- Replace second form field (extra_pairs) with target_version
- target_version sets version B ("to"); empty = highest available tag
- Pair every tag in [floor_version, target_version) with target_version
The create step was calling postgres.start / mongo.start immediately after
create. In Dokploy, .start only scales an already-deployed Swarm service;
on a freshly created resource the service doesn't exist yet, so the server
ran 'docker service scale ci-pg-db-xxx=1' against a missing service and
returned a 500 (curl -sf → exit 22 → step failure).
.deploy is the mutation that actually builds and creates the Swarm service.
Applications already used .deploy correctly; this aligns the databases.
v0.20.x used a completely different postgres/mongo/app API schema —
projectId fallback returned boolean false, not the expected resource object.
Since the goal is testing v0.29.4+, drop the old-API compatibility shim
and fail fast with a clear message if environmentId is missing.
Also: default floor_version changed from v0.20.0 to v0.29.4 to reduce
the number of matrix jobs and focus on the stable modern API surface.
In Dokploy <= ~v0.24, project.create returns a flat object with just
projectId at the top level; no environment concept exists yet, so
resources (postgres, mongo, apps) are created with projectId.
In Dokploy >= ~v0.25, project.create returns nested {project, environment}
objects and resources require environmentId.
Previous workflow always used .project.projectId and .environment.environmentId,
both of which evaluated to null on old versions, causing curl to exit with
code 22 (HTTP error) on every subsequent resource-create call.
Fix: extract PROJECT_ID with a // fallback, check ENV_ID nullness, and
build a SCOPE fragment (either 'environmentId' or 'projectId') used in
all resource-creation tRPC calls.
Previous run failed with 'This script must be run as root'. install.sh also
respects DOKPLOY_VERSION and ADVERTISE_ADDR env vars, so we can install
version A directly instead of installing latest and downgrading (which
would risk B's migrations running on A's expected schema).
When called with an empty array, `docker service ps` receives no
SERVICE argument and fails with "requires at least 1 argument".
Return early with [] when appNames is empty.
getSwarmNodes was changed to return [] on error, but the existing
SwarmMonitorCard checks `if (!nodes)` to detect failures. Since ![]
is false, the error state was silently skipped, breaking the Overview
tab for users without Docker Swarm initialized.
Reverted to return undefined on error (original behavior) so the
existing Overview tab error handling continues to work. The Containers
tab already handles nodes === undefined explicitly.
TL;DR: New "Containers" tab on the Swarm page showing which containers
run on which nodes, with live CPU/Memory/Block I/O/Network I/O metrics
refreshing every 5 seconds. Comprehensive edge case handling guides
users through prerequisites (swarm init, registry, service deployment).
---
## New Files
- `apps/dokploy/components/dashboard/swarm/containers/show-swarm-containers.tsx`
Main component: data fetching, error/empty states, summary cards,
and the node-grouped container layout.
- `apps/dokploy/components/dashboard/swarm/containers/node-section.tsx`
Collapsible per-node section with container table, role badge,
and down-node indicator.
- `apps/dokploy/components/dashboard/swarm/containers/container-row.tsx`
Table row for a single container: state badge with error tooltip,
formatted CPU/memory/IO metrics.
- `apps/dokploy/components/dashboard/swarm/containers/utils.ts`
Formatting helpers for docker stats values (CPU %, memory, I/O).
- `apps/dokploy/components/dashboard/swarm/containers/types.ts`
Shared TypeScript interfaces (ContainerStat, ContainerInfo, SwarmNode,
NodeGroup).
## Modified Files
- `apps/dokploy/pages/dashboard/swarm.tsx`
Added Tabs (Overview / Containers) wrapping existing SwarmMonitorCard
and the new ShowSwarmContainers component.
- `apps/dokploy/server/api/routers/swarm.ts`
Added `getContainerStats` tRPC endpoint calling `getAllContainerStats`,
following existing auth/validation patterns.
- `packages/server/src/services/docker.ts`
- Added `getAllContainerStats()` — runs `docker stats --no-stream` for
cluster-wide container metrics.
- Fixed `getSwarmNodes`, `getNodeApplications`, `getApplicationInfo` to
return `[]` instead of `undefined` on errors (prevents tRPC
serialization crashes) and added `console.error` logging.
- Added empty stdout guard (`if (!stdout.trim()) return []`) to prevent
`JSON.parse("")` crashes when no services exist.
## Features
- Container table per node: name, image, state, CPU %, memory usage,
block I/O, and network I/O
- Resource formatting: values rounded to 1 decimal (2.711MiB → 2.7 MiB),
CPU to 1 decimal (0.00% → 0.0%)
- Node role badges (Leader / Reachable / Worker) on each section header
- Error tooltips: hover the status badge to see Docker error details
- Down/drained node detection with red indicator dot and warning banner
- Multi-node metrics banner explaining docker stats manager-only limitation
- Unscheduled services footer for services scaled to 0 replicas
- Contextual empty/error states with actionable guidance, doc links to
Dokploy docs and Docker Swarm guide, and links to Cluster Settings
## Edge Cases Handled
1. Swarm not initialized (tRPC error or undefined data)
2. Docker command failures (stderr / non-zero exit)
3. Swarm active but no services deployed
4. Services exist but no running containers
5. Containers with Docker errors (shown in tooltip + error alert)
6. Nodes down or drained (cross-referenced from node list)
7. Multi-node setups (metrics only from manager node)
8. Services scaled to 0 replicas (separated from running containers)
9. Empty stdout from docker commands (no JSON.parse crash)
ClaudeandBlas