Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Diagnose vastly varying RocksDB disk read latency #621

Closed
willmeister opened this issue Nov 15, 2018 · 1 comment
Closed

Diagnose vastly varying RocksDB disk read latency #621

willmeister opened this issue Nov 15, 2018 · 1 comment
Assignees
@willmeister

Comments

@willmeister
Copy link
Contributor

willmeister commented Nov 15, 2018 •
edited
Loading

On the mana-ethereum-sync-2 box, single Trie node RocksDB disk reads can take anywhere from 1 microsecond to 1 second and they vary a lot.

This task is to determine the cause and see if it's an issue with the setup or the box or somehow the code.

Things to test:

  • Is IO contention an issue?
    • We noticed this issue with parity writing to the same mount /mnt/blockstorage
  • Is RocksDB the issue?
  • Other?
This was referenced Nov 15, 2018
Open
Open
@willmeister willmeister self-assigned this Nov 15, 2018
@willmeister
Copy link
Contributor Author

Closing this as I was unable to reproduce this on my laptop. I'm guessing IO contention with parity and a slow mounted drive combine to create this variance. It's possible that RocksDB configuration could make this better, but that's tracked in #624.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@willmeister willmeister

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@willmeister