MySQL Bugs: #116933: Function recv_single_rec doesn't handle corrupt

Bug #116933	Function recv_single_rec doesn't handle corrupt_log correctly
Submitted:	11 Dec 2024 1:47	Modified:	11 Dec 2024 5:31
Reporter:	mengchu shi (OCA)	Email Updates:
Status:	Verified	Impact on me:	None
Category:	MySQL Server: InnoDB storage engine	Severity:	S3 (Non-critical)
Version:	8.0,8.4	OS:	Any
Assigned to:		CPU Architecture:	Any
Tags:	Contribution, recovery crash, redo apply

Description:
Function recv_single_rec doesn't handle corrupt_log while function recv_multi_rec return true immediatly.
If this parse round happens to reach the memory bottleneck to call function recv_apply_hashed_log_recs, This bug will cause redo logs after the corrupt_log be applied and maybe corrupt datas.
I think it's not by design, right?

How to repeat:
Just see the codes please:
```c++
static bool recv_single_rec(byte *ptr, byte *end_ptr) {
  ...
  if (recv_sys->found_corrupt_log) {
    recv_report_corrupt_log(ptr, type, space_id, page_no);

#ifdef UNIV_HOTBACKUP
    return true;
#endif /* UNIV_HOTBACKUP */

  } else if (len == 0 || recv_sys->found_corrupt_fs) {
    return true;
  }
  ...
}
```
```c++
static bool recv_multi_rec(byte *ptr, byte *end_ptr) {
  ...
    if (recv_sys->found_corrupt_log) {
      recv_report_corrupt_log(ptr, type, space_id, page_no);

      return true;

    } else if (len == 0) {
      return true;

    } else if ((*ptr & MLOG_SINGLE_REC_FLAG)) {
      recv_sys->found_corrupt_log = true;

      recv_report_corrupt_log(ptr, type, space_id, page_no);

      return true;

    } else if (recv_sys->found_corrupt_fs) {
      return true;
    }
...
}
```

Suggested fix:
Function recv_single_rec should handle corrupt_log the same as function recv_multi_rec. So remove #ifdef UNIV_HOTBACKUP.

recv_single_rec_should_handle_corrupt_log.patch (*) I confirm the code being submitted is offered under the terms of the OCA, and that I am authorized to contribute it.

Contribution: recv_single_rec_should_handle_corrupt_log.patch (application/octet-stream, text), 526 bytes.

Hello mengchu shi,

Thank you for the report and contribution.

regards,
Umesh