The chart told operators that adding a duplicate env name to an installed
release makes the next upgrade fail in the API server, quoting "The order in
patch list ... doesn't match $setElementOrder list". Measured against helm
3.19.0 and Kubernetes 1.35, it does not fail: the upgrade succeeds, helm prints
"Release has been upgraded", and Kubernetes merges the env list by name, so the
duplicate pair collapses to the chart's own entry and the operator's override is
discarded without a word. At install time the same values behave differently -
both entries reach the pod, the later one wins, and the API server says so with
"hides previous definition of ..., which may be dropped when using apply".
The advice that followed from the old claim survives - such a switch only takes
at install time - but the stated reason was wrong in the dangerous direction. A
reader who expects a loud failure treats a green upgrade as proof the override
took effect, and green is exactly what a silently dropped override looks like.
The text now names the real failure mode and tells the reader to read the value
back off the Deployment. README carries the same correction, because every chart
in this repo offers the same additive env block.
Also records what the earlier measurement left open. With the database scaled
back up the pod returns to Ready on its own after about five seconds, restart
count still 0, so leaving the liveness group untouched costs no manual recovery.
And the limit of that signal belongs next to it: the db indicator validates a
connection, not the schema, so a database that came back empty still put the pod
back to Ready with /actuator/health at 200.
MANAGEMENT_ENDPOINT_HEALTH_GROUP_READINESS_INCLUDE=readinessState,db puts the datasource into Spring Boot's readiness group, which shipped with readinessState alone: a total database outage left the pod 1/1 Ready while every request failed. Naming the members replaces the default, so readinessState is repeated. The liveness group stays untouched on purpose, because an outage has to keep traffic away rather than add a crash loop to it.
MANAGEMENT_HEALTH_MAIL_ENABLED=false removes the mail indicator, which pinned the aggregate /actuator/health at 503 for the whole life of a healthy release, because the chart's mail wiring points at localhost:1025 where nothing listens. values.yaml and NOTES.txt document what that costs and when it has to be switched back on.
Teil von #157 und Teil von #43.
#157: multiplyQuantity turned a bare byte count into exponential notation.
Values pass through fromYaml, which parses every number as float64, and Go's
"%v" switches float64 to exponential notation from about 1e6 upwards. Both
places that stringify a number now go through the new formatNumber helper -
the input of multiplyQuantity, which aborted the render with an unsupported
suffix "e+06", and the output of a non-integer product, which did not abort
at all but rendered an invalid quantity that only the API server rejects.
Measured on this chart against ci/urlaubsverwaltung/test-values.yaml:
requests.memory=1048576 before: abort, suffix "e+06"
after: request 1048576, limit "3145728"
requests.ephemeralStorage=2097152 before: abort, suffix "e+06"
after: limit "6291456"
memory=999999, limitFactor=1.5 before: limit "1.4999985e+06", no abort
after: limit "1499998.5"
requests.memory=65536 (regression) 196608 before and after
The correction is the canonical one, word for word as merged into
template-chart, voucher-vault and zipline; only the helper prefix differs.
urlaubsverwaltung has a single resources path and a single render site
(the Deployment), and both were verified.
#43: adds a NOTES.txt. Its content was verified against a running release in
a throwaway k3d cluster rather than taken from upstream documentation - the
identity provider being resolved at boot (a bad issuer-uri crashlooops the
pod instead of breaking the login), the buildpack sizing the JVM heap from
the computed memory limit, /actuator/health answering 503 for the lifetime of
the release because the chart points the mail sender at localhost:1025, the
probe groups not reacting to a database outage, and a name collision in
urlaubsverwaltung.env making the next helm upgrade fail on the patch order.
README.md is deliberately untouched: the line listing the charts that ship a
NOTES.txt collides with every parallel NOTES PR and is pulled up once at the
end of the programme.
Minor bump: a NOTES.txt renders no Kubernetes object and the quantity fix
changes no default. appVersion unchanged.
jklein