All pull requests

Pull requests

• fix(snode): a host's huge pages are given back with its last storage node

  target/main
  #1654·wmousa opened Nov 14, 2026Contributor·

  ·

• feat(snode): the CPU-topology Job configures the kubelet of an RKE2 or k3s node

  #1653·wmousa opened Nov 13, 2026Contributor·

  ·

• feat(snode): the CPU-topology Job configures the kubelet of an EKS node

  target/main
  #1648·wmousa opened Nov 13, 2026Contributor·

  ·

• Implement backup deletion

  target/main
  #1645·mxsrc opened Nov 13, 2026Collaborator·3 tasks done·

  ·

• Replication: never grow the chain (merge-behind hold) + collapse migrated chains + spare cleanup

  #1638·schmidt-scaled opened Nov 11, 2026Contributor·

  ·

• Replication: sweep orphaned landing copies the lineage walk misses (cluster filled to 95% on back-and-forth migration)

  target/main
  #1635·schmidt-scaled opened Nov 10, 2026Contributor·

  ·

• Size the SPDK bdev_io pool from the core count (65535 -> 24576 on small nodes)

  target/main
  #1633·schmidt-scaled opened Nov 10, 2026Contributor·

  ·

• feat(floors): erasure-coding resource floors, raid_devices, relaxed migration at n+k

  target/main
  #1626·schmidt-scaled opened Nov 9, 2026Contributor·

  ·
  7

• feat(lblk): take RAID arrays and multipath maps as storage units

  target/main
  #1625·geoffrey1330 opened Nov 9, 2026Collaborator·

  ·

• feat(replication): replication steps through the peer client (control-plane split, step 2)

  #1623·EbiRider opened Nov 9, 2026Collaborator·2/2·8 tasks done·

  ·
  4

• feat(control-plane): stretched control plane over one FDB, two sites, operator-moved leadership

  target/main
  #1622·schmidt-scaled opened Nov 9, 2026Contributor·

  ·
  7

• fix(cluster create): configure FDB before waiting for the app stack (SFAM-2847)

  target/main
  #1619·boddumanohar opened Nov 6, 2026Member·

  ·

• feat(replication): peer control-plane client and internal /_peer API

  target/main
  #1615·EbiRider opened Nov 6, 2026Collaborator·1/2·

  ·
  4

• Prune docker

  target/main
  #1604·mxsrc opened Nov 4, 2026Collaborator·

  ·
  2

• Container images: pull through a cluster-wide registry prefix

  target/main
  #1596·Hamdy-khader opened Nov 3, 2026Collaborator·3 of 5 tasks·

  ·
  1

• Consistency groups: subsystem moves for a live late join

  target/main
  #1587·geoffrey1330 opened Nov 2, 2026Collaborator·

  ·
  21

• fix(lblk): uniform-only device growth, smallest factor across all LUNs

  target/main
  #1572·schmidt-scaled opened Oct 30, 2026Contributor·3 of 5 tasks·

  ·

• test(two-node-sim): F6 restart precheck on a fenced peer in JCERR

  target/main
  #1571·schmidt-scaled opened Oct 30, 2026Contributor·

  ·

• fix(restart): retry jc_compression status checks on all clusters

  target/main
  #1559·BombermaG opened Oct 29, 2026Collaborator·

  ·
  13

• fix(restart): replace thread with task-based execution

  target/main
  #1492·mxsrc opened Oct 16, 2026Collaborator·

  ·

• feat(snode): Talos huge pages and kubelet CPU topology, pre-provisioned by default or managed through the Talos API

  target/main
  #1491·wmousa opened Oct 16, 2026Contributor·

  ·
  10

• refactor(two-node): translate the Go event collector to a Python service

  target/main
  #1471·mxsrc opened Oct 13, 2026Collaborator·

  ·

• fix: batch-migrated snapshot's primary DB record never moved when a clone sibling migrated with it

  target/main
  #1456·EbiRider opened Oct 9, 2026Collaborator·

  ·
  1

• Ask the removal's admission without starting the removal

  target/main
  #1429·noctarius opened Oct 5, 2026Collaborator·

  ·
  1

• A refused node DELETE says why

  target/main
  #1428·noctarius opened Oct 5, 2026Collaborator·

  ·
  2