Skip to content

v26.12rc1 - #9740

Merged
nGoline merged 1 commit into
ElementsProject:masterfrom
nGoline:release-26.12rc1
Nov 17, 2026
Merged

nGoline merged 1 commit into
ElementsProject:masterfrom
nGoline:release-26.12rc1

Conversation

@nGoline

@nGoline nGoline commented Nov 16, 2026

Copy link
Copy Markdown
Collaborator

Changelog and package versions for v26.12rc1.

Changelog-None

🤖 Generated with Claude Code

@nGoline
nGoline requested a review from cdecker as a code owner November 16, 2026 21:53
@nGoline nGoline added this to the v26.12 milestone Nov 16, 2026
Andezion
Andezion previously approved these changes Nov 16, 2026

@Andezion Andezion left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the maxhops entry duplicates an earlier point release at CHANGELOG.md line 14. The #9627 commits ("askrene: getroutes: add maxhops option", "xpay: call askrene with maxhops", and others) are already in v26.06.10. The 26.06.10 section already says "xpay: Clamp the length of the payment routes using askrene's maxhops option. . ." (line 173). Checklist step 4 says to remove entries that were in earlier point releases. Users on 26.06.10 already have this parameter. So maybe drop line 14. If you want to keep it as a 26.12 highlight, add a note that it was backported to 26.06.10? Also the #8557 splice crash fix is under EXPERIMENTAL at CHANGELOG.md line 101, the commit says Changelog-EXPERIMENTAL, but splicing has been on by default since 26.04 ("Protocol: Splicing is enabled by default", #9021). --experimental-splicing is now a no op (lightningd/options.c at line 1279). Users who skip the EXPERIMENTAL section will miss a crash fix that affects default nodes. Sp maybe do ### Fixed, for example - splicing: fixed crash when splicing a channel which hasn't been announced yet ([#8557])

and v26.06rc1 used the same placeholder as v26.12rc1, is this ok?

sangbida
sangbida previously approved these changes Nov 17, 2026
@daywalker90

Copy link
Copy Markdown
Collaborator

We still have the db migration problem i think:

 ┌────────────────────────────┬──────────────────────────────────────────────────────────────────────────┬──────────────────────────────────────────────────────────────────────────┐
 │ Index                      │ v26.06.10                                                                │ master                                                                   │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 282                        │ ALTER TABLE offers ADD COLUMN force_paths                                │ same                                                                     │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 283                        │ UPDATE channel_funding_inflights (clamp to 1000000)                      │ CREATE TABLE our_outputs                                                 │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 284                        │ UPDATE channel_funding_inflights (clamp to 253)                          │ CREATE TABLE our_txs                                                     │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 285                        │ (none)                                                                   │ migrate_backfill_bwatch_tables                                           │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 286                        │ (none)                                                                   │ ALTER TABLE payments ADD failmsg                                         │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 287                        │ (none)                                                                   │ UPDATE channel_funding_inflights (clamp to 1000000)                      │
 ├────────────────────────────┼──────────────────────────────────────────────────────────────────────────┼──────────────────────────────────────────────────────────────────────────┤
 │ 288                        │ (none)                                                                   │ UPDATE channel_funding_inflights (clamp to 253)                          │
 └────────────────────────────┴──────────────────────────────────────────────────────────────────────────┴──────────────────────────────────────────────────────────────────────────┘

@daywalker90

Copy link
Copy Markdown
Collaborator

#9742

I believe this fix is better than reordering the migrations to not mess with people running master.

@nGoline

nGoline commented Nov 17, 2026

Copy link
Copy Markdown
Collaborator Author

#9742

I believe this fix is better than reordering the migrations to not mess with people running master.

Agreed, creating the tables if they're missing is better than renumbering. The definitions match 283/284, the copy into them already ignores existing rows, and the repeated feerate clamps are idempotent. Let's get it in before RC1.

@nGoline

nGoline commented Nov 17, 2026

Copy link
Copy Markdown
Collaborator Author

the maxhops entry duplicates an earlier point release at CHANGELOG.md line 14. The #9627 commits ("askrene: getroutes: add maxhops option", "xpay: call askrene with maxhops", and others) are already in v26.06.10. The 26.06.10 section already says "xpay: Clamp the length of the payment routes using askrene's maxhops option. . ." (line 173). Checklist step 4 says to remove entries that were in earlier point releases. Users on 26.06.10 already have this parameter. So maybe drop line 14. If you want to keep it as a 26.12 highlight, add a note that it was backported to 26.06.10? Also the #8557 splice crash fix is under EXPERIMENTAL at CHANGELOG.md line 101, the commit says Changelog-EXPERIMENTAL, but splicing has been on by default since 26.04 ("Protocol: Splicing is enabled by default", #9021). --experimental-splicing is now a no op (lightningd/options.c at line 1279). Users who skip the EXPERIMENTAL section will miss a crash fix that affects default nodes. Sp maybe do ### Fixed, for example - splicing: fixed crash when splicing a channel which hasn't been announced yet ([#8557])

and v26.06rc1 used the same placeholder as v26.12rc1, is this ok?

Thanks, both fixed. maxhops stays with "(already in v26.06.10)", since the 26.06.10 notes only mentioned the xpay clamp and never the getroutes parameter. #8557 moved to Fixed with your wording. CODENAME is fine for an RC: the checklist sets the title and namer at the final release, and 26.06rc1 did the same.

@nGoline
nGoline dismissed stale reviews from sangbida and Andezion via 49c827c November 17, 2026 11:18
@nGoline
nGoline force-pushed the release-26.12rc1 branch 2 times, most recently from 49c827c to 1f7a91e Compare November 17, 2026 12:08
Changelog-None

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

@jaonoctus jaonoctus left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

tACK 97a900add7257f74a9c11c95ac165a31a30d94bd

Built with gmake on macOS arm64 (--disable-rust) and tested the database
migration up and down against a real v26.06.10 build on regtest (sqlite3).

v26.06.10 creates the wallet: db version 284, no our_outputs/our_txs.

Up: this commit upgrades it from 284 to 292, creates both tables and adds
payments.failmsg. listfunds is identical before and after, and a new
deposit and a withdraw work afterwards.

Down: tools/lightning-downgrade takes it from 292 to 282 and drops the new
tables and column. v26.06.10 starts on it again (282 to 284) with all
outputs intact.

Up again: from 284 to 292, same listfunds.

No BROKEN log lines in any run. run-migrate_backfill_bwatch_tables and
run-db pass.

Limits: wallet outputs only (no channels, payments or forwards in the db),
and the untagged build needed --database-upgrade=true.
-----BEGIN PGP SIGNATURE-----

iKsEARYKAFMWIQQTBU9gvuuH0rnH1336/g1mZVR61gUCavyeVBsUgAAAAAAEAA5t
YW51MiwyLjUrMS4xMiwwLDMZHGphb25vY3R1c0Bwcm90b25tYWlsLmNvbQAKCRD6
/g1mZVR61utuAQDwV7xrdkVXEU6NYRgWBTsHEwtFs/L5Ij58gJdyxrkP2wD+OAMJ
om/nyacq/Z2zw/F/9l9RWP0pFUphkGkC43lgygg=
=rLoT
-----END PGP SIGNATURE-----

@nGoline
nGoline merged commit 4002c4e into ElementsProject:master Nov 17, 2026
45 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants