Try to fetch multiple time configuration

[thibaultmeyer]

It appears that when the application is started, the plugin continues to try to communicate with Vault (even though all configurations have already been retrieved and mapped to the Quarkus interface (@ConfigMapping).

App start with success (datasoure = OK, flyway = OK, ...). I get this error when :

• First incoming and outgoing HTTP request
• occasionally via opentelemetry

Error occurred while destroying instance of SYNTHETIC bean [
    types=[io.opentelemetry.api.OpenTelemetry, java.lang.Object],
    qualifiers=[@Any, @Default], target=n/a]: VaultClientException{operationName='VAULT [AUTH (approle)] Login',
    requestPath='https://xxxxxxx:8200/v1/auth/approle/login',
    status=400,
    body='{"errors":["invalid role or secret ID"]}'} [Error Occurred After Shutdown]

VaultClientException{operationName='VAULT [AUTH (approle)] Login', requestPath='https://xxxxxxxxx:8200/v1/auth/approle/login', status=400, body='{"errors":["invalid role or secret ID"]}'}
	at io.quarkus.vault.client.http.VaultHttpClient.throwVaultException(VaultHttpClient.java:80)
	at io.quarkus.vault.client.http.VaultHttpClient.lambda$buildResponse$0(VaultHttpClient.java:33)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture.uniApplyNow(CompletableFuture.java:684)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture.uniApplyStage(CompletableFuture.java:662)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture.thenApply(CompletableFuture.java:2200)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture$MinimalStage.thenApply(CompletableFuture.java:2948)
	at io.quarkus.vault.client.http.VaultHttpClient.buildResponse(VaultHttpClient.java:29)
	at io.quarkus.vault.client.http.jdk.JDKVaultHttpClient.lambda$execute$1(JDKVaultHttpClient.java:32)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture$UniCompose.tryFire(CompletableFuture.java:1150)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture.postComplete(CompletableFuture.java:510)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture.postFire(CompletableFuture.java:614)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture$UniWhenComplete.tryFire(CompletableFuture.java:844)
	at java.base@21.0.6/java.util.concurrent.CompletableFuture$Completion.exec(CompletableFuture.java:483)
	at java.base@21.0.6/java.util.concurrent.ForkJoinTask.doExec(ForkJoinTask.java:387)
	at java.base@21.0.6/java.util.concurrent.ForkJoinPool$WorkQueue.topLevelExec(ForkJoinPool.java:1312)
	at java.base@21.0.6/java.util.concurrent.ForkJoinPool.scan(ForkJoinPool.java:1843)
	at java.base@21.0.6/java.util.concurrent.ForkJoinPool.runWorker(ForkJoinPool.java:1808)
	at java.base@21.0.6/java.util.concurrent.ForkJoinWorkerThread.run(ForkJoinWorkerThread.java:188)
	at org.graalvm.nativeimage.builder/com.oracle.svm.core.thread.PlatformThreads.threadStartRoutine(PlatformThreads.java:896)
	at org.graalvm.nativeimage.builder/com.oracle.svm.core.thread.PlatformThreads.threadStartRoutine(PlatformThreads.java:872)

[vsevel]

at least one of your errors is in the context of a shutdown: Error Occurred After Shutdown
this is not strictly a vault issue. it is larger and complex problem that has to do with ordering components shutdown inside the quarkus app.
for that particular case, it is not clear to me what causes what. why are we logging in with vault? why is the role or secret id invalid?
what we see is that we are destroying the open telemetry bean.
apart from shutdown, do you still see an issue?
note that the extension has some background activity to refresh secrets even after the app has started. by default this is being done every 10 minutes. you can change the frequency using secretConfigCachePeriod
please post a reproducer if you still believe there is an issue with the vault extension.