Обсуждение: Rename sync_error_count to tbl_sync_error_count in subscription statistics

Hi,

This patch proposes renaming a column in the view
pg_stat_subscription_stats to disambiguate between table sync and
sequence sync error counts. With the introduction of
seq_sync_error_count, the existing column name sync_error_count
becomes ambiguous, it's unclear whether it refers to tables,
sequences, or both. Renaming it to tbl_sync_error_count makes the
meaning explicit: it represents errors that occurred during table
synchronization. This helps users avoid confusion when interpreting
the view's output. These changes were also suggested by Amit's first
comment at [1] and Peter's fourth comment at [2].
Thoughts?

Since an existing column is being renamed, this change should also be
mentioned in the release notes so that users can update any scripts or
monitoring queries that reference it.

[1] - https://www.postgresql.org/message-id/CAA4eK1LB7u2KQLRFh6xfTSpEB-8gbpR%3DhqzFOfL9Z1R8rj7Q5g%40mail.gmail.com
[2] - https://www.postgresql.org/message-id/CAHut%2BPtoLN0bRu7bNiSeF04dQQecoW-EXKMBX%3DHy0uqCvQa8MA%40mail.gmail.com

Regards,
Vignesh

On Wed, Nov 12, 2025 at 9:54 AM vignesh C <vignesh21@gmail.com> wrote:
>
> Hi,
>
> This patch proposes renaming a column in the view
> pg_stat_subscription_stats to disambiguate between table sync and
> sequence sync error counts. With the introduction of
> seq_sync_error_count, the existing column name sync_error_count
> becomes ambiguous, it's unclear whether it refers to tables,
> sequences, or both. Renaming it to tbl_sync_error_count makes the
> meaning explicit: it represents errors that occurred during table
> synchronization. This helps users avoid confusion when interpreting
> the view's output.

+1 on the intent. It will definitely help avoid confusion in the names.

I’m slightly leaning toward using 'table_sync_error_count' instead of
'tbl_sync_error_count'. The name 'table_sync_error_count' isn’t too
long and should work fine, while 'tbl_..' feels more like an internal
variable name. But I would like to see what others prefer here.

thanks
Shveta

Hi,

On Wed, Nov 12, 2025 at 1:04 PM shveta malik <shveta.malik@gmail.com> wrote:
>
> On Wed, Nov 12, 2025 at 9:54 AM vignesh C <vignesh21@gmail.com> wrote:
> >
> > Hi,
> >
> > This patch proposes renaming a column in the view
> > pg_stat_subscription_stats to disambiguate between table sync and
> > sequence sync error counts. With the introduction of
> > seq_sync_error_count, the existing column name sync_error_count
> > becomes ambiguous, it's unclear whether it refers to tables,
> > sequences, or both. Renaming it to tbl_sync_error_count makes the
> > meaning explicit: it represents errors that occurred during table
> > synchronization. This helps users avoid confusion when interpreting
> > the view's output.

Thanks for the patch.

> +1 on the intent. It will definitely help avoid confusion in the names.
>
> I’m slightly leaning toward using 'table_sync_error_count' instead of
> 'tbl_sync_error_count'. The name 'table_sync_error_count' isn’t too
> long and should work fine, while 'tbl_..' feels more like an internal
> variable name. But I would like to see what others prefer here.
>
> thanks
> Shveta
>
>

+1 for using table_sync_error_count as the user-facing column name —
it’s clearer and consistent with existing catalog naming.
Do we need to bump CATALOG_VERSION_NO for this update?

--
Best,
Xuneng

On Wed, Nov 12, 2025 at 4:04 PM shveta malik <shveta.malik@gmail.com> wrote:
>
> On Wed, Nov 12, 2025 at 9:54 AM vignesh C <vignesh21@gmail.com> wrote:
> >
> > Hi,
> >
> > This patch proposes renaming a column in the view
> > pg_stat_subscription_stats to disambiguate between table sync and
> > sequence sync error counts. With the introduction of
> > seq_sync_error_count, the existing column name sync_error_count
> > becomes ambiguous, it's unclear whether it refers to tables,
> > sequences, or both. Renaming it to tbl_sync_error_count makes the
> > meaning explicit: it represents errors that occurred during table
> > synchronization. This helps users avoid confusion when interpreting
> > the view's output.
>
> +1 on the intent. It will definitely help avoid confusion in the names.
>
> I’m slightly leaning toward using 'table_sync_error_count' instead of
> 'tbl_sync_error_count'. The name 'table_sync_error_count' isn’t too
> long and should work fine, while 'tbl_..' feels more like an internal
> variable name. But I would like to see what others prefer here.
>

+1 to use 'table' instead of 'tbl'.

~~

But, I was thinking about different naming convention. Since the
proposal will already change the member, then this is probably my last
chance to propose this slighty different approach, so I'm just putting
it out there to see what people think.

Background -- there are the following counters:
1x "apply" error counter
2x "synchronization" error counters
7x "conflict" error counters

IMO it would be better for the "synchronization" error counters to
have a common "sync_" prefix; this is consistent with all the
"conflict" counters having a "confl_" prefix.

e.g.

CURRENT
seq_sync_error_count
sync_error_count

MY SUGGESTION
sync_seq_error_count
sync_table_error_count

~~

Thoughts?

======
Kind Regards,
Peter Smith.
Fujitsu Australia

On Thu, Nov 13, 2025 at 4:31 AM Peter Smith <smithpb2250@gmail.com> wrote:
>
> CURRENT
> seq_sync_error_count
> sync_error_count
>
> MY SUGGESTION
> sync_seq_error_count
> sync_table_error_count
>

+1. Along with consistency with other column names, it will also make
it easier for users to group and query all 'sync' related counts
together.

--
With Regards,
Amit Kapila.

> On Nov 13, 2025, at 11:42, Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Thu, Nov 13, 2025 at 4:31 AM Peter Smith <smithpb2250@gmail.com> wrote:
>>
>> CURRENT
>> seq_sync_error_count
>> sync_error_count
>>
>> MY SUGGESTION
>> sync_seq_error_count
>> sync_table_error_count
>>
>
> +1. Along with consistency with other column names, it will also make
> it easier for users to group and query all 'sync' related counts
> together.
>

+1. A common prefix looks good and brings benefits such as what Amit mentioned.

Best regards,
--
Chao Li (Evan)
HighGo Software Co., Ltd.
https://www.highgo.com/

On Thu, Nov 13, 2025 at 10:48 AM Chao Li <li.evan.chao@gmail.com> wrote:

> >
> > +1. Along with consistency with other column names, it will also make
> > it easier for users to group and query all 'sync' related counts
> > together.
> >
>
> +1. A common prefix looks good and brings benefits such as what Amit mentioned.
>
+1 this makes sense.


--
Regards,
Dilip Kumar
Google

On Thu, 13 Nov 2025 at 09:12, Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Thu, Nov 13, 2025 at 4:31 AM Peter Smith <smithpb2250@gmail.com> wrote:
> >
> > CURRENT
> > seq_sync_error_count
> > sync_error_count
> >
> > MY SUGGESTION
> > sync_seq_error_count
> > sync_table_error_count
> >
>
> +1. Along with consistency with other column names, it will also make
> it easier for users to group and query all 'sync' related counts
> together.

This seems better, the attached v2 version patch has the changes for the same.

Regards,
Vignesh

On Thu, Nov 13, 2025 at 8:52 PM vignesh C <vignesh21@gmail.com> wrote:
>
> On Thu, 13 Nov 2025 at 09:12, Amit Kapila <amit.kapila16@gmail.com> wrote:
> >
> > On Thu, Nov 13, 2025 at 4:31 AM Peter Smith <smithpb2250@gmail.com> wrote:
> > >
> > > CURRENT
> > > seq_sync_error_count
> > > sync_error_count
> > >
> > > MY SUGGESTION
> > > sync_seq_error_count
> > > sync_table_error_count
> > >
> >
> > +1. Along with consistency with other column names, it will also make
> > it easier for users to group and query all 'sync' related counts
> > together.
>
> This seems better, the attached v2 version patch has the changes for the same.
>

LGTM.

======
Kind Regards,
Peter Smith.
Fujitsu Australia

> On Nov 13, 2025, at 17:52, vignesh C <vignesh21@gmail.com> wrote:
>
> On Thu, 13 Nov 2025 at 09:12, Amit Kapila <amit.kapila16@gmail.com> wrote:
>>
>> On Thu, Nov 13, 2025 at 4:31 AM Peter Smith <smithpb2250@gmail.com> wrote:
>>>
>>> CURRENT
>>> seq_sync_error_count
>>> sync_error_count
>>>
>>> MY SUGGESTION
>>> sync_seq_error_count
>>> sync_table_error_count
>>>
>>
>> +1. Along with consistency with other column names, it will also make
>> it easier for users to group and query all 'sync' related counts
>> together.
>
> This seems better, the attached v2 version patch has the changes for the same.
>
> Regards,
> Vignesh
> <v2-0001-Rename-sync_error_count-to-sync_table_error_count.patch>

V2 overall looks good, a small comment is that the commit message need to be updated:

```
Subject: [PATCH v2] Rename sync_error_count to sync_table_error_count in
 subscription statistics

This patch renames the sync_error_count column to sync_table_error_count in
the pg_stat_subscription_stats view. The new name makes the purpose explicit
```

We now actually rename two fields.

Best regards,
--
Chao Li (Evan)
HighGo Software Co., Ltd.
https://www.highgo.com/

On Wed, Nov 12, 2025 at 7:42 PM Amit Kapila <amit.kapila16@gmail.com> wrote:
>
> On Thu, Nov 13, 2025 at 4:31 AM Peter Smith <smithpb2250@gmail.com> wrote:
> >
> > CURRENT
> > seq_sync_error_count
> > sync_error_count
> >
> > MY SUGGESTION
> > sync_seq_error_count
> > sync_table_error_count
> >
>
> +1. Along with consistency with other column names, it will also make
> it easier for users to group and query all 'sync' related counts
> together.

+1

--
Masahiko Sawada
Amazon Web Services: https://aws.amazon.com