[Info-vax] Are queue manager updates written to disk immediately ?

[Jan-Erik Soderholm]

David Froble wrote 2013-04-12 00:02:
> Jan-Erik Soderholm wrote:
>
>> I guess accounting (acc/queue=ba0) also shows both jobs as runed?
>> Nothing weird with the start/finish timestamps in accounting?
>>
>> I would look mare at something with the system clock at
>> startup that made the holding job to be released. Such
>> as a startup with wrong time setting or similar.
>>
>> That is, queue ba0 is started by the startup before the
>> clock is corrected. Or something like that.
>>
>> Jan-Erik.
>
> Good idea of something to look at.
>
> I too wondered about the clock, but what Simon posted didn't seem to have
> any time stamps that looked improper.
>
> If it's as reported, then I'd say it's definitely a problem.
>
> That said, I'd also suggest that if the batch job running more than once is
> a problem, then perhaps some type of flag showing last run time of the
> procedure might be in order.  If it is at least written to disk in a timely
> manner.  Regardless, there could be multiple reasons for a batch job to run
> when you don't want it to run.  Now, if it's just a report, so what if it's
> run twice.
>
> I usually design in such sequence info in my designs when appropriate.

Well, from the name of the procedure ("CHECK_QUEUES") I thought that
it would be safe to run it twice, but the question about what actualy
happend is still interesting. :-)

And maybe you *do* want to "check your queues" right after
a reboot anyway! :-)

Actualy, it might even be designed to run that particualr
script right after any boot...

Jan-Erik.

Jan-Erik.