[Bug] Incorrect dledger commitlog min offset after mappedFile re delete failed.

[yuz10]

Before Creating the Bug Report

• I found a bug, not just asking a question, which should be created in GitHub Discussions.

• I have searched the GitHub Issues and GitHub Discussions of this repository and believe that this is not a duplicate.

• I have confirmed that this bug belongs to the current repository, not other repositories of RocketMQ.

Runtime platform environment

Ubuntu

RocketMQ version

5.1.0

JDK Version

1.8.0

Describe the Bug

after "re-delete" mapped file failed, the first mapped file is not available, so get min offset should return the first available file's offset.

Steps to Reproduce

1. start a dledger mode cluster
2. produce large amount of messages, the disk is full,
3. the log "the mappedFile re delete failed" is expected:

2023-08-30 16:15:42 WARN DLedgerCleanSpaceService - destroy mapped file[REF:1922311] /opt/data/store/dledger-D2/data/00000000318901321728 Failed. cleanupOver: false
2023-08-30 16:15:42 INFO DLedgerCleanSpaceService - Clean space count=0 timeUp=false checkExpired=true forceClean=true enableForceClean=true diskFull=false storeBaseRatio=0.86 dataRatio=0.86
2023-08-30 16:15:42 WARN DLedgerCleanSpaceService - the mappedFile was destroyed once, but still alive, /opt/data/store/dledger-D2/data/00000000318901321728
2023-08-30 16:15:42 WARN DLedgerCleanSpaceService - destroy mapped file[REF:1922311] /opt/data/store/dledger-D2/data/00000000318901321728 Failed. cleanupOver: false
2023-08-30 16:15:42 WARN DLedgerCleanSpaceService - the mappedFile re delete failed, /opt/data/store/dledger-D2/data/00000000318901321728

4. consume message from earliest offset.

What Did You Expect to See?

consume success.

What Did You See Instead?

consume fail.

Additional Context

No response