Horton

Enable data checksums by default

Greg Sabino Mullane <htamfids@gmail.com>

View thread
From: Greg Sabino Mullane <htamfids@gmail.com>
To: pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2024-08-06T22:46:52Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Avoid BufferGetLSNAtomic() calls during nbtree scans.

  2. doc PG 18 relnotes: Add incompatibility note about checksums now default

  3. Fix pg_upgrade's cross-version tests when old < 18

  4. initdb: Change default to using data checksums.

  5. Allow TAP tests to force checksums off when calling init()

  6. initdb: Add new option "--no-data-checksums"

  7. Tweak docs to reduce possible impact of data checksums

Attachments

Please find attached a patch to enable data checksums by default.

Currently, initdb only enables data checksums if passed the
--data-checksums or -k argument. There was some hesitation years ago when
this feature was first added, leading to the current situation where the
default is off. However, many years later, there is wide consensus that
this is an extraordinarily safe, desirable setting. Indeed, most (if not
all) of the major commercial and open source Postgres systems currently
turn this on by default. I posit you would be hard-pressed to find many
systems these days in which it has NOT been turned on. So basically we have
a de-facto standard, and I think it's time we flipped the switch to make it
on by default.

The patch is simple enough: literally flipping the boolean inside of
initdb.c, and adding a new argument '--no-data-checksums' for those
instances that truly want the old behavior. One place that still needs the
old behavior is our internal tests for pg_checksums and pg_amcheck, so I
added a new argument to init() in PostgreSQL/Test/Cluster.pm to allow those
to still pass their tests.

This is just the default - people are still more than welcome to turn it
off with the new flag. The pg_checksums program is another option that
actually argues for having the default "on", as switching to "off" once
initdb has been run is trivial.

Yes, I am aware of the previous discussions on this, but the world moves
fast - wal compression is better than in the past, vacuum is better now,
and data-checksums being on is such a complete default in the wild, it
feels weird and a disservice that we are not running all our tests like
that.

Cheers,
Greg