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

Expose listener address when configured for port 0 #69

Open
bmoylan opened this issue Apr 16, 2019 · 0 comments
Open

Expose listener address when configured for port 0 #69

bmoylan opened this issue Apr 16, 2019 · 0 comments

Comments

@bmoylan
Copy link
Contributor

bmoylan commented Apr 16, 2019

Configuring the server on port 0 tells the kernel to assign a random port and return the address. We could handle this more nicely:

  • easy: pass instantiated listener to ServeTLS instead of using ListenAndServeTLS so we can log the address
  • more involved: expose this info over an API (maybe on the *Server object?)
blakehawkins pushed a commit to blakehawkins/witchcraft-go-server that referenced this issue Feb 13, 2020
* support dynamic allocation of executors

* Adds a SoftReservationStore which keeps track of in-memory reservations for extra executors and handles failover between schedulers

* Removes leftover method

* Fixes tests

* Checkstyle fixes

* Modifies CircleCI config to publish a snapshot release

* Circle branch updat

* Updates Circle published image tag

* Revert "Updates Circle published image tag"

This reverts commit 3800676c14ebb3a3243484134ea57626080263fe.

* Addresses some PR comments

* Fixes storeLock removal

* checkstyle fixes

* Modifies Circle config to publish dirty release

* Fixes a bug in failover for stale drivers which was specifying bounds out of range when executor count is less than min

* Reverts Circle config to develop

* Renames dynamic allocation annotation keys to be more explicit

* Removes the logic for handling Fifo priority from the extra executor code path to reduce complexity for now

It is still a correct behavior we probably want to reintroduce in the long term, but for now will make it easier to reason about dynamic allocation

* Always marks the executor status as false in the SoftReservation object to eliminate a race between the event handling and the executor scheduling request

* Removes redundant debug log line

* Adds tests for different dynamic allocation scenarios the extender might get

* Resets unintentional changes after merge

* Fixes merge conflicts

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

No branches or pull requests

1 participant