Download Latest Version v2.10.1 source code.tar.gz (20.4 MB)
Email in envelope

Get an email when there's a new version of Seldon Core

Home / v2.10.1
Name Modified Size InfoDownloads / Week
Parent folder
servers.yaml 2025-10-20 762 Bytes
0
certs.yaml 2025-10-20 3.6 kB
0
components.yaml 2025-10-20 50.6 kB
0
crds.yaml 2025-10-20 2.3 MB
0
runtime.yaml 2025-10-20 1.4 kB
0
seldon-core-v2-certs-2.10.1.tgz 2025-10-20 721 Bytes
0
seldon-core-v2-crds-2.10.1.tgz 2025-10-20 256.7 kB
0
seldon-core-v2-runtime-2.10.1.tgz 2025-10-20 1.2 kB
0
seldon-core-v2-servers-2.10.1.tgz 2025-10-20 814 Bytes
0
seldon-core-v2-setup-2.10.1.tgz 2025-10-20 16.7 kB
0
seldon-linux-amd64 2025-10-20 74.8 MB
0
README.md 2025-10-20 1.9 kB
0
v2.10.1 source code.tar.gz 2025-10-20 20.4 MB
0
v2.10.1 source code.zip 2025-10-20 21.8 MB
6 6 weekly downloads
Totals: 14 Items   119.6 MB 6

Overview

Core 2.10.1 is a patch release fixing a significant partial scheduling bug that existed since 2.9.0 but started being visible to users from 2.10.0.

We also take eliminate a set of scenarios where the scheduler experienced a slow-start due to waiting for connections from server replicas that were never be created due to configuration errors (i.e updating a ServerConfig after Servers with more than zero replicas referencing that config have been deployed as StatefulSets)

Bug details:

Starting in 2.10.0, after pods of an inference server hosting a model were restarted (irrespective of the reason), the model ended up scheduled only on approximately model.spec.minReplicas server replicas rather than the requested (and expected) model.spec.replicas. The variation in the actual model replicas being scheduled was dependent on the timing/sequencing of server replica connection to the scheduler after restart.

This regression appeared because an existing bug in the partial scheduling logic (there since 2.9.0) started manifesting itself consistently after fixing a data race bug (not directly related to partial scheduling) in 2.10. Before, the data race bug was difficult to trigger under most cluster operation scenarios, so was not experienced by users.

In 2.10.1, we fix the underlying bug so that partial scheduling works as expected.

Changelog

Dates are displayed in UTC. Generated by auto-changelog.

v2.10.1

20 October 2025

  • fix(operator): incorrect expected replicas notification #6890
  • fix(scheduler): not all models deployed to Servers when minReplicas on Model is set #6885
Source: README.md, updated 2025-10-20