From previous posts you are definitely doing "time" based archives instead of "full"....
Time: Every x minutes you say "move to next ai extent".
Full: Every x minutes you check to see if you have any full AI extents.
I like using the Full method for a couple of reasons:
1) No wasted...
More fun!
If you look at the mbpro shell script in your $DLC/bin dir you'll see (pardon me, my v8 is showing):
$ cat mbpro
#!/bin/sh
DLC=${DLC-/gpapp/dlc83b};export DLC
PROEXE=${PROEXE-$DLC/bin/_progres}
echo Batch processing will be performed using:
$DLC/version
echo
trap "" 1...
As Murray said, use 1!
Why: Using a value of one says "use as much space as you want" when applied to indexes only.
At a 8,192 byte /128 recs per block you are going with a fixed size of 64 bytes per rec. If you run a idxanal and find a mean size of 32 bytes (I am deliberately not not...
This is a fun one!
The issue is that values in excess of 2 billion roll into negative values. This is because the sign bit gets written into.
I've got a workaround I did for values into the 4 billion range that goes something like:
assign lr = _Buffer-LogicRds / 10.
if lr < 0 then lr =...
Hey Keith,
Do as Toby said! Here is a quick script example for 1 GB backup extents (using 8k block sizes). Once it actually fills what it needs the rest of the entries just get disarded.
probkup $db $backupdir/$db.bkup01 -vs 125000 <<END
$backupdir/$db.bkup02
$backupdir/$db.bkup03...
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.