I have two databases aside from the default database and have been trying
to set up timed rewalks without success.
I telnet into the ISP to work with Webinator (we've set up a cron job to
launch gw to rewalk the default database).
There's reference to running gw "in the background if you want to use the
scheduler" but I don't see a command to do that.
I invoke gw -dDB -rewalk="TIME_SPEC" and it seems to start alright but if
it finishes I don't get the telnet prompt back. Sometimes the rewalk
doesn't finish, leaving the temporary database stranded.
Cron is the unix scheduler. You said you are already using that for one
rewalk already. If you want to let gw schedule itself, you run it in the
background and use the -rewalk=TIME_SPEC option. -rewalk=TIME_SPEC is
provided mainly for those without access to cron like facilities. But it
doesn't hurt to use it instead of cron. Cron has the advantage of working
automatically after a system boot. If you use TIME_SPEC and the system boots,
you will have to log back in to restart it.
-rewalk="TIME_SPEC" will never return to the shell prompt unless it is killed
or placed in the background. The idea is to stay running to wait around for
the next rewalk time.
Running programs in the background is a standard unix feature. Just
put an & at the end of the command line you want to run in
the background. Normally all of your jobs will be killed when
you logout (standard unix behavior). To prevent that use the
standard unix "nohup" command.
To run gw in background and let it continue when you logout:
nohup gw ...options... &
See your system man pages about the nohup command and the shell for
backgrounding jobs.
This is very interesting for the UNIX user. Is there anyone out there
running the Windows NT version? The cron equivalent here is 'at' and so
far, I've not been able to get it to run properly in the background. It
kicks off, but nothing happens. We have all the permissions set up for
us, so far as I know. What should we be looking at and what's the best
way to debug it?
we have provided the -rewalk and -rewalk=TIMESPEC options so that gw
can rewalk on a schedule via a task scheduler, such as cron or at, or
if the process can be left running it can take schedule itself.
While we can not debug the AT program, a couple of suggestions which
might produce some more helpful output follow. You might want to
create a batch file which runs gw, and redirects output to a file so
that you can see the output that is produced by gw. If the job does
not run properly you might also want to check if there are any logs
produced by AT as to why it does not start the job.
