[30.x] 30.3rc1 #35452

1. 
   fanquake commented at 5:02 PM on June 3, 2026: member

   Backports:

   • #34918
   • #35465

   Final changes for v30.3rc1.

   Backports since 30.2:

   • #34283
   • #34459
   • #34689
   • #34754
   • #34805
   • #34856
   • #35211
   • #35232
2. fanquake added this to the milestone 30.3 on Jun 3, 2026
3. DrahtBot added the label Backport on Jun 3, 2026
4. 
   DrahtBot commented at 5:02 PM on June 3, 2026: contributor

   <!--e57a25ab6845829454e8d69fc972939a-->

   The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

   <!--006a51241073e994b41acfe9ec718e94-->

   Code Coverage & Benchmarks

   For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/35452.

   <!--021abf342d371248e50ceaed478a90ca-->

   Reviews

   See the guideline for information on the review process.

   Type	Reviewers
   ACK	sedited, marcofleon

   If your review is incorrectly listed, please copy-paste <code>&lt;!--meta-tag:bot-skip--&gt;</code> into the comment that the bot should ignore.

   <!--5faf32d7da4f0f540f40219e4f7537a3-->

5. 
   andrewtoth commented at 8:50 PM on June 3, 2026: contributor

   Can we have something like this in the release notes. Also for 29 point release (and possibly 31 if we don't have an automatic compaction patch merged):

   This fixes an issue where the chainstate database would repeatedly rewrite large portions of itself, causing excessive disk reads and writes during normal operation.

   As a side effect, the chainstate will be around 4 GB larger after an initial sync. To reclaim this space, restart the node with -forcecompactdb=1 after the initial sync has finished. It should be removed on subsequent restarts. Leaving it set will cause significant disk reads and writes every time the node is restarted.

6. lint: disable leveldb subtree check

   This is no-longer a proper subtree, because of direct cherry-picks.

   d09bd0f402
7. fanquake force-pushed on Jun 17, 2026
8. 
   fanquake commented at 6:07 AM on June 17, 2026: member

   Can we have something like this in the release notes.

   Pulled that into the release notes.

9. coins: test chainstate flush baseline

   Add `CDBWrapper::GetProperty()` and expose it through `CCoinsViewDB::GetDBProperty()` so coins tests can inspect LevelDB runtime properties through the coins view.
   Use it in a coins DB flush baseline that records the LevelDB layout after flushing while keeping readback coverage for the flushed coin and best block.
   
   Co-authored-by: Andrew Toth <andrewstoth@gmail.com>
   
   Github-Pull: #35465
   Rebased-From: b10889d10752c5d5e4954af2959f7bdff47bd67c

   55ead701a3
10. fanquake force-pushed on Jun 22, 2026
11. fanquake marked this as ready for review on Jun 22, 2026
12. fanquake force-pushed on Jun 22, 2026
13. fanquake force-pushed on Jun 22, 2026
14. fanquake force-pushed on Jun 22, 2026
15. fanquake force-pushed on Jun 22, 2026
16. fanquake force-pushed on Jun 22, 2026
17. validation: randomly compact chainstate

    Full chainstate flushes are convenient maintenance points for long-term LevelDB cleanup because the chainstate was just written.
    Randomize the trigger so nodes that flush near the same height do not compact together.
    
    Add blocking chainstate compaction through `CCoinsViewDB::CompactFull()` and give each post-IBD full flush on the normal chainstate a 1/320 chance to start compaction.
    With hourly flushes this averages roughly every two weeks and makes a six-month miss about one in a million.
    This keeps the schedule stateless and leaves last-compaction height or timestamp bookkeeping out of chainstate metadata.
    
    Co-authored-by: Andrew Toth <andrewstoth@gmail.com>
    
    Github-Pull: #35465
    Rebased-From: aa021b26f39fd231b2a3aac5780d5113a4aea639

    992b1dd39f
18. coins: compact chainstate in background

    Full chainstate compaction can take minutes on large databases.
    Move `CCoinsViewDB::CompactFull()` to a named `utxocompact` one-shot background thread so validation only schedules the work.
    
    When validation selects compaction after a full flush, the chainstate was just written and another write is less likely to be needed immediately.
    The coins view destructor waits for completion, and a mutex prevents compaction from using `m_db` while `ResizeCache()` replaces it.
    
    Co-authored-by: Andrew Toth <andrewstoth@gmail.com>
    
    Github-Pull: #35465
    Rebased-From: 394e473d42ba1383dfec45a3eafa8a73a09dbe8b

    b7c7d1ea28
19. fuzz: Remove unused g_setup pointers

    These are unused and removing them avoids clang warnings like:
    
    src/test/fuzz/deserialize.cpp:42:26: error: variable g_setup set but not used [-Werror,-Wunused-but-set-variable]
    
    Github-Pull: #34918
    Rebased-From: fabbfec3b00c138a28034a4f5594305d2220b9bb

    e7c6d391ef
20. doc: update release notes for v30.3rc1 1bdd2b1334
21. build: bump version to v30.3rc1 a1406d6333
22. doc: update manual pages for v30.3rc1 d52747d8e7
23. in src/validation.cpp:2917 in cbdb11da24

    2915 | +        if (m_chainman.m_options.signals) {
    2916 | +            // Update best block in wallet (so we can detect restored wallets).
    2917 | +            m_chainman.m_options.signals->ChainStateFlushed(this->GetRole(), GetLocator(m_chain.Tip()));
    2918 | +        }
    2919 | +
    2920 | +        if (!m_chainman.m_interrupt && m_chainman.m_chainstates.size() == 1) { // Skip AssumeUTXO
    

    ---

    ---

    

    sedited commented at 12:07 PM on June 22, 2026:

    In commit cbdb11da241cd39cc2d1813369fa25e1987d98c5

    I think you want m_chainman.GetAll().size() here instead of m_chainman.m_chainstates.size() (and probably in the 29 backport too to make the ancestor commit CI pass).

24. fanquake force-pushed on Jun 22, 2026
25. sedited approved
26. 
    sedited commented at 1:33 PM on June 22, 2026: contributor

    ACK d52747d8e74cc0ebec252c4f5680d184fb655e80

27. fanquake requested review from marcofleon on Jun 22, 2026
28. 
    marcofleon commented at 4:40 PM on June 22, 2026: contributor

    ACK d52747d8e74cc0ebec252c4f5680d184fb655e80

29. sedited merged this on Jun 23, 2026
30. sedited closed this on Jun 23, 2026

    ---

31. 
    l0rinc commented at 7:50 PM on June 23, 2026: contributor

    nit: we might want to mention that #35083 was also backported since 30.2

    ACK d52747d8e74cc0ebec252c4f5680d184fb655e80

32. fanquake deleted the branch on Jun 24, 2026

Milestone
30.3