[sr-dev] db_cassandra - db_cassa_raw_query & INSERT query

Wed Mar 12 11:26:34 CET 2014

Im not sure whats going on...
I have a bunch of patches and im now having all kinds of trouble after
stashing, and moving up to git head.

back to all sorts of things being broken.

Sorry to trouble....
hopefully I can work through these issues again and update patches..

On 12 March 2014 19:52, Daniel-Constantin Mierla <miconda at gmail.com> wrote:

>  I tried this one and it doesn't apply either. Is the right patch for
> master head?
>
> Cheers,
> Daniel
>
>
> On 12/03/14 10:49, jay binks wrote:
>
> Sigh... seems my last patch was not against head...
>
>  new patch... git log..
>  commit 99d50e2da0753d7482ed2884e665e08e235daf5e
> Author: Jason Penton <jason.penton at gmail.com>
> Date:   Mon Mar 10 19:49:39 2014 +0200
>
>  Sorry all !
>
>
>
> On 12 March 2014 19:32, jay binks <jaybinks at gmail.com> wrote:
>
>> I just explicitly testing this.
>>
>>  Results :
>>
>>  A sane query, but table dosnt exist performed as expected :
>>
>>   avp_db_query("INSERT INTO tablenothere ( KEY, added ) VALUES ( '$si',
>> '$Ts' );");
>>  0(26936) ERROR: db_cassandra [dbcassa_base.cpp:729]:
>> db_cassa_raw_query(): Invalid Request caused error details: unconfigured
>> columnfamily tablenothere.
>>
>>  And insane query where its virtually just crap in a statement also
>> behaved well :
>>
>>  avp_db_query("INSERT INTO tablenothere ( idont enven K'now how to Sql");
>>  0(26913) ERROR: db_cassandra [dbcassa_base.cpp:729]:
>> db_cassa_raw_query(): Invalid Request caused error details: line 1:55
>> mismatched character '<EOF>' expecting '''.
>>
>>  Id say the answer to your question is yes, my patch works as expected
>> in this regard.
>>
>>  Jay
>>
>>
>>
>>
>>
>>
>> On 12 March 2014 19:27, Daniel-Constantin Mierla <miconda at gmail.com>wrote:
>>
>>>
>>> On 12/03/14 10:16, jay binks wrote:
>>>
>>> In my test case I was doing an INSERT query...
>>> yet db_cassandra would complain there was no result... ( both the log
>>> message and return code )
>>>
>>>  In understand that and you are right here -- even select can have no
>>> result (but maybe is setting some other fields there). What I want to
>>> clarify is that in case of a query error (e.g., wrong statement or
>>> something happened with the connection), is it detected? Not to behave like
>>> it was all ok.
>>>
>>> Cheers,
>>> Daniel
>>>
>>>
>>>
>>>  This is the reason I provided the patch.
>>>
>>>  after a little more testing I have found that I get this log message :
>>>
>>>  0(23827) ERROR: <core> [db_res.c:130]: db_free_result(): invalid
>>> parameter
>>>
>>>  So far in my testing everything has performed flawlessly, just with a
>>> few less log lines :)
>>>
>>>  in essence this patch simply makes db_cassandra act the same when
>>> there is no result set as it does when there are now rows.
>>> ( previously it would act like no result set was a big deal )
>>>
>>>  Jay
>>>
>>>
>>>
>>>
>>>
>>> On 12 March 2014 18:49, Daniel-Constantin Mierla <miconda at gmail.com>wrote:
>>>
>>>>  What would be the situation when the query is like SELECT but there is
>>>> no result. Is the behaviour as expected with the new patch?
>>>>
>>>> Anyone here using cassandra having comments? From my point of view is
>>>> no problem to push the patch, but I am not using cassandra, so cannot do a
>>>> proper review.
>>>>
>>>> Cheers,
>>>> Daniel
>>>>
>>>>
>>>> On 12/03/14 08:53, jay binks wrote:
>>>>
>>>>  If doing a query that returns no results ( Insert etc )
>>>> db_cassa_raw_query would cause these ERRORS to be logged
>>>>
>>>>  0(22283) ERROR: db_cassandra [dbcassa_base.cpp:739]:
>>>> db_cassa_raw_query(): The resultype rows was not set, no point trying to
>>>> parse result.
>>>> 0(22283) ERROR: avpops [avpops_db.c:333]: db_query_avp(): cannot do
>>>> the query
>>>>
>>>>  db_cassa_raw_query would also return -1 as a failure code which
>>>> caused avpops_db to log the query failure.
>>>>
>>>>  my patch changes the db_cassa_raw_query log message to debug level,
>>>> and returns success from the function.
>>>>
>>>>  I had a quick look to see if there was an elegant way to determine if
>>>> we should expect results, so we can vary the response code based on query
>>>> type, but I was unable to find anything other than doing string comparisons
>>>> on the query, so I opted to not bother with this as it would be erroneous.
>>>>
>>>>  Please find attached patch.
>>>>
>>>>  --
>>>> Sincerely
>>>>
>>>> Jay
>>>>
>>>>
>>>>  _______________________________________________
>>>> sr-dev mailing listsr-dev at lists.sip-router.orghttp://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>>>>
>>>>
>>>> --
>>>> Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
>>>> Kamailio World Conference - April 2-4, 2014, Berlin, Germanyhttp://www.kamailioworld.com
>>>>
>>>>
>>>> _______________________________________________
>>>> sr-dev mailing list
>>>> sr-dev at lists.sip-router.org
>>>> http://lists.sip-router.org/cgi-bin/mailman/listinfo/sr-dev
>>>>
>>>>
>>>
>>>
>>>  --
>>> Sincerely
>>>
>>> Jay
>>>
>>>
>>> --
>>> Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
>>> Kamailio World Conference - April 2-4, 2014, Berlin, Germanyhttp://www.kamailioworld.com
>>>
>>>
>>
>>
>>   --
>> Sincerely
>>
>> Jay
>>
>
>
>
>  --
> Sincerely
>
> Jay
>
>
> --
> Daniel-Constantin Mierla - http://www.asipto.comhttp://twitter.com/#!/miconda - http://www.linkedin.com/in/miconda
> Kamailio World Conference - April 2-4, 2014, Berlin, Germanyhttp://www.kamailioworld.com
>
>

-- 
Sincerely

Jay
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.sip-router.org/pipermail/sr-dev/attachments/20140312/5ec706fd/attachment.html>