[Info-vax] OpenVMS Development Annoyances

[Kerry Main]

> -----Original Message-----
> From: Info-vax <info-vax-bounces at rbnsn.com> On Behalf Of Arne Vajhøj
> via Info-vax
> Sent: April 22, 2019 6:40 PM
> To: info-vax at rbnsn.com
> Cc: Arne Vajhøj <arne at vajhoej.dk>
> Subject: Re: [Info-vax] OpenVMS Development Annoyances
> 
> On 4/22/2019 12:37 PM, Kerry Main wrote:
> >> From: Info-vax <info-vax-bounces at rbnsn.com> On Behalf Of Arne Vajhøj
> >> On 4/21/2019 10:23 AM, seasoned_geek wrote:
> >>> On Friday, February 8, 2019 at 3:58:07 PM UTC-6, Stephen Hoffman
> wrote:
> >>>> I'd like to have a way to register an app for startup, and to
> >>>> register callbacks when the app tips over, to automatically try a
> >>>> few restarts but avoid a crash-loop, to collect failure-related
> >>>> data, and a way to check if a certain app is running via API and via shell.
> >>>> And I'd prefer to not to poll process names.  Fairly standard stuff.
> >>>> Is it running?  Why not?  Can it restart?  Can I get a notification
> >>>> when some part of production tips over?  Where do I send my
> >>>> run-time logs, and my error-, failure-, and event-related logging?
> >>>>
> >>>
> >>> You've basically described what ACMS does. It will even give you
> >>> throttle control, allowing you to set minimum and maximum numbers
> >>> for task servers.
> >>
> >> I would have considered a job scheduler and a transaction monitor to
> >> be two different things.
> >
> > Agree - they are two separate products which address two separate and
> > different sets of requirements.
> >
> >> Apparently someone at DEC agreed since they had both DECScheduler
> and
> >> ACMS.
> >
> > Yes.  Same for other OS platforms - they have separate products for
> > Job Schedulers and TP Monitors. Neither of which are typically native to the
> OS.
> > They are usually both addressed with commercial LP offerings from
> vendors.
> 
> I don't even see them target the same people.
> 
> To me a scheduler is an OPS thingy while a transaction monitor is DEV thingy.
> 
> Arne
> 

Agree, but to clarify, schedulers will often be used by both OPS and APPs teams to coordinate their batch activities.

As a simple example, 

- Initiate OPS Job A - Anti-virus scanning on specific disk areas
- if Job A completes successfully, initiate App Job B maint on specific App files, or app transfers etc. 
- If Job A fails - page on call person1
- If Job B fails - page on call person2
- if Job A and Job B both execute successfully, then initiate Job C - system backups
- If Job C fails - page on call person3

No one wants App maint / transfers or backups to start if a prev job detected a virus on that system.


Regards,

Kerry Main
Kerry dot main at starkgaming dot com