Reindexing download issue

[dwarf_racoon]

On a standard btc node, I had a problem akin to that of https://bitcointalk.org/index.php?topic=5319613.0 so went to reindex with a './bitcoind --reindex' and the daemon seems unable to download 100% of the blockchain. I started the command about a week ago, and for the past ~4 days it has been stuck right around 90.8%. Not that the daemon does not continue to attempt, but the percentage downloaded fluctuates between ~90.65 and 90.90%.

I have found a few related issues on the stackexhange that say this could be due to faulty peers or even a storage issue, but neither of these seem immediately apparent. Beyond this, several answers report that the reindex cannot be cancelled, so I am unsure of how to solve this issue. Is my node fried? Is there a way to change my 'dbcache' mid download (as https://bitcoin.stackexchange.com/questions/38422/possible-to-speed-up-reindex gives this as a potential alleviation, though without specifying the location of dbcache)?

Any info on the above would be significantly appreciated as I attempt to re-activate my node.

[BitMaxz]

Can you add more details about your nodes your current hard drive space, ram, and OS?

Have you tried to upgrade the Bitcoin core? There was a new release last month Bitcoin Core version 27.0 it might fix your issue if you upgraded it to the latest version if the drive is not the issue.

About dbcache you can stop the node and edit the Bitcoin.conf file and add this "dbcache=(Value)" The value depends on your current RAM size.

[Cricktor]

Do you see any errors in your debug.log file? If yes, can you post portions of those here (sanitize private data before)?

Your first linked thread is about issues with content of settings.json file. Are you sure you have a similar problem?

If your Bitcoin Core is stuck or slow with syncing but doesn't spit out some errors, I don't see an immediate reason for a reindex run, unless you killed bitcoind by force and corrupted some of its databases or indices.

Remember to always shutdown bitcoind gracefully via bitcoin-cli stop and monitor the tail of debug.log to finally say:
yyyy-mm-ddThh:mm:ssZ Shutdown: done

[nc50lc]

-snip- Beyond this, several answers report that the reindex cannot be cancelled, so I am unsure of how to solve this issue. Is my node fried? Is there a way to change my 'dbcache' mid download

It should continue where you left it.
Probably, those reports have restarted Bitcoin Core without removing --reindex arg, doing so will restart the reindex process from start.

If you closed Core properly (./bitcoin-cli stop) and considering that there's no other issue, then, your node should resume the progress when you shutdown your node.

But it should be SOP to check the logs before using commands like --reindex since it may not be the best solution that you need.
And now, there should be an indication why it's stalled or at least tell that there's no issue but slow sync progress.