Compare commits

..

1 Commits

Author SHA1 Message Date
Andrew Stoltz
40b5027ee3 Mirror printer state alert routing 2026-05-19 15:17:01 -05:00
6 changed files with 247 additions and 202 deletions

View File

@@ -103,7 +103,6 @@ curl -sk -X DELETE https://dns.iamworkin.lan/api/v1/servers/<serverId>/zones/iam
- **Public read-only hosts**: if a public host fronts a service that also exposes admin writes internally, add a Traefik route match like `Host(...) && (Method(GET) || Method(HEAD))` on the public edge instead of trusting the app to reject unsafe methods.
- **Public read-write allowlist hosts**: if a public host accepts a tightly bounded write surface (e.g. bootstrap-JWT POST), pin the allowlist as `(Method(GET) || Method(HEAD) || Method(POST) || Method(OPTIONS))`. PUT/PATCH/DELETE must still 404 at the route. Track A's `updatecenter.iamworkin.lan` / `updates.iamworkin.lan` are the canonical example. The lint test enforces this invariant.
- **Traefik VIP netpols**: when a `NetworkPolicy` allows `10.0.56.200`, also allow the post-DNAT backend ports (`8443` for TLS plus `8080` or `8000` for HTTP) or Calico will drop the rewritten flow.
- **RemoteDesktop isolation**: `apps/fc-desktop/network-policies.yaml` intentionally keeps desktop pod egress to named CoreDNS, `intranet-web:5300/TCP`, and noc1 step-ca `10.0.56.10:9000/9443` only. Guacamole display egress is owned separately by `apps/guacamole/guacamole.yaml` through `guacd-desktop-egress` on `5901/TCP`.
- **Auth-safe probes**: services behind API-key or global auth middleware should prefer `tcpSocket` probes unless `/health` is explicitly exempted before the middleware runs.
- **ArgoCD must use internal Gitea URL**: `http://gitea-clusterip.gitea.svc.cluster.local:3000/bluejay/bluejay-infra.git`, not the external HTTPS URL (step-ca cert isn't trusted by ArgoCD). The `ApplicationSet` and any hand-created `Application` must both use the internal URL.

View File

@@ -20,12 +20,9 @@
# 1) desktop-isolation — Browser Lab session pods.
#
# Locks down pods labeled `app.kubernetes.io/name=remote-desktop` (every
# session pod regardless of template). Allows guacd ingress for the display
# lane and remotedesktop-web's pre-handoff probing. Egress is deliberately
# narrow: named CoreDNS, direct Intranet web, and noc1 step-ca only. There is
# no broad Traefik/VIP or internet egress from desktop sessions. If a future
# Browser Lab path needs a public-style host, prefer an explicit Service rule
# or include the post-DNAT backend port per the Traefik VIP lint.
# session pod regardless of template). Allows guacd ingress for the VNC/RDP
# display lane and remotedesktop-web's pre-handoff probing. Egress: NFS to
# Synology, DNS, Traefik (cluster + LB VIP), Intranet (Browser Lab home).
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
@@ -68,22 +65,51 @@ spec:
- port: 5901
protocol: TCP
egress:
# CoreDNS only. The old to: [] DNS rule accidentally allowed any DNS
# listener in any namespace or routed network.
# NFS to Synology
- to:
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: kube-system
podSelector:
matchLabels:
k8s-app: kube-dns
- ipBlock:
cidr: 10.0.58.3/32
ports:
- port: 2049
protocol: TCP
- port: 2049
protocol: UDP
- port: 111
protocol: TCP
- port: 111
protocol: UDP
- to:
- ipBlock:
cidr: 10.0.58.3/32
ports:
- port: 445
protocol: TCP
- to: []
ports:
- port: 53
protocol: UDP
- port: 53
protocol: TCP
# Browser Lab home / internal docs target. Use the real service port
# directly rather than public Traefik host aliases.
- to:
- ipBlock:
cidr: 10.0.56.200/32
- ipBlock:
cidr: 10.43.33.87/32
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: traefik-system
podSelector:
matchLabels:
app.kubernetes.io/name: traefik
ports:
- port: 80
protocol: TCP
- port: 443
protocol: TCP
- port: 8000
protocol: TCP
- port: 8443
protocol: TCP
- to:
- namespaceSelector:
matchLabels:
@@ -94,17 +120,6 @@ spec:
ports:
- port: 5300
protocol: TCP
# noc1 step-ca ACME endpoint. The lane brief called out 9000/TCP; the live
# ACME directory currently answers on 9443/TCP, so both stay pinned to the
# same host rather than reopening Traefik or internet egress.
- to:
- ipBlock:
cidr: 10.0.56.10/32
ports:
- port: 9000
protocol: TCP
- port: 9443
protocol: TCP
---
# 2) fc-desktop-default-deny — namespace-wide catch-all.
#
@@ -315,11 +330,3 @@ spec:
protocol: UDP
- port: 53
protocol: TCP
- to:
- ipBlock:
cidr: 10.0.56.10/32
ports:
- port: 9000
protocol: TCP
- port: 9443
protocol: TCP

View File

@@ -254,68 +254,6 @@ spec:
targetPort: 4822
name: guacd
---
# Guacd display egress isolation.
#
# Guacamole web talks to guacd on TCP/4822. Guacd then opens the desktop
# display connection to the per-session pod. Keep that second hop at raw VNC
# 5901/TCP for the current RemoteDesktop Browser Lab/openSUSE images. Do not
# grant guacd broad fc-desktop namespace egress; desktop-to-desktop lateral
# paths remain blocked by apps/fc-desktop/network-policies.yaml.
apiVersion: networking.k8s.io/v1
kind: NetworkPolicy
metadata:
name: guacd-desktop-egress
namespace: guacamole
labels:
app.kubernetes.io/part-of: remotedesktop
app.kubernetes.io/component: display-isolation
spec:
podSelector:
matchLabels:
app: guacd
policyTypes:
- Ingress
- Egress
ingress:
- from:
- podSelector:
matchLabels:
app: guacamole
ports:
- port: 4822
protocol: TCP
egress:
- to:
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: kube-system
podSelector:
matchLabels:
k8s-app: kube-dns
ports:
- port: 53
protocol: UDP
- port: 53
protocol: TCP
# kubectl-proxy sidecar reaches the Kubernetes API; keep it explicit
# because this NetworkPolicy selects the whole guacd pod.
- to: []
ports:
- port: 443
protocol: TCP
- port: 6443
protocol: TCP
- to:
- namespaceSelector:
matchLabels:
kubernetes.io/metadata.name: fc-desktop
podSelector:
matchLabels:
app.kubernetes.io/name: remote-desktop
ports:
- port: 5901
protocol: TCP
---
# Guacamole Web Application
apiVersion: apps/v1
kind: Deployment

View File

@@ -656,14 +656,15 @@ data:
summary: "Print queue backlog on edge2 ({{ $value }} active jobs)"
description: "CUPS has {{ $value }} active jobs queued. Possible printer jam, USB disconnect, or paper out."
# Paper roll lifecycle alerts (XL Track I, 2026-04-26).
# Source-of-truth gauge: print_paper_remaining_percent (Print.Web OTEL,
# hydrated on startup from the active PaperRoll row).
# alert_channel=thermal_print routes through irc-notify -> Print.Web
# /api/print/alert so the printer announces its own paper-out warning
# on its remaining paper. Self-referential humor + operator nudge.
# Printer hardware and paper-roll lifecycle alerts.
# print_printer_online: 1 when the transport is reachable/selected.
# print_printer_state enum: 0 unknown, 1 online, 2 offline,
# 3 paper_depleted, 4 jam, 5 head_error, 6 cover_open.
# Offline/jam/cover alerts stay IRC-only. Paper depleted and head
# error may route to the thermal digest only when the printer is
# online enough to make that useful.
- alert: PrintPaperRollLow
expr: print_paper_remaining_percent{job="printweb-otel"} < 10 and print_paper_remaining_percent{job="printweb-otel"} > 5
expr: (print_paper_remaining_percent{job="printweb-otel"} < 10 and print_paper_remaining_percent{job="printweb-otel"} > 5) and print_printer_online{job="printweb-otel"} == 1
for: 5m
labels:
severity: warning
@@ -672,15 +673,59 @@ data:
summary: "Print roll low on edge2 ({{ $value | printf \"%.1f\" }}% remaining)"
description: "NuPrint 210 paper roll has {{ $value | printf \"%.1f\" }}% remaining. Operator should load a fresh roll soon. Run /api/paper/status for the precise mm + estimated jobs left."
- alert: PrinterOfflineWarning
expr: print_printer_state{job="printweb-otel"} == 2
for: 2m
labels:
severity: warning
service: print-web
alert_channel: irc
annotations:
summary: "Print.Web printer offline on edge2"
description: "Print.Web reports the NuPrint 210 transport is offline or unreachable. IRC-only by design: do not thermal-print an alert when the thermal printer itself is offline."
- alert: PrintPaperRollCritical
expr: print_paper_remaining_percent{job="printweb-otel"} <= 5
expr: print_printer_state{job="printweb-otel"} == 3 and print_printer_online{job="printweb-otel"} == 1
for: 2m
labels:
severity: critical
alert_channel: thermal_print
annotations:
summary: "Print roll critical on edge2 ({{ $value | printf \"%.1f\" }}% remaining)"
description: "NuPrint 210 paper roll at {{ $value | printf \"%.1f\" }}% — load a new roll NOW. The 50ft roll has a ~12% red-stripe zone; once paper passes that, the printer can run dry mid-job."
summary: "Print paper depleted on edge2"
description: "NuPrint 210 reports paper depleted while the printer is still online. Load a new roll, drain the hardware buffer if needed, then replay DeadLetter jobs from /print-log."
- alert: PrinterJamWarning
expr: print_printer_state{job="printweb-otel"} == 4
for: 2m
labels:
severity: warning
service: print-web
alert_channel: irc
annotations:
summary: "Print.Web printer jam on edge2"
description: "Print.Web reports a paper/cutter jam state. IRC-only: clear the jam, drain the hardware buffer if bytes were queued, then retry affected jobs."
- alert: PrinterHeadErrorCritical
expr: print_printer_state{job="printweb-otel"} == 5
for: 2m
labels:
severity: critical
service: print-web
alert_channel: thermal_print
annotations:
summary: "Print.Web printer head error on edge2"
description: "Print.Web reports a thermal head or unrecoverable printer error. Critical routing may enter the thermal digest per existing policy; IRC remains the primary triage stream."
- alert: PrinterCoverOpenWarning
expr: print_printer_state{job="printweb-otel"} == 6
for: 2m
labels:
severity: warning
service: print-web
alert_channel: irc
annotations:
summary: "Print.Web printer cover open on edge2"
description: "Print.Web reports the printer cover/lid is open. IRC-only: close the cover and verify /api/print/status before retrying jobs."
- alert: PrintJobDeadLetter
expr: increase(print_jobs_dead_letter_total[15m]) > 0
@@ -3635,6 +3680,146 @@ data:
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: threshold, expression: B, conditions: [{evaluator: {params: [600], type: gt}}], refId: C}
- orgId: 1
name: Print Services
folder: Print Alerts
interval: 1m
rules:
- uid: printer-offline-warning
title: PrinterOfflineWarning
condition: C
for: 2m
noDataState: OK
execErrState: OK
annotations:
summary: "Print.Web printer offline on edge2"
description: "Print.Web reports the NuPrint 210 transport is offline or unreachable. IRC-only by design: do not thermal-print an alert when the thermal printer itself is offline."
runbook: "1. Check edge2 power/network 2. Check USB/CUPS queue 3. Open https://print.iamworkin.lan/admin 4. Do not force thermal routing for offline alerts."
labels:
severity: warning
service: print-web
alert_channel: irc
data:
- refId: A
relativeTimeRange: {from: 120, to: 0}
datasourceUid: prometheus
model: {expr: 'print_printer_state{job="printweb-otel"} == 2', instant: true, refId: A}
- refId: B
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: reduce, expression: A, reducer: last, refId: B}
- refId: C
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: threshold, expression: B, conditions: [{evaluator: {params: [0], type: gt}}], refId: C}
- uid: print-paper-roll-critical
title: PrintPaperRollCritical
condition: C
for: 2m
noDataState: OK
execErrState: OK
annotations:
summary: "Print paper depleted on edge2"
description: "NuPrint 210 reports paper depleted while the printer is still online. Load a new roll, drain the hardware buffer if needed, then replay DeadLetter jobs from /print-log."
runbook: "1. Load a fresh roll 2. Drain the hardware buffer if paper-out happened mid-job 3. Open https://print.iamworkin.lan/print-log 4. Retry DeadLetter jobs after the state clears."
labels:
severity: critical
service: print-web
alert_channel: thermal_print
data:
- refId: A
relativeTimeRange: {from: 120, to: 0}
datasourceUid: prometheus
model: {expr: 'print_printer_state{job="printweb-otel"} == 3 and print_printer_online{job="printweb-otel"} == 1', instant: true, refId: A}
- refId: B
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: reduce, expression: A, reducer: last, refId: B}
- refId: C
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: threshold, expression: B, conditions: [{evaluator: {params: [0], type: gt}}], refId: C}
- uid: printer-jam-warning
title: PrinterJamWarning
condition: C
for: 2m
noDataState: OK
execErrState: OK
annotations:
summary: "Print.Web printer jam on edge2"
description: "Print.Web reports a paper/cutter jam state. IRC-only: clear the jam, drain the hardware buffer if bytes were queued, then retry affected jobs."
runbook: "1. Clear paper/cutter path 2. Drain hardware buffer if CUPS queued bytes 3. Verify /api/print/status 4. Retry affected jobs."
labels:
severity: warning
service: print-web
alert_channel: irc
data:
- refId: A
relativeTimeRange: {from: 120, to: 0}
datasourceUid: prometheus
model: {expr: 'print_printer_state{job="printweb-otel"} == 4', instant: true, refId: A}
- refId: B
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: reduce, expression: A, reducer: last, refId: B}
- refId: C
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: threshold, expression: B, conditions: [{evaluator: {params: [0], type: gt}}], refId: C}
- uid: printer-head-error-critical
title: PrinterHeadErrorCritical
condition: C
for: 2m
noDataState: OK
execErrState: OK
annotations:
summary: "Print.Web printer head error on edge2"
description: "Print.Web reports a thermal head or unrecoverable printer error. Critical routing may enter the thermal digest per existing policy; IRC remains the primary triage stream."
runbook: "1. Let the printer cool if overheated 2. Power-cycle only after checking queued jobs 3. Verify /api/print/status 4. Retry jobs after the state clears."
labels:
severity: critical
service: print-web
alert_channel: thermal_print
data:
- refId: A
relativeTimeRange: {from: 120, to: 0}
datasourceUid: prometheus
model: {expr: 'print_printer_state{job="printweb-otel"} == 5', instant: true, refId: A}
- refId: B
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: reduce, expression: A, reducer: last, refId: B}
- refId: C
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: threshold, expression: B, conditions: [{evaluator: {params: [0], type: gt}}], refId: C}
- uid: printer-cover-open-warning
title: PrinterCoverOpenWarning
condition: C
for: 2m
noDataState: OK
execErrState: OK
annotations:
summary: "Print.Web printer cover open on edge2"
description: "Print.Web reports the printer cover/lid is open. IRC-only: close the cover and verify /api/print/status before retrying jobs."
runbook: "1. Close the printer cover 2. Verify /api/print/status returns online 3. Retry affected jobs only after the state clears."
labels:
severity: warning
service: print-web
alert_channel: irc
data:
- refId: A
relativeTimeRange: {from: 120, to: 0}
datasourceUid: prometheus
model: {expr: 'print_printer_state{job="printweb-otel"} == 6', instant: true, refId: A}
- refId: B
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: reduce, expression: A, reducer: last, refId: B}
- refId: C
relativeTimeRange: {from: 120, to: 0}
datasourceUid: __expr__
model: {type: threshold, expression: B, conditions: [{evaluator: {params: [0], type: gt}}], refId: C}
- orgId: 1
name: CI Runners
folder: CI Alerts

View File

@@ -304,7 +304,7 @@ public sealed class FleetManifestLintTests
}
[Fact]
public void Monitoring_MustAlertWhenLinuxRunnerDeploymentIsUnavailable()
public void Monitoring_MustIncludeRequiredAlertRoutingGuards()
{
var monitoring = File.ReadAllText(Path.Combine(Inventory.BluejayRoot, "apps", "monitoring", "noc-monitoring.yaml"));
@@ -315,6 +315,15 @@ public sealed class FleetManifestLintTests
monitoring.Should().Contain("folder: CI Alerts");
monitoring.Should().Contain("uid: linux-runner-offline");
monitoring.Should().Contain("alert_channel: irc");
monitoring.Should().Contain("PrinterOfflineWarning");
monitoring.Should().Contain("expr: print_printer_state{job=\"printweb-otel\"} == 2");
monitoring.Should().Contain("IRC-only by design: do not thermal-print an alert when the thermal printer itself is offline.");
monitoring.Should().Contain("PrintPaperRollCritical");
monitoring.Should().Contain("expr: print_printer_state{job=\"printweb-otel\"} == 3 and print_printer_online{job=\"printweb-otel\"} == 1");
monitoring.Should().Contain("PrinterJamWarning");
monitoring.Should().Contain("PrinterHeadErrorCritical");
monitoring.Should().Contain("PrinterCoverOpenWarning");
}
[Fact]

View File

@@ -1,93 +0,0 @@
using FluentAssertions;
using Xunit;
namespace BluejayInfraLint.Tests;
[Trait("Category", "Unit")]
public sealed class RemoteDesktopNetworkPolicyTests
{
private static readonly ManifestInventory Inventory = ManifestInventory.Load();
[Fact]
public void LiveDesktopIsolation_AllowsOnlyCoreDnsIntranetAndStepCaEgress()
{
var policy = NetworkPolicy("fc-desktop", "desktop-isolation");
var ports = policy.EgressPorts().ToHashSet(StringComparer.Ordinal);
ports.Should().BeEquivalentTo("53", "5300", "9000", "9443");
policy.AllScalars().Should().Contain(new[]
{
"kube-system",
"kube-dns",
"intranet",
"intranet-web",
"10.0.56.10/32"
});
}
[Fact]
public void LiveDesktopIsolation_RemovesInternetNfsAndTraefikEgress()
{
var policy = NetworkPolicy("fc-desktop", "desktop-isolation");
var scalars = policy.AllScalars().ToList();
var ports = policy.EgressPorts().ToHashSet(StringComparer.Ordinal);
scalars.Should().NotContain(new[] { "10.0.58.3/32", "10.0.56.200/32", "10.43.33.87/32", "traefik-system" });
ports.Should().NotContain(new[] { "80", "443", "445", "111", "2049", "8000", "8080", "8443" });
policy.MappingSequence("spec", "egress")
.Should()
.NotContain(rule => EgressRuleHasEmptyTo(rule), "desktop sessions must not use to: [] internet-style egress");
}
[Fact]
public void LiveGuacdIsolation_AllowsRawVncToDesktopPodsOnly()
{
var policy = NetworkPolicy("guacamole", "guacd-desktop-egress");
var scalars = policy.AllScalars().ToList();
var ports = policy.EgressPorts().ToHashSet(StringComparer.Ordinal);
ports.Should().Contain("5901");
scalars.Should().Contain(new[] { "fc-desktop", "remote-desktop" });
ports.Should().NotContain(new[] { "3000", "3001", "3389", "80", "8080", "8443" });
}
[Fact]
public void LiveGuacdIsolation_KeepsGuacamoleWebIngressOnGuacdPort()
{
var policy = NetworkPolicy("guacamole", "guacd-desktop-egress");
policy.Scalar("spec", "podSelector", "matchLabels", "app").Should().Be("guacd");
policy.AllScalars().Should().Contain(new[] { "guacamole", "4822" });
}
[Fact]
public void HelperSmoke_FindsExpectedRemoteDesktopPolicies()
{
NetworkPolicy("fc-desktop", "desktop-isolation").Name.Should().Be("desktop-isolation");
NetworkPolicy("guacamole", "guacd-desktop-egress").Name.Should().Be("guacd-desktop-egress");
}
[Fact]
public void HelperSmoke_EgressPortExtractionKeepsDistinctPorts()
{
var ports = NetworkPolicy("fc-desktop", "desktop-isolation")
.EgressPorts()
.ToHashSet(StringComparer.Ordinal);
ports.Should().HaveCount(4);
ports.Should().Contain(new[] { "53", "5300", "9000", "9443" });
}
private static ManifestDocument NetworkPolicy(string ns, string name)
=> Inventory.Documents.Single(document =>
document.Kind == "NetworkPolicy"
&& string.Equals(document.Namespace, ns, StringComparison.Ordinal)
&& string.Equals(document.Name, name, StringComparison.Ordinal));
private static bool EgressRuleHasEmptyTo(YamlDotNet.RepresentationModel.YamlMappingNode rule)
=> rule.Children.Any(entry =>
entry.Key is YamlDotNet.RepresentationModel.YamlScalarNode key
&& string.Equals(key.Value, "to", StringComparison.Ordinal)
&& entry.Value is YamlDotNet.RepresentationModel.YamlSequenceNode sequence
&& sequence.Children.Count == 0);
}