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.

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

Node allocator latency_groups causes just single node to be used for writes #8496

Open
guymguym opened this issue Oct 31, 2024 · 0 comments
Open

Comments

@guymguym
Copy link
Member

Environment info

  • NooBaa Version: 5.15
  • Platform: ODF-4.15

Actual behavior

  1. Using a PVPool of 2 or 4 we can see that just a single PV writes at a time.
  2. This happens due to this code -
    if (nodes_with_avg_disk_write.length >= config.NODE_ALLOCATOR_NUM_CLUSTERS) {
  3. We can workaround this issue by setting the following config which eliminated this code path:
oc set env sts noobaa-core CONFIG_JS_NODE_ALLOCATOR_NUM_CLUSTERS=99999

Expected behavior

  1. All PV's should be writing in parallel.

Steps to reproduce

  1. Write to a bucket that uses a PVpool.

More information - Screenshots / Logs / Other output

This is a benchmark running before the workaround - in the write stage you can see a single PV active, and in the read stage both are active:
image

After the workaround both are writing together.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant