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

Rack Unlock: Parallelize share retrieval and only retrieve shares not already retrieved #514

Closed
andrewjstone opened this issue Dec 14, 2021 · 1 comment
Closed

Rack Unlock: Parallelize share retrieval and only retrieve shares not already retrieved #514

andrewjstone opened this issue Dec 14, 2021 · 1 comment
Assignees
@andrewjstone
Labels
bootstrap services For those occasions where you want the rack to turn on

Comments

@andrewjstone
Copy link
Contributor

Currently in the rack unlock code (as of #487), shares are retrieved serially, and we always try to contact all remote sleds to retrieve them when we fail to get a quorum. We really only want to contact sleds that haven't responded when we retry, and also make retrievals work in parallel.
@andrewjstone andrewjstone added the bootstrap services For those occasions where you want the rack to turn on label Dec 14, 2021
@andrewjstone andrewjstone self-assigned this Dec 14, 2021
@andrewjstone
Copy link
Contributor Author

This code has changed significantly now that sprockets is in play.

leftwo pushed a commit that referenced this issue Sep 11, 2023
Update Crucible and Propolis to latest
246a139 
Crucible:
update rust crate base64 to 0.21.3 (#913)
update rust crate slog-async to 2.8 (#915)
update rust crate async-recursion to 1.0.5 (#912)
Move active jobs into a separate data structure and optimize `ackable_work` (#908)
Check repair IDs correctly (#910)
update actions/checkout action to v4 (#903)
update rust crate tokio to 1.32 (#890)
Remove single-item Vecs (#898)
Move "extent under repair" into a helper function (#907)
offset_mod shouldn't be randomized (#905)
Only rehash if a write may have failed (#899)
Make negotiation state an enum (#901)
Test update for fast write ack and gather errors on test failure (#897)

Propolis:
Update cpuid-gen util for better coverage
Make storage backend config more flexible and consistent
Use correct register sizes for PIIX3 PM device
Update bitflags dependency
fix softnpu port order (#517)
Use hex formatting for unhandled MMIO/PIO/MSRs
Update deps for new crucible and oximeter
Update standalone-with-crucible docs (#514)
@leftwo leftwo mentioned this issue Sep 11, 2023
Merged
leftwo added a commit that referenced this issue Sep 12, 2023
@leftwo
Update Crucible and Propolis to latest (#4079)
25c73f4 
Crucible:
update rust crate base64 to 0.21.3 (#913)
update rust crate slog-async to 2.8 (#915)
update rust crate async-recursion to 1.0.5 (#912)
Move active jobs into a separate data structure and optimize
`ackable_work` (#908) Check repair IDs correctly (#910)
update actions/checkout action to v4 (#903)
update rust crate tokio to 1.32 (#890)
Remove single-item Vecs (#898)
Move "extent under repair" into a helper function (#907) offset_mod
shouldn't be randomized (#905)
Only rehash if a write may have failed (#899)
Make negotiation state an enum (#901)
Test update for fast write ack and gather errors on test failure (#897)

Propolis:
Update cpuid-gen util for better coverage
Make storage backend config more flexible and consistent Use correct
register sizes for PIIX3 PM device
Update bitflags dependency
fix softnpu port order (#517)
Use hex formatting for unhandled MMIO/PIO/MSRs
Update deps for new crucible and oximeter
Update standalone-with-crucible docs (#514)

Co-authored-by: Alan Hanson <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@andrewjstone andrewjstone

Labels
bootstrap services For those occasions where you want the rack to turn on
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@andrewjstone