Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Rob Oxspring
    @roxspring
    OK - confirmed that RulesManager was broken
    It was loading from the system ClassLoader but once Spring's BOOT-INF magic is engaged then we're actually running inside a org.springframework.boot.loader.LaunchedURLClassLoader
    Vadim Shaigorodskiy
    @vadeg
    @roxspring wrong Classloader?
    Rob Oxspring
    @roxspring
    yup
    have got a fix lined up - seems to work, just running a gradlew check before opening a MR
    Rob Oxspring
    @roxspring
    Hopefully that'll get through CI and fix the problem - sorry to have caused the headache in the first place :)
    Vadim Shaigorodskiy
    @vadeg
    Great job @roxspring! Thank you!
    Gregor Zeitlinger
    @zeitlinger
    dito :)
    Tronje Krop
    @tkrop
    thanks.
    Tronje Krop
    @tkrop
    @vadeg @zeitlinger how about merging zalando/zally#1117
    Gregor Zeitlinger
    @zeitlinger
    done
    Tronje Krop
    @tkrop
    @roxspring we have another problem since the update:
    java.lang.RuntimeException: check invocation failed: id=101 title=Provide API Specification using OpenAPI checkName=validateSchema reason=java.lang.IllegalStateException: node.at("/instance/pointer").textValue() must not be null
        at de.zalando.zally.core.RulesValidator.invoke(RulesValidator.kt:63) ~[zally-core.jar!/:na]
        at de.zalando.zally.core.RulesValidator.validate(RulesValidator.kt:40) ~[zally-core.jar!/:na]
        at de.zalando.zally.core.CompositeRulesValidator.validate(CompositeRulesValidator.kt:10) ~[zally-core.jar!/:na]
        at de.zalando.zally.apireview.ApiViolationsController.validate(ApiViolationsController.kt:47) ~[classes!/:na]
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[na:1.8.0_232]
    Can you see something similar on your server?
    Origin is:
    Caused by: java.lang.reflect.InvocationTargetException: null
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[na:1.8.0_232]
        at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) ~[na:1.8.0_232]
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) ~[na:1.8.0_232]
        at java.lang.reflect.Method.invoke(Method.java:498) ~[na:1.8.0_232]
        at de.zalando.zally.core.RulesValidator.invoke(RulesValidator.kt:61) ~[zally-core.jar!/:na]
        ... 99 common frames omitted
    Caused by: java.lang.IllegalStateException: node.at("/instance/pointer").textValue() must not be null
        at de.zalando.zally.core.JsonSchemaValidator.toValidationMessage(JsonSchemaValidator.kt:43) ~[zally-core.jar!/:na]
        at de.zalando.zally.core.JsonSchemaValidator.validate(JsonSchemaValidator.kt:36) ~[zally-core.jar!/:na]
        at de.zalando.zally.ruleset.zalando.UseOpenApiRule.validateSchema(UseOpenApiRule.kt:56) ~[zally-ruleset-zalando.jar!/:na]
        ... 104 common frames omitted
    Rob Oxspring
    @roxspring
    Eek. Gotta get to a meeting now, but will take a look this afternoon. Maybe we need to revert and remake the changes more carefully???
    Tronje Krop
    @tkrop
    Testing through on our apis revealed that it is a bug that appears the moment the document started with openapi: ...
    Rob Oxspring
    @roxspring
    Are you seeing this with any openapi spec?
    Tronje Krop
    @tkrop
    I found none, that worked.
    correction I found none openapi 3.0 ... api that worked.
    Rob Oxspring
    @roxspring
    yeah - that's what I was driving at :)
    Tronje Krop
    @tkrop
    2.0 works.
    Rob Oxspring
    @roxspring
    right. that'll contribute to why we're not seeing the problem here. migrating from 2 to 3 is still on our todo list :(
    Tronje Krop
    @tkrop
    damn. I assume it has something to do with the parser configuration, but why are our tests not failing already?
    Rob Oxspring
    @roxspring
    huh. I just tried UseOpenApiRuleTest in IDEA and it failed with this stacktrace.
    ... but I'm sure ./gradle check was passing the other day - and it must have done in CI... does that mean that the subproject tests aren't being run??
    Tronje Krop
    @tkrop
    I agree. It was not obvious that it something was not running. I even thought it was working alright on Friday evening.
    I now see tons of exceptions when compiling/testing.
    Rob Oxspring
    @roxspring
    yeah - gradle shows 996 tests completed, 851 failed, 1 skipped for me !? :(
    Tronje Krop
    @tkrop
    Do we have something that makes the server reload libs or schemas on the fly?
    Rob Oxspring
    @roxspring
    not that I'm aware of
    Tronje Krop
    @tkrop
    same for me.
    we have the travis log, that showed things were running successful.
    Rob Oxspring
    @roxspring
    I think that's whats happening though
    debugging my failing UseOpenApiRuleTest I can see the underlying processing message before the null pointer
    message -> {TextNode@4171} ""content at URI \"http://openapis.org/v3/schema.json#\" is not valid JSON""
    https://www.openapis.org/v3/schema.json does indeed resolve to an html 404 page
    ( i should add that idea also tells me parsingMessage -> {TextNode@4175} ""Unexpected character ('<' (code 60)): expected a valid value (JSON String, Number, Array, Object or token 'null', 'true' or 'false')"" which is why an html page seems relevant)
    Tronje Krop
    @tkrop
    oh ja. I'm currently running the command from the .travis
    it shows the failures but does not fail :)
    Rob Oxspring
    @roxspring
    wtf?
    Tronje Krop
    @tkrop
    what a mess.
    oh, no wrong. it fails.
    lets see in the logs, what Friday happend.
    seems to have worked on Friday.
    Rob Oxspring
    @roxspring
    have we just been unlucky and openapis.org have moved that schema since Friday??
    Tronje Krop
    @tkrop
    that may be the reason.
    Rob Oxspring
    @roxspring
    Feel I have to point out that I did bump json-schema-core to a pre-released version - which could be related to all this https://github.com/zalando/zally/blob/master/server/build.gradle.kts#L67
    Tronje Krop
    @tkrop
    it may even just be temporary down.
    Rob Oxspring
    @roxspring
    true