Re: BUG #17385: "RESET transaction_isolation" inside serializable transaction causes Assert at the transaction end

Tom Lane <tgl@sss.pgh.pa.us>

From: Tom Lane <tgl@sss.pgh.pa.us>
To: Masahiko Sawada <sawada.mshk@gmail.com>
Cc: Michael Paquier <michael@paquier.xyz>, Justin Pryzby <pryzby@telsasoft.com>, Dmitry Koval <d.koval@postgrespro.ru>, Dilip Kumar <dilipbalaut@gmail.com>, andrewbille@gmail.com, pgsql-bugs@lists.postgresql.org, Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Date: 2022-09-24T20:57:51Z
Lists: pgsql-bugs

Attachments

Masahiko Sawada <sawada.mshk@gmail.com> writes:
> I've attached patches.

Per the cfbot, this incorrectly updates the plpgsql_transaction
test.  Here's an updated version that fixes that.

The proposed error message for the GUC_ACTION_SAVE case,

    errmsg("parameter \"%s\" cannot be set temporarily or in functions locally",

does not seem like very good English to me.  In the attached I changed
it to

    errmsg("parameter \"%s\" cannot be set temporarily or locally in functions",

but that isn't great either: it seems redundant and confusing.
I think we could just make it

    errmsg("parameter \"%s\" cannot be set locally in functions",

because that's really the only case users should ever see.

Other than the message-wording issue, I think v6-0001 is OK.

> I did the reorganization of GUC flags in a separate patch (0002 patch)
> since it's not relevant with the new flag GUC_NO_RESET.

0002 seems quite invasive and hard to review compared to what it
accomplishes.  I think we should just keep the flags in the same
order, stick in GUC_NO_RESET in front of GUC_NO_RESET_ALL, and
renumber the flag values as needed.  I did not change 0002 below,
though.

			regards, tom lane

Commits

  1. Renumber GUC flags for a bit more sanity.

  2. Introduce GUC_NO_RESET flag.