Jacek Becla a écrit :
> Emmanuel
>
> Do you still have this problem? I think we mentioned this
> several times, but maybe we didn't explain in details: it
> looks like you are using a database called
> "qms_clrlsst-dbwkr2.in2p3.fr". Please change it to a name
> that does not have dots, eg qms_clrlsst-dbwkr2_in2p3_fr
>
> Mysql tends to get very confused when there is a dot in
> database name. I added a code that disallows that over
> the weekend (it is still in #2725 though)
>
Hello Jacek,
Thanks, Fabrice and I have solved this issue.
Could you please :
- review ticket #2796 (Douglas)
- review ticket #2767 (Daniel)
Sorry for insisting, but we are waiting for reviews since a few weeks and we are
ready to merge.
Thanks in advance for your answer.
Emmanuel
> Jacek
>
>
>
>
>
> On 04/28/2013 11:20 PM, Emmanuel Medernach wrote:
>> Jacek Becla a écrit :
>>> Emmanuel
>>>
>>> Feel free to dump any quick & dirty scripts you write that are
>>> far from production quality but nevertheless might be useful
>>> (reused) in the future. Also, it helps with reproducing what
>>> we have done...
>>>
>>> Jacek
>>>
>>
>> Hello Jacek,
>>
>> Here is the QMS branch I am currently working on :
>> Qserv branch: u/medernac/tickets_2837
>>
>> I installed it with scons and get these errors when running
>> $ qserv-start
>>
>> Here is an extract from the QMS log:
>>
>> 2013-04-25 16:42:33,802 DEBUG connected to mysql (port)
>> 2013-04-25 16:42:35,821 DEBUG Connected to db qms_clrlsst-dbwkr2.in2p3.fr
>> 2013-04-25 16:42:35,821 DEBUG Executing SELECT dbName FROM DbMeta
>> 2013-04-25 16:42:35,822 ERROR MySQL Error [1146]: Table
>> 'qms_clrlsst-dbwkr2.in2p3.fr.DbMeta' doesn't exist
>> 2013-04-25 16:42:35,823 ERROR Exception in listDbs: Internal MySQL
>> error. (MySQL
>> Error [1146]: Table 'qms_clrlsst-dbwkr2.in2p3.fr.DbMeta' doesn't exist)
>> 2013-04-25 16:53:18,436 DEBUG connected to mysql (port)
>> 2013-04-25 16:53:20,419 DEBUG Connected to db qms_clrlsst-dbwkr2.in2p3.fr
>> 2013-04-25 16:53:20,419 DEBUG Executing SELECT dbName FROM DbMeta
>> 2013-04-25 16:53:20,422 ERROR MySQL Error [1146]: Table
>> 'qms_clrlsst-dbwkr2.in2p3.fr.DbMeta' doesn't exist
>> 2013-04-25 16:53:20,424 ERROR Exception in listDbs: Internal MySQL
>> error. (MySQL
>> Error [1146]: Table 'qms_clrlsst-dbwkr2.in2p3.fr.DbMeta' doesn't exist)
>> 20
>>
>>
>>
>> I have another issue, I extracted some Object/Source from PT12 in
>> case04 tests.
>> I used Object column 3 as RA and 5 as declination and Source 34 as RA and
>> columns 35 as declination. Here are the tables created when trying to
>> partition/load it :
>>
>> | Object_100 |
>> | Source_180 |
>>
>> Something seems wrong here, as the chunkID is not the same for Object
>> and Source.
>>
>> Indeed, the SourcePartitions seems wrong :
>>
>> mysql> select * from SourcePartitions ;
>> +---------+------------+---------+-------+-------+---------+---------------+---------+-----------------+
>>
>> | chunkId | subChunkId | numRows | raMin | raMax | declMin |
>> declMax |
>> overlap | alpha |
>> +---------+------------+---------+-------+-------+---------+---------------+---------+-----------------+
>>
>> | 179 | 5 | 76797 | 1368 | 360 | 63
>> | 72 |
>> 0.01667 | 0.0539452603939 |
>> | 180 | 5 | 83979 | 0 | 360 | 81 |
>> 90.0000002778 |
>> 0.01667 | 180 |
>> +---------+------------+---------+-------+-------+---------+---------------+---------+-----------------+
>>
>> 2 rows in set (0.00 sec)
>>
>> The first row has a raMin of 1368, which is fishy and the second row
>> has a
>> declMax greater than 90 degrees.
>>
>> The tables are correctly extracted :
>>
>> zcat Object.csv.gz | awk -F',' '/./ { print $1, $3, $5; }' >
>> Object.radecl
>> sort Object.radecl | uniq > Object.radecl.u
>>
>> zcat Source.csv.gz | awk -F',' '/./ { print $4, $34, $35; }' >
>> Source.radecl
>> sort Source.radecl | uniq > Source.radecl.u
>>
>> diff Object.radecl.u Source.radecl.u
>>
>> => returns nothing
>>
>>
>> You could find the case04 data used here:
>> Qservdata branch: u/medernac/pt12
>>
>> An the corresponding branch here:
>> Qserv branch: u/medernac/dataTests
>>
>> Help would be much appreciated.
>>
>> Regards,
>> --
>>
>
>
########################################################################
Use REPLY-ALL to reply to list
To unsubscribe from the QSERV-L list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=QSERV-L&A=1
