Skip to content
Discussion options

You must be logged in to vote

Hi,

You're experiencing a classic NBD issue where I/O operations get stuck in an uninterruptible state when the server disconnects unexpectedly.

When the NBD server crashes, any pending I/O operations (like those from zpool trim) end up waiting indefinitely in the kernel, which is why you can't kill the process. This happens because:

  1. ZFS issues TRIM operations to the NBD device
  2. The NBD server crashes before completing these operations
  3. The kernel keeps waiting for a response that will never come

You might be able to forcibly disconnect the NBD device with:

nbd-client -d /dev/nbdX

If this is a concern, you may want to look at the "timeout" option of nbd-client https://manpages.debian.org/…

Replies: 1 comment 1 reply

Comment options

You must be logged in to vote
1 reply
@overflowz
Comment options

Answer selected by overflowz
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants
Converted from issue

This discussion was converted from issue #34 on July 22, 2025 23:12.