All of lore.kernel.org
 help / color / mirror / Atom feed
* widening Messenger::create
@ 2015-01-29 16:34 Matt W. Benjamin
  2015-01-29 16:42 ` Sage Weil
  0 siblings, 1 reply; 7+ messages in thread
From: Matt W. Benjamin @ 2015-01-29 16:34 UTC (permalink / raw)
  To: ceph-devel

Hi,

I'm looking at ways to pass potentially messenger-specific options through the Messenger factory.
The immediate parameter I need to provide is the number of Accelio portals to set up (this should
vary by Messenger instance), but I expect others will be needed in future.  I don't think this fits
cleanly on to the Policy mechanism, though I looked at that.  I think I'd like to provide a
generalized argument to the Messenger factory, which can do the right thing for a given Messenger
(e.g., constructor).

Thanks,

Matt

-- 
Matt Benjamin
CohortFS, LLC.
315 West Huron Street, Suite 140A
Ann Arbor, Michigan 48103

http://cohortfs.com

tel.  734-761-4689 
fax.  734-769-8938 
cel.  734-216-5309 

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: widening Messenger::create
  2015-01-29 16:34 Matt W. Benjamin
@ 2015-01-29 16:42 ` Sage Weil
  2015-01-29 16:44   ` Matt W. Benjamin
  0 siblings, 1 reply; 7+ messages in thread
From: Sage Weil @ 2015-01-29 16:42 UTC (permalink / raw)
  To: Matt W. Benjamin; +Cc: ceph-devel

On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> Hi,
> 
> I'm looking at ways to pass potentially messenger-specific options through the Messenger factory.
> The immediate parameter I need to provide is the number of Accelio portals to set up (this should
> vary by Messenger instance), but I expect others will be needed in future.  I don't think this fits
> cleanly on to the Policy mechanism, though I looked at that.  I think I'd like to provide a
> generalized argument to the Messenger factory, which can do the right thing for a given Messenger
> (e.g., constructor).

In the past we've mostly gotten away with impl-specific config options.  
Is that not sufficient here?  Because you need to pass different 
parameters to different instances maybe?

sage

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: widening Messenger::create
  2015-01-29 16:42 ` Sage Weil
@ 2015-01-29 16:44   ` Matt W. Benjamin
  2015-01-29 17:05     ` Haomai Wang
  2015-01-29 17:13     ` Sage Weil
  0 siblings, 2 replies; 7+ messages in thread
From: Matt W. Benjamin @ 2015-01-29 16:44 UTC (permalink / raw)
  To: Sage Weil; +Cc: ceph-devel

Hi,

----- "Sage Weil" <sage@newdream.net> wrote:

> On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> > Hi,
> > 
> > I'm looking at ways to pass potentially messenger-specific options
> through the Messenger factory.
> > The immediate parameter I need to provide is the number of Accelio
> portals to set up (this should
> > vary by Messenger instance), but I expect others will be needed in
> future.  I don't think this fits
> > cleanly on to the Policy mechanism, though I looked at that.  I
> think I'd like to provide a
> > generalized argument to the Messenger factory, which can do the
> right thing for a given Messenger
> > (e.g., constructor).
> 
> In the past we've mostly gotten away with impl-specific config
> options.  
> Is that not sufficient here?  Because you need to pass different 
> parameters to different instances maybe?

Yes, exactly.

> 
> sage


-- 
Matt Benjamin
CohortFS, LLC.
315 West Huron Street, Suite 140A
Ann Arbor, Michigan 48103

http://cohortfs.com

tel.  734-761-4689 
fax.  734-769-8938 
cel.  734-216-5309 

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: widening Messenger::create
  2015-01-29 16:44   ` Matt W. Benjamin
@ 2015-01-29 17:05     ` Haomai Wang
  2015-01-29 17:13       ` Matt W. Benjamin
  2015-01-29 17:13     ` Sage Weil
  1 sibling, 1 reply; 7+ messages in thread
From: Haomai Wang @ 2015-01-29 17:05 UTC (permalink / raw)
  To: Matt W. Benjamin; +Cc: Sage Weil, ceph-devel

I think it may better that Messenger impl can introspect itself by
"type", "policy", config and related infos?

On Fri, Jan 30, 2015 at 12:44 AM, Matt W. Benjamin <matt@cohortfs.com> wrote:
> Hi,
>
> ----- "Sage Weil" <sage@newdream.net> wrote:
>
>> On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
>> > Hi,
>> >
>> > I'm looking at ways to pass potentially messenger-specific options
>> through the Messenger factory.
>> > The immediate parameter I need to provide is the number of Accelio
>> portals to set up (this should
>> > vary by Messenger instance), but I expect others will be needed in
>> future.  I don't think this fits
>> > cleanly on to the Policy mechanism, though I looked at that.  I
>> think I'd like to provide a
>> > generalized argument to the Messenger factory, which can do the
>> right thing for a given Messenger
>> > (e.g., constructor).
>>
>> In the past we've mostly gotten away with impl-specific config
>> options.
>> Is that not sufficient here?  Because you need to pass different
>> parameters to different instances maybe?
>
> Yes, exactly.
>
>>
>> sage
>
>
> --
> Matt Benjamin
> CohortFS, LLC.
> 315 West Huron Street, Suite 140A
> Ann Arbor, Michigan 48103
>
> http://cohortfs.com
>
> tel.  734-761-4689
> fax.  734-769-8938
> cel.  734-216-5309
> --
> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html



-- 
Best Regards,

Wheat

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: widening Messenger::create
  2015-01-29 16:44   ` Matt W. Benjamin
  2015-01-29 17:05     ` Haomai Wang
@ 2015-01-29 17:13     ` Sage Weil
  1 sibling, 0 replies; 7+ messages in thread
From: Sage Weil @ 2015-01-29 17:13 UTC (permalink / raw)
  To: Matt W. Benjamin; +Cc: ceph-devel

On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> Hi,
> 
> ----- "Sage Weil" <sage@newdream.net> wrote:
> 
> > On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> > > Hi,
> > > 
> > > I'm looking at ways to pass potentially messenger-specific options 
> > > through the Messenger factory. The immediate parameter I need to 
> > > provide is the number of Accelio portals to set up (this should vary 
> > > by Messenger instance), but I expect others will be needed in 
> > > future.  I don't think this fits cleanly on to the Policy mechanism, 
> > > though I looked at that.  I think I'd like to provide a generalized 
> > > argument to the Messenger factory, which can do the right thing for 
> > > a given Messenger (e.g., constructor).
> > 
> > In the past we've mostly gotten away with impl-specific config
> > options.  
> > Is that not sufficient here?  Because you need to pass different 
> > parameters to different instances maybe?
> 
> Yes, exactly.

Hmm, in that case I'd say if you can make it as generic as possible that's 
about all we can do that this point.  Just a single arg now?  We can 
easily expand the interface to a param struct later.  (Or do that now if 
you prefer.. whatever works and looks reasonable!)

sage

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: widening Messenger::create
  2015-01-29 17:05     ` Haomai Wang
@ 2015-01-29 17:13       ` Matt W. Benjamin
  0 siblings, 0 replies; 7+ messages in thread
From: Matt W. Benjamin @ 2015-01-29 17:13 UTC (permalink / raw)
  To: Haomai Wang; +Cc: Sage Weil, ceph-devel

There is obvious overlap with policy, but policy is currently not well
structured for this use.

Changing or extending the policy interface seemed plausible, but also seemed
likely to run into more objections.

Matt

----- "Haomai Wang" <haomaiwang@gmail.com> wrote:

> I think it may better that Messenger impl can introspect itself by
> "type", "policy", config and related infos?
> 
> On Fri, Jan 30, 2015 at 12:44 AM, Matt W. Benjamin <matt@cohortfs.com>
> wrote:
> > Hi,
> >
> > ----- "Sage Weil" <sage@newdream.net> wrote:
> >
> >> On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> >> > Hi,
> >> >
> >> > I'm looking at ways to pass potentially messenger-specific
> options
> >> through the Messenger factory.
> >> > The immediate parameter I need to provide is the number of
> Accelio
> >> portals to set up (this should
> >> > vary by Messenger instance), but I expect others will be needed
> in
> >> future.  I don't think this fits
> >> > cleanly on to the Policy mechanism, though I looked at that.  I
> >> think I'd like to provide a
> >> > generalized argument to the Messenger factory, which can do the
> >> right thing for a given Messenger
> >> > (e.g., constructor).
> >>
> >> In the past we've mostly gotten away with impl-specific config
> >> options.
> >> Is that not sufficient here?  Because you need to pass different
> >> parameters to different instances maybe?
> >
> > Yes, exactly.
> >
> >>
> >> sage
> >
> >
> > --
> > Matt Benjamin
> > CohortFS, LLC.
> > 315 West Huron Street, Suite 140A
> > Ann Arbor, Michigan 48103
> >
> > http://cohortfs.com
> >
> > tel.  734-761-4689
> > fax.  734-769-8938
> > cel.  734-216-5309
> > --
> > To unsubscribe from this list: send the line "unsubscribe
> ceph-devel" in
> > the body of a message to majordomo@vger.kernel.org
> > More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 
> 
> 
> -- 
> Best Regards,
> 
> Wheat
> --
> To unsubscribe from this list: send the line "unsubscribe ceph-devel"
> in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
Matt Benjamin
CohortFS, LLC.
315 West Huron Street, Suite 140A
Ann Arbor, Michigan 48103

http://cohortfs.com

tel.  734-761-4689 
fax.  734-769-8938 
cel.  734-216-5309 

^ permalink raw reply	[flat|nested] 7+ messages in thread

* Re: widening Messenger::create
       [not found] <1917094814.68.1422552099121.JavaMail.root@thunderbeast.private.linuxbox.com>
@ 2015-01-29 17:23 ` Matt W. Benjamin
  0 siblings, 0 replies; 7+ messages in thread
From: Matt W. Benjamin @ 2015-01-29 17:23 UTC (permalink / raw)
  To: Sage Weil; +Cc: ceph-devel

I was going to do a proplist arg type thing?  I'll adapt it until hopefully shared happiness breaks out...

Matt

----- "Sage Weil" <sage@newdream.net> wrote:

> On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> > Hi,
> > 
> > ----- "Sage Weil" <sage@newdream.net> wrote:
> > 
> > > On Thu, 29 Jan 2015, Matt W. Benjamin wrote:
> > > > Hi,
> > > > 
> > > > I'm looking at ways to pass potentially messenger-specific
> options 
> > > > through the Messenger factory. The immediate parameter I need to
> 
> > > > provide is the number of Accelio portals to set up (this should
> vary 
> > > > by Messenger instance), but I expect others will be needed in 
> > > > future.  I don't think this fits cleanly on to the Policy
> mechanism, 
> > > > though I looked at that.  I think I'd like to provide a
> generalized 
> > > > argument to the Messenger factory, which can do the right thing
> for 
> > > > a given Messenger (e.g., constructor).
> > > 
> > > In the past we've mostly gotten away with impl-specific config
> > > options.  
> > > Is that not sufficient here?  Because you need to pass different 
> > > parameters to different instances maybe?
> > 
> > Yes, exactly.
> 
> Hmm, in that case I'd say if you can make it as generic as possible
> that's 
> about all we can do that this point.  Just a single arg now?  We can 
> easily expand the interface to a param struct later.  (Or do that now
> if 
> you prefer.. whatever works and looks reasonable!)
> 
> sage

-- 
Matt Benjamin
CohortFS, LLC.
315 West Huron Street, Suite 140A
Ann Arbor, Michigan 48103

http://cohortfs.com

tel.  734-761-4689 
fax.  734-769-8938 
cel.  734-216-5309 

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2015-01-29 17:23 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
     [not found] <1917094814.68.1422552099121.JavaMail.root@thunderbeast.private.linuxbox.com>
2015-01-29 17:23 ` widening Messenger::create Matt W. Benjamin
2015-01-29 16:34 Matt W. Benjamin
2015-01-29 16:42 ` Sage Weil
2015-01-29 16:44   ` Matt W. Benjamin
2015-01-29 17:05     ` Haomai Wang
2015-01-29 17:13       ` Matt W. Benjamin
2015-01-29 17:13     ` Sage Weil

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.