compaction job won't fail and keep running in the job management when one of the nodes stopped

[henry202108]

Please check the FAQ documentation before raising an issue

When start the compaction job if one of the nodes stop the job won't fail and will keep running until manally stop.If another job starts,it will be in queue status

Your Environments (required)

• OS: uname -a
• Compiler: g++ --version or clang++ --version
• CPU: lscpu
• Commit id (e.g. a3ffc7d8)

Linux 192-168-8-191 3.10.0-1160.36.2.el7.x86_64 #1 SMP Wed Jul 21 11:57:15 UTC 2021 x86_64 x86_64 x86_64 GNU/Linux
g++ (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44)
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 4
On-line CPU(s) list: 0-3
Thread(s) per core: 1

f895632

How To Reproduce(required)

Steps to reproduce the behavior:

1. Step 1
   start the compaction job
2. Step 2
   kill one of the storage services
3. Step 3
   the job will not fail and keep running

Expected behavior

In the job management,the job should not keep running and fail when one of the nodes stopped

Additional context

[critical27]

In the job management,the job should not keep running and fail when one of the nodes stopped

I don't agree:

1. Even if some nodes run the job, it still has meaning, which is not useless.
2. Job Mananger will only run a job in async, it is expected behavior when some nodes failed after the job has started, this is by design.