![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Hi, I have a CSV file (24,584 recs, ~230-240) fields that fails DTS on data insert with "not enough storage to complete operation" listed twice in the message box. Although I have numeric data in many fields, I am importing all as text because the text fields are not delimited with quotes or apostrophes. I have tried ending the last field with comma linefeed and linefeed alone. The table is created with all the field names from the first row correctly. Each field is a varchar(8000). I've tried the following: - Service pack 4, no difference - Bulk insert in Query Analyzer which fails complaining about a field being too large. It would be the last field on the first row, or the the first field on the second (sorry, I forget what I changed to make this happen). - Looked at the file in a hex editor and saw that each record is terminated by 0x0A, and insured my record separator in DTS and in the bulk query is set to 'LF'. - Imported same file into Microsoft Access without modification or difficulty and it chose column types more appropriately. I can "punk out" at this point, but I really would like to know what I'm doing wrong in DTS. It seems like a relatively simple import, and I've spent some time on it! I've also tried DTS to import Paradox files in the past and choked on that as well. I'm usually fairly proficient at learning new tools, but this one is an exception. Thanks in Advance for any help! Regards, Jim Shedden |
#3
| |||
| |||
|
#4
| |||
| |||
|
|
Hi Ilya, Thanks for the reply. I remain ignorant on a few things however, so if you have the time... I see the error metioned with regard to "lookups" and "data pumps". Is streaming in a CSV file considered a lookup? Is a "data pump" also used when I try the "bulk insert" query, or is that a component only of DTS? I assume that we are saying that a W2K service pack broke the process somehow and that SQL2Ksp4 does not actually correct it? Shall I abandon DTS for good to avoid wasting more time? Right now it seems best to import to Access, then push the data from there, or use only BCP, correct? If so, it doesn't seem very user friendly to a newcomer, nor well explained in the MSDN knowledge base (great resource!). Please note that performance in my case will rarely be an issue if success can be achieved. Thanks Again! Regards, Jim Shedden |
![]() |
| Thread Tools | |
| Display Modes | |
| |