Hi Lewis,
Thanks for your updates.
Here is the current status of gora-cassandra test.
1) GORA-151, 152, and 153
I have committed their patches.
2) GORA-154
We need to implement delete() and deleteByQuery() method.
3) NullPointerException on CassandraClient.getFamilyMap() method twice.
testGetWebPageDefaultFields(org.apache.gora.cassandra.store.TestCassandraStore)
Time elapsed: 0.525 sec <<< ERROR!
java.lang.NullPointerException
at
org.apache.gora.cassandra.store.CassandraClient.getFamilyMap(CassandraClient.java:342)
at
org.apache.gora.cassandra.store.CassandraStore.execute(CassandraStore.java:125)
at
org.apache.gora.cassandra.store.CassandraStore.get(CassandraStore.java:246)
at
org.apache.gora.store.DataStoreTestUtil.testGetWebPage(DataStoreTestUtil.java:371)
at
org.apache.gora.store.DataStoreTestUtil.testGetWebPageDefaultFields(DataStoreTestUtil.java:382)
at
org.apache.gora.store.DataStoreTestBase.testGetWebPageDefaultFields(DataStoreTestBase.java:295)
4) StackOverflowException .. 7 times
testQuery(org.apache.gora.cassandra.store.TestCassandraStore) Time
elapsed: 1.492 sec <<< ERROR!
java.lang.StackOverflowError
at
org.apache.gora.persistency.ListGenericArray.compareTo(ListGenericArray.java:32)
at org.apache.avro.generic.GenericData.compare(GenericData.java:487)
at
org.apache.gora.persistency.ListGenericArray.compareTo(ListGenericArray.java:102)
at
org.apache.gora.persistency.ListGenericArray.compareTo(ListGenericArray.java:32)
...
Thanks for the info about StackOverflowException on gora-hbase.
So, the remaining failures or errors are 10.
Tests run: 28, Failures: 1, Errors: 9, Skipped: 0, Time elapsed: 16.924
sec <<< FAILURE!
-Kaz
On 7/19/12 1:53 PM, Lewis John Mcgibbney wrote:
> Kaz,
>
> Having done a 'bit' of digging here it looks like gora-hbase also
> suffers from problems with some tests which are subsequently skipped
> at runtime (although not officially as maven doesn't pick this up).
>
> If you look at testDeleteByQuery for example as this one gave us the
> StackOverFlow exception within gora-cassandra you'll see
>
> @Override
> public void testDeleteByQuery() throws IOException {
> //We need to skip this test since gora considers endRow
> inclusive, while its exclusinve for HBase.
> //TODO: We should raise an issue for HBase to allow us to
> specify if the endRow will be inclussive or exclusive.
> }
>
> Therefore I think it essential that we find out exactly how many of
> these tests Gora makes assumptions for which do not agree semantically
> with the Cassandra specific data model.
>
> I'll keep digging.
>
> Lewis
>
> On Thu, Jul 19, 2012 at 7:39 PM, Lewis John Mcgibbney
> <[email protected]> wrote:
>> This is brilliant news and my priority is to get my local workspace
>> looking like both yours and Jenkins workspace.
>>
>> Thanks for quick response Kazuomi. I'll also not have time until much
>> later on to get deep into this but we're making good progress
>> none-the-less.
>>
>> Lewis
>>
>> On Thu, Jul 19, 2012 at 7:30 PM, Kazuomi Kashii <[email protected]> wrote:
>>> I saw
>>> https://builds.apache.org/view/G-L/view/Gora/job/gora-trunk/ws/trunk./gora-cassandra/target/surefire-reports/org.apache.gora.cassandra.store.TestCassandraStore.txt
>>> and those exceptions are exactly what I see on my test envrionment.
>>>
>>> Last night, I have fixed 3 bugs, but not reported yet.
>>> I fixed :
>>> a)AssertionFailedError on schemaExist()
>>> b)ComparisonFailure on gora-core test bug
>>> c) AssertionFailedError on gora-cassanda code for Map
>>>
>>> The remainings are StackOverflow and NullPoineterException.
>>>
>>> I will update the status and continue working on remaining stuffs this
>>> evening.
>>>
>>> Regards,
>>> -Kaz
>>>
>>>
>>>
>>> On 7/19/12 11:15 AM, Lewis John Mcgibbney wrote:
>>>> Hi Kaz,
>>>>
>>>> On Wed, Jul 18, 2012 at 11:08 PM, Kazuomi Kashii <[email protected]>
>>>> wrote:
>>>>
>>>>> I have not seen such an exception on my environments, both on Linux
>>>>> (CentOS)
>>>>> and Mac OSX.
>>>> Neither have I and I am working on Linux (Ubuntu 11.10). However now I
>>>> am getting a whole host of exceptions when running the Cassandra tests
>>>> locally. In all honesty I'm getting vastly different results from you
>>>> and the CI build so I just assume that this is my local setUp which is
>>>> affecting my results.
>>>>
>>>>> Is it in gora-cassandra test console output, or in
>>>>>
>>>>> target/surefire-reports/org.apache.gora.cassandra.store.TestCassandraStore.txt
>>>> This is exclusively from the surefire-reports on TestCassandraStore.
>>>>
>>>>> What exactly is that exception ?
>>>> The exception that we are seeing on the Jenkins CI build regarding the
>>>> snappy compression (which is also linked to CASSANDRA-4400) seems to
>>>> have vanished today. Today the embedded Cassandra was able to initiate
>>>> and some tests ran, out of 28 which did run there was however quite a
>>>> few failures [0]. If you look at the link I gave you, you will see
>>>> that most of them are actually the ones you highlighted earlier so at
>>>> least we are reading from the same page now.
>>>>
>>>> [0]
>>>> https://builds.apache.org/view/G-L/view/Gora/job/gora-trunk/ws/trunk./gora-cassandra/target/surefire-reports/org.apache.gora.cassandra.store.TestCassandraStore.txt
>>>>
>>>>> By the way, I tested Cassandra 1.0.0, 1.0.10, 1.1.0, and 1.1.2, and got
>>>>> the
>>>>> same results as
>>>>>
>>>>> https://issues.apache.org/jira/browse/GORA-53?focusedCommentId=13414694#comment-13414694
>>>> OK. It's really just my intention to get to that stage as well.
>>>> Therefore I'm going to deleted local m2 cache and start from beginning
>>>> solely on getting the tests sorted.
>>>>
>>>> Thanks
>>>> Lewis
>>>
>>
>>
>> --
>> Lewis
>
>
