Also, if I do something like this

curl -H "Content-Type: application/json" -iv -u admin:admin
> atlas-server.service.consul:21000/api/atlas/v2/search/basic?classification=PII2&includeClassificationAttributes=true


The attributes don't come back either.

On Tue, Dec 12, 2017 at 12:01 PM, Connie Chen <[email protected]> wrote:

> Hi, I am still trying to resolve this issue and it is somewhat difficult
> as I don't understand the gremlin query language. I am trying to just run
> the query that outputs on gremlin, and it is giving me this error
>
> > GET /api/atlas/discovery/search/gremlin?gremlinQuery=def r=(([]) as
>> Set);def f1={GremlinPipeline x->x.as('theTrait').in().as('theInstance')};
>> [0..<25].select(['theTrait', 'theInstance']).fill(r)};f1(g.
>> V().has('__typeName','PII2'));f1(g.V().has('__
>> superTypeNames','PII2'));r._().as('__tmp').transform({((Row)
>> it).getColumn('theTrait')}).as('theTrait').back('__tmp').
>> transform({((Row)it).getColumn('theInstance')}).as(
>> 'theInstance').select(['theTrait', 'theInstance'],{[it]},{[it]}).
>> as('a0').has('PII2.pii_type',T.'eq','example').back('a0')
>> [0..<25].toList() HTTP/1.1
>> > Authorization: Basic YWRtaW46YWRtaW4=
>> > User-Agent: curl/7.35.0
>> > Host: atlas-server.service.consul:21000
>> > Accept: */*
>> >
>> < HTTP/1.1 400 Unknown Version
>> < Content-Length: 0
>> < Connection: close
>> * Server Jetty(9.2.12.v20150709) is not blacklisted
>> < Server: Jetty(9.2.12.v20150709)
>> <
>> * Closing connection 0
>
>
> If it works in your env do you think you can add an example of this in the
> QuickStartV2? That entire quick start works on our setup. But, there is no
> example of a attribute def on a classification (and searching for it). I am
> not sure if something is wrong with the query it is generating (see above)
> or our setup. Thanks again for your help.
>
> Connie
>
> On Fri, Dec 8, 2017 at 4:03 PM, Connie Chen <[email protected]> wrote:
>
>> Hi Madhan, thanks for the response. It is still producing the same
>> result. Do you have any suggestions for where to debug this? It looks like
>> it is in the jersey layer, but where is that being controlled? thank you.
>>
>> Connie
>>
>> On Thu, Dec 7, 2017 at 7:58 PM, Madhan Neethiraj <[email protected]>
>> wrote:
>>
>>> Connie,
>>>
>>>
>>>
>>> The error was due to missing request header: -H 'Content-Type:
>>> application/json'. Can you try the following?
>>>
>>>
>>>
>>> curl -iv -H 'Content-Type: application/json' -u admin:admin -X POST
>>> atlas-server/api/atlas/v2/search/basic -d '{
>>>
>>>    "excludeDeletedEntities": true,
>>>
>>>    "classification"        : "PII",
>>>
>>>    "query"                 : "",
>>>
>>>    "limit"                 : 50,
>>>
>>>    "offset"                : 0,
>>>
>>>    "entityFilters"         : null,
>>>
>>>    "tagFilters"            : {
>>>
>>>      "attributeName" : "pii_type",
>>>
>>>      "operator"      : "eq",
>>>
>>>      "attributeValue": "example"
>>>
>>>    },
>>>
>>>    "attributes": [""]
>>>
>>> }'
>>>
>>>
>>>
>>>
>>>
>>> Hope this helps.
>>>
>>>
>>>
>>> Madhan
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> *From: *Connie Chen <[email protected]>
>>> *Reply-To: *"[email protected]" <[email protected]>
>>> *Date: *Wednesday, December 6, 2017 at 11:38 AM
>>>
>>> *To: *"[email protected]" <[email protected]>
>>> *Subject: *Re: 500 returns when I try to basic search
>>>
>>>
>>>
>>> I tried an example from the suggested queries applied to classification,
>>> like `PII PII.type="name"` and got this error
>>>
>>>
>>>
>>> 2017-12-06 19:36:35,743 ERROR - [pool-1-thread-7 -
>>> 4cdc2d8b-f710-4c59-9eb2-ef7560801a74:] ~ Error handling a request:
>>> e83408f35ae8fcd6 (ExceptionMapperUtil:32)
>>> org.apache.atlas.exception.AtlasBaseException: Discovery query failed
>>> L:{def r=(([]) as Set);def f1={GremlinPipeline x->x.as
>>> ('theTrait').in().as('theInstance') [0..<25].select(['theTrait',
>>> 'theInstance']).fill(r)};f1(g.V().has('__typeName','PII'));f
>>> 1(g.V().has('__superTypeNames','PII'));r._().as('__tmp').tra
>>> nsform({((Row)it).getColumn('theTrait')}).as('theTrait').
>>> back('__tmp').transform({((Row)it).getColumn('theInstance
>>> ')}).as('theInstance').select(['theTrait',
>>> 'theInstance'],{[it]},{[it]}).as('a0').has('PII.type').back('a0')
>>> [0..<25].toList()}
>>> at org.apache.atlas.discovery.EntityDiscoveryService.searchUsin
>>> gDslQuery(EntityDiscoveryService.java:143)
>>> at org.apache.atlas.web.rest.DiscoveryREST.searchUsingDSL(Disco
>>> veryREST.java:94)
>>>
>>>
>>>
>>> On Wed, Dec 6, 2017 at 11:30 AM, Connie Chen <[email protected]>
>>> wrote:
>>>
>>> Hi Madhan, thanks for looking. I changed the config properties but
>>> getting the same error. I don't see the "Given type" error in the startup
>>> anymore though:
>>>
>>>
>>>
>>> https://gist.github.com/connie-stripe/597e087433869036e63e5d
>>> 8363d61bd3#file-taxonomy-enable-false
>>>
>>>
>>>
>>> I also tried an advanced search in the UI I think that was causing those
>>> `Discovery query` errors. Let me know what you think, thanks
>>>
>>>
>>>
>>> Connie
>>>
>>>
>>>
>>> On Wed, Dec 6, 2017 at 11:00 AM, Madhan Neethiraj <[email protected]>
>>> wrote:
>>>
>>> Connie,
>>>
>>>
>>>
>>> The query text is correct; it works in my env. The log file shows
>>> failures in bunch of other calls as well:
>>>
>>>   - com.sun.jersey.api.NotFoundException: null for uri:
>>> http://atlas-server.service.consul:21000/api/atlas/discovery
>>> /search/basic
>>>
>>>   - com.sun.jersey.api.NotFoundException: null for uri:
>>> http://atlas-server.service.consul:21000/api/atlas/v2/types/
>>> typedefs/classification
>>>
>>>   - com.sun.jersey.api.NotFoundException: null for uri:
>>> http://atlas-server.service.consul:21000/api/atlas/v2/discov
>>> ery/search/attribute?attrName=dataType
>>>
>>>   - org.apache.atlas.exception.AtlasBaseException: Discovery query
>>> failed L:{def r=(([]) as Set);def f1={GremlinPipeline x->x.as
>>> ('theTrait').in().as('theInstance') [0..<25].select(['theTrait',
>>> 'theInstance']).fill(r)};f1(g.V().has('__typeName','PII'));f
>>> 1(g.V().has('__superTypeNames','PII'));r._().as('__tmp').tra
>>> nsform({((Row)it).getColumn('theTrait')}).as('theTrait').
>>> back('__tmp').transform({((Row)it).getColumn('theInstance
>>> ')}).as('theInstance').select(['theTrait',
>>> 'theInstance'],{[it]},{[it]}).as('a0').has('PII.dataType',T.'eq','example').back('a0')
>>> [0..<25].toList()}
>>>
>>>   - org.apache.atlas.exception.AtlasBaseException: Discovery query
>>> failed PII OR Metric
>>>
>>>   - org.apache.atlas.exception.AtlasBaseException: Discovery query
>>> failed Column WHERE Column isa PII
>>>
>>>
>>>
>>> Log file shows a restart of Atlas running into the following error
>>> during initialization: “Given type TaxonomyTerm already exists”. If
>>> taxonomy feature (which is in tech-preview) is enabled, can you try after
>>> disabling it? Please set the following configuration in
>>> atlas-application.properties:
>>>
>>>     atlas.feature.taxonomy.enable=false
>>>
>>>
>>>
>>> Madhan
>>>
>>>
>>>
>>> *From: *Nixon Rodrigues <[email protected]>
>>> *Reply-To: *"[email protected]" <[email protected]>
>>> *Date: *Wednesday, December 6, 2017 at 1:04 AM
>>> *To: *"[email protected]" <[email protected]>
>>> *Subject: *Re: 500 returns when I try to basic search
>>>
>>>
>>>
>>> Connie,
>>>
>>>
>>>
>>> Can you try this json as the body, I think to search on attribute needs
>>> to be added to in criterion attribute.
>>>
>>>
>>>
>>>  http://atlasUrl:21000/api/atlas/v2/search/basic method - post
>>>
>>>
>>>
>>> {
>>>
>>>     "excludeDeletedEntities": true,
>>>
>>>     "entityFilters": null,
>>>
>>>     "tagFilters": {
>>>
>>>       "criterion": [
>>>
>>>         {
>>>
>>>           "attributeName": "type",
>>>
>>>           "operator": "eq",
>>>
>>>           "attributeValue": "example"
>>>
>>>         }
>>>
>>>       ]
>>>
>>>     },
>>>
>>>     "attributes": [],
>>>
>>>     "limit": 25,
>>>
>>>     "offset": 0,
>>>
>>>     "classification": "PII"
>>>
>>>   }
>>>
>>>
>>>
>>>
>>>
>>> On Wed, Dec 6, 2017 at 11:50 AM, Connie Chen <[email protected]>
>>> wrote:
>>>
>>> Hi Madhan, thanks for the fast response.
>>>
>>>
>>>
>>> I added the snippet of the application log that shows the stack trace
>>> when I make the search request. https://gist.github.c
>>> om/connie-stripe/597e087433869036e63e5d8363d61bd3#gistcomment-2277245
>>>
>>>
>>>
>>> I also added as much of the application log as I could in another file
>>> https://gist.github.com/connie-stripe/597e087433869036e
>>> 63e5d8363d61bd3#file-application-log
>>>
>>>
>>>
>>> Let me know if that helps, thanks!
>>>
>>> Connie
>>>
>>>
>>>
>>> On Tue, Dec 5, 2017 at 6:41 PM, Madhan Neethiraj <[email protected]>
>>> wrote:
>>>
>>> Connie,
>>>
>>>
>>>
>>> Thanks for the details provided on this issue. Classification definition
>>> and basic-search REST API call details look good. To troubleshoot further,
>>> can you add complete application.log file?
>>>
>>>
>>>
>>> Madhan
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>> *From: *Connie Chen <[email protected]>
>>> *Reply-To: *"[email protected]" <[email protected]>
>>> *Date: *Tuesday, December 5, 2017 at 5:17 PM
>>> *To: *"[email protected]" <[email protected]>
>>> *Subject: *500 returns when I try to basic search
>>>
>>>
>>>
>>> Hi, I am new to using Atlas and I am testing out the API to create my
>>> data models. I've created a classification with a string attribute
>>> "dataType" where "isIndexable" is set to true. However, I cannot figure out
>>> how to query for a specific value via the API without getting a 500.
>>>
>>>
>>>
>>> Here is my configuration:
>>>
>>> https://gist.github.com/connie-stripe/597e087433869036e63e5d8363d61bd3
>>>
>>>
>>>
>>> This is the field I am indexing on: https://gist.github.com/co
>>> nnie-stripe/597e087433869036e63e5d8363d61bd3#file-gistfile1-txt-L22
>>>
>>>
>>>
>>> Here is the curl command I am running to try and retrieve a value where
>>> that attribute field is a specific value: https://gist.github.com
>>> /connie-stripe/597e087433869036e63e5d8363d61bd3#gistcomment-2277045
>>>
>>>
>>>
>>> Then I am getting these logs back: https://gist.github.com/
>>> connie-stripe/597e087433869036e63e5d8363d61bd3#gistcomment-2277047
>>>
>>>
>>>
>>> Am I doing something wrong? Is there an easier way to debug these sorts
>>> of errors? The stack trace hasn't been very helpful. Thanks!
>>>
>>>
>>>
>>> Connie
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>
>>
>

Reply via email to