Skip to content
  1. Dec 09, 2020
  3. Dec 08, 2020
  5. Dec 07, 2020
  7. Dec 04, 2020
    • Bastian Köcher's avatar
      Add an upper number of maximum parallel runtime api requests (#2069) · f619cdbb
      Bastian Köcher authored 
      

* Add an upper number of maximum parallel runtime api requests

Instead of spawning all runtime api requests in the background and using
all wasm instances. This pr adds a maximum number of parallel requests.

* Update node/core/runtime-api/src/lib.rs

Co-authored-by: default avatarSergei Shulepov <[email protected]>

* Review feedback

* Increase instances

* Add warning

* Update node/core/runtime-api/src/lib.rs

Co-authored-by: default avatarSergei Shulepov <[email protected]>

Co-authored-by: default avatarSergei Shulepov <[email protected]>
      f619cdbb
    • Sergey Pepyakin's avatar
      Add tracing for candidate validation failures in backing (#2071) · ad32069b
      Sergey Pepyakin authored
      ad32069b
    • Peter Goodspeed-Niklaus's avatar
      do not store backed candidates in the provisioner (#1909) · e7e9605f
      Peter Goodspeed-Niklaus authored 
      * guide: non-semantic changes

* guide: update per the issue description

* GetBackedCandidates operates on multiple hashes now

* GetBackedCandidates still needs a relay parent

* implement changes specified in guide

* distinguish between various occasions for canceled oneshots

* add tracing info to getbackedcandidates

* REVERT ME: add tracing messages for GetBackedCandidates

Note that these messages are only sometimes actually passed on to the
candidate backing subsystem, with the consequence that it is
unexpectedly frequent that the provisioner fails to create its
provisionable data.

* REVERT ME: more tracing logging

* REVERT ME: log when CandidateBackingJob receives any message at all

* REVERT ME: log when send_msg sends a message to a job

* fix candidate-backing tests

* streamline GetBackedCandidates

This uses table.attested_candidate instead of table.get_candidate, because
it's not obvious how to get a BackedCandidate from just a
CommittedCandidateReceipt.

* REVERT ME: more logging tracing job lifespans

* promote warning about job premature demise

* don't terminate CandiateBackingJob::run_loop in event of failure to process message

* Revert "REVERT ME: more logging tracing job lifespans"

This reverts commit 7365f2fb3dec988d95cfcd317eba75587fe7fd16.

* Revert "REVERT ME: log when send_msg sends a message to a job"

This reverts commit 58e46aad038e6517d6d56390c8be65b046a21884.

* Revert "REVERT ME: log when CandidateBackingJob receives any message at all"

This reverts commit 0d6f38413c7c66b5e9e81dabc587906fa9f82656.

* Revert "REVERT ME: more tracing logging"

This reverts commit 675fd2628e84d1596965280e7314155ef21b28e6.

* Revert "REVERT ME: add tracing messages for GetBackedCandidates"

This reverts commit e09e156493430b33b6c8ab4b5cedb3f2f91afd51.

* formatting

* add logging message to CandidateBackingJob::run_loop start

* REVERT ME: add tracing to candidate-backing job creation

* run candidatebacking loop even if no assignment

* use unique error variants for each canceled oneshot

* Revert "REVERT ME: add tracing to candidate-backing job creation"

This reverts commit 8ce5f4f0bd7186dade134b118751480f72ea1fd6.

* try_runtime_api more to reduce silent exits

* add sanity check that returned backed candidates preserve ordering

* remove redundant err attribute
      e7e9605f
  9. Dec 03, 2020
  11. Dec 02, 2020
  13. Dec 01, 2020
  15. Nov 30, 2020
    • asynchronous rob's avatar
      small improvements for parachains consensus (#2040) · 414acdfc
      asynchronous rob authored 
      

* introduce a waiting period before selecting candidates and bitfields

* add network_bridge=debug tracing for rep

* change to 2.5s timeout in proposer

* pass timeout to proposer

* move timeout back to provisioner

* grumbles

* Update node/core/provisioner/src/lib.rs

* Fix nitpicks

* Fix bug

Co-authored-by: default avatarBastian Köcher <[email protected]>
Co-authored-by: default avatarBastian Köcher <[email protected]>
      414acdfc
    • asynchronous rob's avatar
      Move candidate validation to the background (#2028) · 10084654
      asynchronous rob authored 
      * refactor some functions to not rely on `self`

* factor out common elements of seconding and attesting

* Add Spawn to backing FromJob

* do candidate validation in background

* tests

* address grumbles
      10084654
    • Bastian Köcher's avatar
      Simplify subsystem jobs (#2037) · 536dceb4
      Bastian Köcher authored 
      * Simplify subsystem jobs

This pr simplifies the subsystem jobs interface. Instead of requiring an
extra message that is used to signal that a job should be ended, a job
now ends when the receiver returns `None`. Besides that it changes the
interface to enforce that messages to a job provide a relay parent.

* Drop ToJobTrait

* Remove FromJob

We always convert this message to FromJobCommand anyway.
      536dceb4
    • Bastian Köcher's avatar
      Process runtime api requests in the background (#2035) · 9ce18622
      Bastian Köcher authored 
      This pr changes how the runtime api subsystem processes runtime api
requests. Instead of answering all of them in the subsystem task and
thus, making all requests sequential, we now answer them in a background
task. This enables us to serve multiple requests at once.
      9ce18622
  17. Nov 28, 2020
  19. Nov 27, 2020
  21. Nov 26, 2020
    • Bastian Köcher's avatar
      Make sure we don't send messages multiple times in the availability distribution subsystem (#2015) · e761c998
      Bastian Köcher authored
      e761c998
    • ordian's avatar
      past-session validator discovery APIs (#2009) · 39a12b68
      ordian authored 
      

* guide: fix formatting for SessionInfo module

* primitives: SessionInfo type

* punt on approval keys

* ah, revert the type alias

* session info runtime module skeleton

* update the guide

* runtime/configuration: sync with the guide

* runtime/configuration: setters for newly added fields

* runtime/configuration: set codec indexes

* runtime/configuration: update test

* primitives: fix SessionInfo definition

* runtime/session_info: initial impl

* runtime/session_info: use initializer for session handling (wip)

* runtime/session_info: mock authority discovery trait

* guide: update the initializer's order

* runtime/session_info: tests skeleton

* runtime/session_info: store n_delay_tranches in Configuration

* runtime/session_info: punt on approval keys

* runtime/session_info: add some basic tests

* Update primitives/src/v1.rs

* small fixes

* remove codec index annotation on structs

* fix off-by-one error

* validator_discovery: accept a session index

* runtime: replace validator_discovery api with session_info

* Update runtime/parachains/src/session_info.rs

Co-authored-by: default avatarSergei Shulepov <[email protected]>

* runtime/session_info: add a comment about missing entries

* runtime/session_info: define the keys

* util: expose connect_to_past_session_validators

* util: allow session_info requests for jobs

* runtime-api: add mock test for session_info

* collator-protocol: add session_index to test state

* util: fix error message for runtime error

* fix compilation

* fix tests after merge with master

Co-authored-by: default avatarSergei Shulepov <[email protected]>
      39a12b68
  23. Nov 25, 2020
    • Bastian Köcher's avatar
      Some code cleanup in overseer (#2008) · 4ce74481
      Bastian Köcher authored 
      * Some code cleanup in overseer

- Switches to select! in the overseer run loop to be more fair about
message processing between the different sources.
- Added a check to only send `ActiveLeaves` if the update actually
contains any data.

* Move the check

* Restore old behavior

* Simplify message sending and signal sending to subsystems

* Update node/subsystem/src/lib.rs
      4ce74481
    • Fedor Sakharov's avatar
      PoV Distribution optimization (#1990) · 994d621f
      Fedor Sakharov authored 
      * Initial commit

* Remove unnecessary struct

* Some review nits

* Update node/network/pov-distribution/src/lib.rs

* Update parachain/test-parachains/adder/collator/tests/integration.rs

* Review nits

* notify_all_we_are_awaiting

* Both ways of peers connections should work the same

* Add mod-level docs to error.rs

* Avoid multiple connection requests at same parent

* Dont bail on errors

* FusedStream for ConnectionRequests

* Fix build after merge

* Improve error handling

* Remove whitespace formatting
      994d621f
  25. Nov 24, 2020
  27. Nov 23, 2020
  29. Nov 22, 2020
    • Bastian Köcher's avatar
      Improve collator side of the collator-protocol (#1955) · b9a9e836
      Bastian Köcher authored 
      * Improve collator side of the collator-protocol

This pr improves the collator-protocol implementation of the collator
side. Besides cleaning up code and rewriting it, the following changed:

- Before on `PeerViewChange` we send an advertisment to every peer, now
this only happens for validators.
- It also adds a check that we send an advertisment message only once
for a connected peer.
- If the same validator was part of the current and next group, we
requested to be connected to this validator two times. This is also
fixed now.
- Instead of having only one connection request, we now are being able
to store multiple of them. This is required as we can have multiple
active leafs at any point of time.

* Switch to common `ConnectionRequests`

* Update node/network/collator-protocol/src/collator_side.rs
      b9a9e836