Hi!
In a system, which collects data and the summarises it, we collect a
large amount of data.
Now, there seems to be a problem, when we try to delete just 16000
rows (entire table btw) or some 250000 rows in another table. My test
table is totally 94 MB, but larger have been seen :-)
The error message reads: "Record is deleted" (16k rows table) or "The
search key is not found in any record" (250000 rows, leaving some rows
to survive)
Solution: probably the solution is to delete data patially. I tried to
do that.
Deleting is done as: delete from whatever where date < now-
some_days_say_3 and source_id=56
It loops for each source, but that it not enough.
We do not know how old data we have, but a solution could be to read
in min(date), then do it one day at a time. But is there a better way
to do it in one sql, one step?
We use Delphi and ADO connection through ODBC.
WBR
Sonnich
In a system, which collects data and the summarises it, we collect a
large amount of data.
Now, there seems to be a problem, when we try to delete just 16000
rows (entire table btw) or some 250000 rows in another table. My test
table is totally 94 MB, but larger have been seen :-)
The error message reads: "Record is deleted" (16k rows table) or "The
search key is not found in any record" (250000 rows, leaving some rows
to survive)
Solution: probably the solution is to delete data patially. I tried to
do that.
Deleting is done as: delete from whatever where date < now-
some_days_say_3 and source_id=56
It loops for each source, but that it not enough.
We do not know how old data we have, but a solution could be to read
in min(date), then do it one day at a time. But is there a better way
to do it in one sql, one step?
We use Delphi and ADO connection through ODBC.
WBR
Sonnich
Comment