Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

Portal hub stream-subscription attach times out during silo churn, latching the circuit into cross-process-routing-disabled

Closed
  • Is something wrong?
#6,429 2 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
22/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Active
Tech stack
csharp
Domain
backend, distributed-systems

Research direction

Start in src/MeshWeaver.Connection.Orleans/OrleansRoutingService.cs at SubscribeWhenStreamingReadyAsync (around line 1949) and trace where the RegisterConsumer TimeoutException goes, including the IsTransientFailure check and the retry budget added in #2633. Done means establishing whether a 30 s timeout is retried or latches the circuit, then agreeing the fix with maintainers, since the issue asks for that decision.

Written by the indexing model from the issue text.

Description

bug sev:M

What is failing

A portal hub's Orleans Memory stream subscription failed to attach: the IPubSubRendezvousGrain.RegisterConsumer call from OrleansRoutingService.SubscribeWhenStreamingReadyAsync timed out after the full 30 s budget against the rendezvous grain on a remote silo. The hub was consequently latched into "cross-process routing DISABLED" for its circuit — in-process routing remains active, but the circuit cannot be reached over the stream fallback leg until it re-registers.

Probable cause

The RegisterConsumer call could not complete because the silo hosting the rendezvous grain was mid-departure or unresponsive — this fired in the closing minutes of a rolling-deploy churn window that dominates the log today (the SiloUnavailableException stream-leg incidents on the memory-* pods immediately precede it). Per Doc/Architecture/OrleansStreamPubSubDurability, since #2633 the attach is retried on a bounded, loud budget gated on the same IsTransientFailure the delivery leg uses — a 30 s TimeoutException either exhausted that budget or was not classified as transient by it, and the failure latched. Confidence: medium — the churn correlation is circumstantial; the open question for whoever picks this up is whether System.TimeoutException from the RegisterConsumer call reaches the IsTransientFailure predicate at all, and whether a 30 s per-attempt timeout can outrun the retry budget in practice.

Impact

Narrow so far, and narrower than this log reads: since the #1742 transport swap, the primary route to a pod-process hub is a directed grain call — the stream carries only the roll window and client-owned hubs. So the loss is confined to the stream fallback leg for the one affected portal circuit on the one pod that logged it (counts in the footer): other silos' deliveries to that circuit will be refused by the HasLiveSubscriber check with a DeliveryFailure{NotFound} rather than silently dropped, and in-process traffic is unaffected. The circuit stays in this state until it re-registers.

Where to look

MeshWeaver.Connection.Orleans/OrleansRoutingService.cs — SubscribeWhenStreamingReadyAsync (the top frame, line 1949 in the sample) and the bounded retry budget added in #2633. Check how the RegisterConsumer TimeoutException flows into that budget, and whether the budget should be sized against the 30 s grain-call timeout rather than attempt count. Context: Orleans Stream Pub-Sub Durability (residual "attach failed" section — this is exactly the genuine race plus the permanent-attach-failure case that page says still fires at Critical); related: #2633, #1742, #2320, #2322, Pod-Hub Delivery Roll Plan.

Already known?

Not a duplicate of any open incident, but it is the documented residual: the durability page predicts precisely this signal surviving after #2633 whenever a genuinely dead target meets the retry budget. The surrounding SiloUnavailableException incidents share the roll-churn trigger but are distinct fingerprints; if this recurs on every roll, consider folding it into the #2320/#2322 stream-leg work rather than fixing the retry loop in isolation.

Evidence
Fingerprint fb31e284cf14f42d
Category MeshWeaver.Connection.Orleans.OrleansRoutingService
Severity Critical
Exception System.TimeoutException
Top frame MeshWeaver.Connection.Orleans.OrleansRoutingService.SubscribeWhenStreamingReadyAsync(Address address, AsyncDelivery callback, CancellationToken ct)
Namespace memex-cloud
Pods memex-portal-deployment-85587f8869-prgnp
Occurrences 1
First seen 2026-10-10 08:15:52Z
Last seen 2026-10-10 08:15:52Z
Routing not determined — no configured route matches the category MeshWeaver.Connection.Orleans.OrleansRoutingService. This repository is the configured fallback, not a finding about who owns the fault; the category names the LOGGER, which may not be the subject.
Recent log lines 
2026-10-10 08:15:52Z memex-portal-deployment-85587f8869-prgnp crit: MeshWeaver.Connection.Orleans.OrleansRoutingService[0]
      Orleans 'Memory' stream subscription could not be attached for portal/E55k_sEouET-v-GxdzPVVsJFmTD8m7ksYyrEtNnU6JI — cross-process routing for this hub is DISABLED (in-process routing remains active)
      System.TimeoutException: Response did not arrive on time in 00:00:30 for message: Request [S10.244.17.58:11111:150624528 sys.client/hosted-10.244.17.58:11111@150624528]->[S10.244.6.128:11111:150624604 pubsubrendezvous/Memory/null/portal/E55k_sEouET-v-GxdzPVVsJFmTD8m7ksYyrEtNnU6JI] Orleans.Streams.IPubSubRendezvousGrain.RegisterConsumer(Orleans.Runtime.GuidId, Orleans.Runtime.QualifiedStreamId, Orleans.Runtime.GrainId, System.String, System.Threading.CancellationToken) #3DCF28F662E35FFF. 
         at Orleans.Serialization.Invocation.ResponseCompletionSource.System.Threading.Tasks.Sources.IValueTaskSource.GetResult(Int16 token) in /_/src/Orleans.Serialization/Invocation/ResponseCompletionSource.cs:line 100
         at System.Threading.Tasks.ValueTask.ValueTaskSourceAsTask.<>c.<.cctor>b__4_0(Object state)
      --- End of stack trace from previous location ---
         at Orleans.Streams.StreamConsumer`1.SubscribeAsyncImpl(IAsyncObserver`1 observer, IAsyncBatchObserver`1 batchObserver, StreamSequenceToken token, Nullable`1 startPosition, String filterData) in /_/src/Orleans.Streaming/Internal/StreamConsumer.cs:line 132
         at MeshWeaver.Connection.Orleans.OrleansRoutingService.SubscribeWhenStreamingReadyAsync(Address address, AsyncDelivery callback, CancellationToken ct) in /home/runner/work/MeshWeaver/MeshWeaver/src/MeshWeaver.Connection.Orleans/OrleansRoutingService.cs:line 1949

Opened automatically from Admin/_LogIncident/fb31e284cf14f42d. Recurrences are folded into this issue rather than opening new ones.
It also stands for the whole log site 2735188207964df5: other fingerprints of this site fold in here as comments rather than opening tickets of their own.

Dominant language
C#
Stars
12
Forks
5
Avg merge
4h 24m
Merged PRs (30d)
961

Getting set up

  • No Dockerfile or Docker Compose file
  • Has a pull request template
  • No contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from Systemorph/MeshWeaver

All issues in Systemorph/MeshWeaver

Similar issues

More C# issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.