Project
Docker Engine is the core runtime that builds, runs, and manages containers, including networking, image pulls, and Swarm orchestration features.
Docker Engine 29.7.0 and 29.8.2 started zero Swarm overlay tasks on a test host lacking IPv6, where 29.6.2 ran all 20, according to a dev.to post. It is one single-node reproduction, so Swarm teams leaving 29.6.x on similar hosts have reason to rerun it before upgrading.
Reality
Docker 29.8.2 cut two healthy containers off their overlay network within 20 seconds when a swarm service kept failing on a taken port, a dev.to test found. The containers keep showing as running, and after enough failures only a daemon restart brings them back.
Reality
Docker's healthcheck defaults probe every 30 seconds with no start period, so a container ready at second 2 can wait almost half a minute to turn healthy. Setting start_period and start_interval yourself shortens rollouts, as long as a proxy or Compose acts on the health status.
Reality
Docker Compose v5.5.1 hit the registry on all three explicit pulls in a developer's test, though its v5.5.0 changelog says pull now honors refresh windows. For CI pipelines built around an explicit pull step, the upgrade bought no fewer registry calls in that test.
Reality
containerd 2.2's mount manager took 29.6 to 47.2 ms per Activate call in a dev.to benchmark, against 23.5 to 25.9 ms for truncate, mkfs, losetup and mount. Most of the gap is design overhead, and a panic plus an orphaned loop device from the same runs are the better reason for shim authors to wait.
Reality
Engine 29.8.0's --umask flag sets one mask for the main process, execs and healthchecks. A side-by-side test on a second daemon also shows it accepting a four-digit value and applying only the last three.
Reality
Compose has carried a restart flag on depends_on since 2023. A read-only pass over 38 running containers on eight servers found seven declared dependencies, all of them waiting on health and all of them recording restart as false.
Reality
On an ext4 host the overlay2 graphdriver refused --storage-opt size outright with exit code 125. The containerd snapshotter that Docker 29 makes the default accepts the same flag on the same filesystem and enforces nothing.
Reality
The word healthcheck names three different behaviours depending on which layer reads it, and the layer where it means nothing for routing is the one most production traffic goes through. Portability was the assumption.
Reality