pgjdbc/pgjdbc GitHub issues and pull requests (mirror)  
help / color / mirror / Atom feed
From: atorik (@atorik) <[email protected]>
To: pgjdbc/pgjdbc <[email protected]>
Subject: [pgjdbc/pgjdbc] PR #3906: fix:  incorrect pg_stat_replication.reply_time calculation
Date: Mon, 12 Jan 2026 05:30:54 +0000
Message-ID: <[email protected]> (raw)

Hi,

When creating a replication connection using the PostgreSQL JDBC driver,
`pg_stat_replication.reply_time` is reported as a timestamp more than 20 years in the past.

Example:

```
=# select application_name, reply_time from pg_stat_replication;
    application_name    |          reply_time
------------------------+-------------------------------
 PostgreSQL JDBC Driver | 1999-12-21 10:01:58.904961+09
(1 row)
```

I initially noticed this issue while using Debezium, but it can also be reproduced with a minimal replication connection app like [this](https://github.com/user-attachments/files/24557645/LogicaldecodingTest.java).

`reply_time` was calculated using `System.nanoTime()`, which is only suitable for measuring elapsed time and not wall-clock time. Additionally, `TimeUnit.convert()` was used with an incorrect source unit.

This change addresses the issue by:

- Replacing `System.nanoTime()` with `System.currentTimeMillis()` to obtain a proper wall-clock timestamp.
- Correcting the `TimeUnit.convert()` usage so that the source unit matches the actual time value being converted.

What do you think?

### All Submissions:

* [x]  Have you followed the guidelines in our [Contributing](https://github.com/pgjdbc/pgjdbc/blob/master/CONTRIBUTING.md) document?
* [x]  Have you checked to ensure there aren't other open [Pull Requests](../../pulls) for the same update/change?

### Changes to Existing Features:

* [x] Does this break existing behaviour? If so please explain.
* [x] Have you added an explanation of what your changes do and why you'd like us to include them?
* [x] Have you written new tests for your core changes, as applicable?
  * No new tests have been added at this time. I considered adding a test that verifies `reply_time` is not excessively in the past (e.g., more than 24 hours earlier than the current time). However, I'm not sure it's worth adding. 
* [x] Have you successfully run tests with your changes locally?


reply

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Reply to all the recipients using the --to and --cc options:
  reply via email

  To: github://pgjdbc/pgjdbc
  Cc: [email protected], [email protected]
  Subject: Re: [pgjdbc/pgjdbc] PR #3906: fix:  incorrect pg_stat_replication.reply_time calculation
  In-Reply-To: <<[email protected]>>

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox