[pgpool-hackers: 4471] [PATCH] printing empty row first in query "show pool_process"

[keiseo]

Hello, I want to fix a minor thing, which is might not a critical thing.

kashif_zeeshan([kashif.z at bitnine.net](https://www.pgpool.net/mailman/listinfo/pgpool-hackers)) also shared this to me, and I had the same thing.

-- reporting (cutoff)

The problem is "show processes" query showing empty row result first.

postgres at master:/tmp$ psql -h localhost -p 11000 -c "show pool_processes;"

pool_pid | start_time | cl

----------+------------+---

(0 rows)

pool_pid |

----------+----------------

56 | 2024-06-12 09:2

57 | 2024-06-12 09:2

58 | 2024-06-12 09:2

(3 rows)

( tested with pool: 4.4.6 (nurikoboshi), pg: PostgreSQL 15.7 on x86_64-pc-linux-gnu, compiled by gcc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0, 64-bit )

you can reproduce this since pool V4_2, with pg 15,16.

-- analysis

In short, because of send_row_description calling twice, psql like client showing emptry row result first before actual contents, that are, process informations.

Since branch V4_2, anew function was added which is the send_row_description_and_data_rows().

That function sends these FE/BE protocol data to the destination as properly like this.

RowDescription

DataRow

CommandComplete

ReadyForQuery

But the problem is, since V4_2, process_reporting() function itself sends FE/BE protocol data as a result like this.

RowDescription

RowDescription

DataRow

CommandComplete

ReadyForQuery

First RowDescription called without following Datarow protocol data. So I think, this is the main reason why zero row printed

and the reason why RowDescription is called twice, send_row_description() is still called before send_row_description_and_data_rows().

by the way, V4_1 is fine because calling send_row_description() then sending DataRow.

-- additional reporting.

The one thing confusing my decision is thepg14 does not print empty row first so totally looks fine like this.

postgres at master:/tmp$ psql -h localhost -p 11000 -c "show pool_processes;"

pool_pid | start_time

----------+---------------------

56 | 2024-06-12 09:58:08

58 | 2024-06-12 09:58:08

59 | 2024-06-12 09:58:08

(3 rows)

( tested with pool: 4.4.6 (nurikoboshi), pg: PostgreSQL 14.12 on x86_64-pc-linux-gnu, compiled by gcc (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0, 64-bit )

-- suggestion

Even if it looks like fine, I think that we should following FE/BE protocol properly.

Oh. This is my suggestion, and you can suggest or tell me whatever you want.

anyway, Could you please review my opinion?

-- attachment

I made patch files similarly like others do. well...

- level 5 logs for FE/BE protocol tracking.
- master branch patch
- V4_2 branch patch
- V4_3 branch patch
- V4_4 branch patch
- V4_5 branch patch

Regards

Kwangwon Seo

Bitnine Global Inc.

www.bitnine.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bug_pool_process.log
Type: text/x-log
Size: 18287 bytes
Desc: not available
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0006.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-pgpool-show-processes-showing-empty-row-first_V4_2_STABLE.patch
Type: text/x-patch
Size: 1102 bytes
Desc: not available
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0007.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-pgpool-show-processes-showing-empty-row-first_master.patch
Type: text/x-patch
Size: 1102 bytes
Desc: not available
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0008.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-pgpool-show-processes-showing-empty-row-first_V4_5_STABLE.patch
Type: text/x-patch
Size: 1102 bytes
Desc: not available
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0009.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-pgpool-show-processes-showing-empty-row-first_V4_4_STABLE.patch
Type: text/x-patch
Size: 1102 bytes
Desc: not available
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0010.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-Fix-pgpool-show-processes-showing-empty-row-first_V4_3_STABLE.patch
Type: text/x-patch
Size: 1102 bytes
Desc: not available
URL: <http://www.pgpool.net/pipermail/pgpool-hackers/attachments/20240612/2c2433e7/attachment-0011.bin>