Re: monit '-i' option?

[Rory Toma]

If you are going to have monit start services at boot time, then we can
do something like with poll cycles. Each process can have a "skip x
amount of cycles..."

Starting services with monit is not always desirable... In my case, it
becomes difficult, as all the 10 processes or so have a dependency
chain.

I currently solve this now by having monit sleep for 5 minutes on
boot...

On Mon, 2003-09-15 at 16:27, Jan-Henrik Haukeland wrote:
> Martin Pala <address@hidden> writes:
> 
> > My vote is +0 for readding '-i' with initial test cycles skip count
> > configurable extension (as Rory recommended) and -1 for readding it
> > in original form (skip validate in first cycle only).
> 
> Thanks for the explanation Rory, I now understand the rationale for
> adding -i and how it in fact might be a functional and practical
> workaround in the scenario you describe, but as Martin augmented it
> *is* a workaround and does not correct the fundamental problem. 
> 
> The problem with this solution is that it _might_ work, but on the
> other hand it might _not_ work. The problem is that it is impossible
> to know _a priori_ the value of -i and for how long monit should wait
> before it should start monitoring services. E.g. should the value for
> -i be 3 or 10 and if you add a new service to the system, which value
> should now be used for -i or if the daemon poll time was changed how
> does this affect the value of -i and so on.
> 
> In other words I think we should try to find a solution to the problem
> otherwise and it might very well be that Martin's suggestion is the
> best solution. That is, to have monit start services at boot time
> instead of init. I don't know, but I do not think that re-adding -i is
> the best solution so my vote is, -1 in both cases.
-- 
Rory Toma               address@hidden
VP of Run Level 9       http://www.trs80.net
Digeo Digital           http://www.digeo.com

signature.asc
Description: This is a digitally signed message part