ci: Downgrade DEBUG=1 to -D_GLIBCXX_ASSERTIONS in centos task #32586

1. 
   maflcko commented at 11:48 am on May 22, 2025: member

   to work around #32524 (comment)

   closes #32524

2. 
   DrahtBot commented at 11:48 am on May 22, 2025: contributor

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

   Code Coverage & Benchmarks

   For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/32586.

   Reviews

   See the guideline for information on the review process.

   Type	Reviewers
   ACK	laanwj, fanquake

   If your review is incorrectly listed, please react with 👎 to this comment and the bot will ignore it on the next update.

   Conflicts

   Reviewers, this pull request conflicts with the following ones:

   • #31802 (Add bitcoin-{node,gui} to release binaries for IPC by Sjors)

   If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

3. DrahtBot added the label Tests on May 22, 2025
4. ci: Downgrade DEBUG=1 to -D_GLIBCXX_ASSERTIONS in centos task

   to work around https://github.com/bitcoin/bitcoin/issues/32524#issuecomment-2900903169

   fa079538e3
5. maflcko force-pushed on May 22, 2025
6. laanwj approved
7. 
   laanwj commented at 12:47 pm on May 22, 2025: member

   ACK fa079538e32d20aec6786c93e1117da1e8ea0cab
8. 
   fanquake commented at 1:06 pm on May 22, 2025: member

   Not sure; now we wouldn’t have any (non-msan, non-32 bit task) CI using DEBUG=1? It’d at least be good to note in the CI config, why this is being changed this way / when it could be removed.
9. 
   maflcko commented at 1:12 pm on May 22, 2025: member

   Not sure; now we wouldn’t have any (non-msan, non-32 bit task) CI using DEBUG=1? It’d at least be good to note in the CI config, why this is being changed this way / when it could be removed.

   Of course there is no right or wrong answer here, but I’d say that the libc++ debug build (msan) and glibc++ debug build (32-bit) are probably enough and redundant with this one anyway, so one could even fully remove the debug setting here. (Also, there is other, possibly more important stuff, only run in nightly CIs outside this repo, including a full gcc debug run, so based on that it should also be fine)

10. 
    laanwj commented at 1:37 pm on May 22, 2025: member

    Not sure; now we wouldn’t have any (non-msan, non-32 bit task) CI using DEBUG=1? It’d at least be good to note in the CI config, why this is being changed this way / when it could be removed.

    In the short term, getting the CI to pass reliably again is most important imo. Adding another DEBUG run can always be considered, but shouldn’t come at the cost of CI flakiness.

11. fanquake approved
12. 
    fanquake commented at 4:58 pm on May 22, 2025: member

    ACK fa079538e32d20aec6786c93e1117da1e8ea0cab - we can followup
13. fanquake merged this on May 22, 2025
14. fanquake closed this on May 22, 2025

    ---

15. maflcko deleted the branch on May 23, 2025
16. 
    maflcko commented at 3:55 pm on May 23, 2025: member

    (edited a prior comment of mine, which said there are two libc++ debug runs in the CI. In reality there is the libc++ debug build (msan) and the glibc++ debug build (32-bit).)

Contributors
maflcko DrahtBot laanwj fanquake

Labels
Tests