You are not performing a point-in-time-restore, you
are performing a point-in-log-restore.
Are you sure that the physical restore was ok ?
If you are getting the error at logical log number 150 and
you have issued the command "onbar -r -w -n 2817" there
are about 2667 logical logs still left to process. This sounds
not very plausible.
I would recommend that you split the restore in two parts.
First do a physical restore:
- onbar -r -w -p
Second do the logical point-in-time-restore (if it is
necessary, otherwise you might execute 'onmode -m'
after the physical restore):
- onbar -r -r -t "<datetime_value>"
Check you online.log and your onbar-activity-log for
any abnormal messages during the restore.
And my last recommendation is to choose a reliable
platform for your database, i.e. Unix or Linux.