [1] indexing returning NULL when the array is empty.
  2. Schema-qualify builtin function
    Please use pg_catalog.pg_blocking_pids(...) to avoid any search_path surprises.
  3. Fix comment typo
    The comment mentions PG_STAT_ACTIVITY.START — I think you meant pg_stat_activity.backend_start. Could you update that so we don’t mislead future maintainers?
  4. MACHINE may be NULL in many environments
    client_hostname is often NULL. To get closer to Oracle’s “usually populated” behavior, consider something like COALESCE(client_hostname, host(client_addr), client_addr::text) (casted to VARCHAR2(...)).
  5. Privileges / visibility
    I agree with not granting V$SESSION to PUBLIC given it exposes SQL_TEXT.
    However, V$PROCESS and V$PARAMETER are still granted to PUBLIC — please confirm that’s intentional for IvorySQL’s security model. If we want to align more with typical Oracle / PG monitoring privilege separation, we might want to restrict these to a monitoring role (e.g. similar to pg_monitor / pg_read_all_stats).

@yuanyl630
Copy link
Collaborator

Hi @pierreforstmann, thanks for the updates on this PR.
I’m currently on vacation this week and won’t be able to do a timely review. I’ll take a close look as soon as I’m back (around <25/02>).
Thanks for your patience!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants

Comments

Close sidebar
Morty Proxy This is a proxified and sanitized view of the page, visit original site.