Database corruption during warp sync

[koenmtb1]

Is there an existing issue?

• I have searched the existing issues

Experiencing problems? Have you tried our Stack Exchange first?

• This is not a support question.

Description of bug

I've tried using warp sync on a new node a few times to get the node up and running quickly.
But twice now have I run into issues with database corruption half way through the sync. The state sync goes fine and it will also successfully import the state; but it will eventually crash when downloading block history.

When I start the node it will throw an error that looks like this:

Run with COLORBT_SHOW_HIDDEN=1 environment variable to disable frame filtering.
Run with RUST_BACKTRACE=full to include source snippets.
2024-08-16 12:23:20 Parity Polkadot    
2024-08-16 12:23:20 ✌️  version 1.15.0-743dc632fd6    
2024-08-16 12:23:20 ❤️  by Parity Technologies <admin@parity.io>, 2017-2024    
2024-08-16 12:23:20 📋 Chain specification: Polkadot    
2024-08-16 12:23:20 🏷  Node name:     
2024-08-16 12:23:20 👤 Role: AUTHORITY    
2024-08-16 12:23:20 💾 Database: RocksDb at /parity/chains/polkadot/db/full    
Error: 
   0: Database error: Corruption: block checksum mismatch: stored = 750900125, computed = 2049841005, type = 4  in /parity/chains/polkadot/db/full/140831.sst offset 1297885 size 27162
   1: Corruption: block checksum mismatch: stored = 750900125, computed = 2049841005, type = 4  in /parity/chains/polkadot/db/full/140831.sst offset 1297885 size 27162

  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ BACKTRACE ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
                                 ⋮ 1 frame hidden ⋮                               
   2: polkadot::main::hd30df8ede3b815fa
      at <unknown source file>:<unknown line>
   3: std::sys_common::backtrace::__rust_begin_short_backtrace::h9e678cfc7f31d6ed
      at <unknown source file>:<unknown line>
   4: main<unknown>
      at <unknown source file>:<unknown line>
   5: __libc_start_main<unknown>
      at <unknown source file>:<unknown line>
   6: _start<unknown>
      at <unknown source file>:<unknown line>

I also have logs from when the node crashed for the first time if that's useful. Let me know if any more information is helpful.
Just as a side note; I have successfully warp synced other nodes perfectly fine so I haven't ruled out hardware issues. From a software perspective these nodes are the exact same (same OS installed with the same ISO and configured exactly the same with Ansible).

Steps to reproduce

Start a node with the following paramaters with the docker image parity/polkadot:stable2407

      --base-path=/parity
      --chain polkadot
      --name ""
      --rpc-methods=unsafe
      --rpc-external
      --prometheus-external
      --rpc-cors all
      --state-pruning 256
      --blocks-pruning 256
      --validator
      --no-private-ip
      --node-key-file /parity/node-key.json
      --sync warp