Index issue?

[Steven Waters]

Thanks for the replies, and giving ideas that pushed me towards the answer.
A few weeks ago I added a new index. I don't know if it slowed things down from the start or not.
My question was a little off. The indexes worked fine if you only used one index for output.
If you used the new index with the old indexes then output was very slow. The index I added is below;

Sort Field:  19 ■  8  ■     ■     ■     
    Length:    4■    8■     ■     ■     
    Descending:    x■    x■     ■     

The other indexes in the file are descending =""

Thanks!



On 3/26/2014 5:41 PM, Steven Waters wrote:
> Filepro 5.0.14
>
> Hello,
> Not sure what happened but on one of our files the request output 
> became very slow on all reports. This has never happened before. The file has 180,000 records.
> Inquire update add works just fine and I rebuilt all of the indexes.
> Thanks in advance with any ideas,

Slow, how?  And when?  What has changed since it used to be "fast"?

Since it's "all reports", did something change in automatic processing recently?  (Something that might cause, for example, a 1/10th second delay on each record.  Something which you wouldn't notice in dclerk, but multiply it 180,000 times in dreport, and it becomes a lot.)

Is it during sort/select or output?

If during sort/select, what happens if you go into dclerk and scan for a record that you know won't be found?  Is that also "slow"?

If you run dreport without any processing, does it also run "slow"?

     dreport filename -a -y qweqwe -z qweqwe -p /tmp/output.txt

--
Kenneth Brody