April 28, 2017 at 10:43 am
Intermittently we are getting column header names corrupted in the output text file when executing a stored proc from a job in SSIS step. This is 15 yr old process that has only started to fail in a 2008 R2 cluster env.
Note: Actual data is unaffected.
The pseudo proc sql looks something like this:
select 'tableA' as XTABLENAME, count(*) as XCOUNT from schemaA.tableA
SELECT col1
FROM schemaA,tableA
ORDER BY col1
select 'tableB' as XTABLENAME, count(*) as XCOUNT from schemaA.tableB
SELECT col1, col2, col3
FROM schemaA,tableB
ORDER BY col1
...
-------------------end proc----------------------------------------
The job executes
USE msdb
GO
EXEC sp_start_job @job_name = 'xpr_extract'
GO
WAIT FOR DELAY '000:05:00'
-----------------------end job-------------------------------------
--- expected
XTABLENAME XCOUNT
------------- -----------
tableA 1
col1
---------------
123456789012345
XTABLENAME XCOUNT
------------- -----------
tableB 3
col1 col2 col3
--------------- --------------- ----------
123456789012345 000000000000012 0000000001
123456789012345 000000000000012 0000000002
123456789012345 000000000000015 0000000001
---------------------------------end expected----------------------
Columns get corrupted like this:
XTABLENAME XCOUNT
------------- -----------
tableA 1
col1
---------------
123456789012345
XTABLENAME XCOUNT
------------- -----------
tableB 3
col1 _xxx2 x_xxx
--------------- --------------- ----------
123456789012345 000000000000012 0000000001
123456789012345 000000000000012 0000000002
123456789012345 000000000000015 0000000001
August 31, 2017 at 7:49 am
Could SQL Agent limitations and concurrent re-indexing have bearing on output?
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply