Date: Tue, 2 Sep 2003 12:59:40 -0700
Reply-To: Prasad S Ravi <prasad.s.ravi@HOUSEHOLD.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Prasad S Ravi <prasad.s.ravi@HOUSEHOLD.COM>
Subject: Re: I/O error: after the 1011th set
Content-type: text/plain; charset=us-ascii
Nothing sticks out from this list but you may be running out of work space
check your directory path to the work library which has 12.5 GB.
Or check to see if this dataset WORK.A1011LL.DATA. is corrupt(or the
variables in this are of different type)
see if the SET works without it.
Also another way, Instead of looping through all 10000, divide it into 5
different 200 datasets in each loop and set the final 5 datasets, see if it
<aldi@WUBIOS.WUSTL.ED To: SAS-L@LISTSERV.UGA.EDU
Sent by: "SAS(r) Subject: Re: I/O error: after the 1011th set
09/02/2003 12:23 PM
Please respond to
Thank you Ravi for the response,
The OS : Linux Redhat 9.0
About the sasv8.cfg: it has the following settings:
I would appreciate it if you point on which one regards to set statement.
Aldi Kraja wrote:
> I have a group of 10000 sets.
> I am putting them all together in one set by the following statement:
> data newone;
> set %do j=1 %to 10000; a&j.ll %end; ;
> Each set has about 430 obs and 3 variables, about 32KB each.
> In total I was expecting a new set of about 3.2 GB (in the SAS work dir
> I have available 12.5GB free).
> The 1011th set is a correct SAS set and is present in the directory
> where I am reading the rest of data.
> SAS complains at 1011th set for I/O error.
> Is there any limitation in how many sets I can concatenate?
> NOTE: There were 430 observations read from the data set WORK.A10000LL.
> NOTE: The data set WORK.A10000LL has 430 observations and 3 variables.
> NOTE: DATA statement used:
> real time 0.00 seconds
> cpu time 0.01 seconds
> ERROR: An I/O error has occurred on file WORK.A1011LL.DATA.