You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@geode.apache.org by Anilkumar Gingade <ag...@pivotal.io> on 2016/06/02 18:10:30 UTC
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Hi Team,
As proposed here, we added support to propagate eviction and expiration
(destroy) operation to AsyncEventQueue using single flag/attribute
"ignoreEvictionAndExpiration" by default which is true (to keep the same
behavior) and one could set (false) to receive eviction/expiration event...
But we come across a product issue, GEODE-1472, that cause data
inconsistency (with eviction destroy)....For this reason we are planning to
break the "ignoreEvictionAndExpiration" attribute to eviction and
expiration specific:
"ignoreEvictionDestroy", "ignoreExpirationDestroy"...
Currently we are planning to support "ignoreExpirationDestroy", and add
"ignoreEvictionDestroy" once GEODE-1472 is fixed...
Looking for comments on this...
Thanks,
-Anil.
https://issues.apache.org/jira/browse/GEODE-1472
On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <ag...@pivotal.io>
wrote:
> Kirk, We could not think of any such requirement...And with this
> application will get all the update operation and can take appropriate
> action (use or ignore)...
>
> -Anil.
>
>
>
> On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io> wrote:
>
>> Would any user ever have a reason to enable forwarding of one type but not
>> the other? If so then I would separate them as forwardEvictionEvents() and
>> forwardExpirationEvents().
>>
>>
>> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io> wrote:
>>
>> > +1 for being more explicit with the "And" conjunction
>> >
>> > -Kirk
>> >
>> >
>> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <ab...@pivotal.io>
>> wrote:
>> >
>> >> I’d prefer to insert a conjunction to clarify the meaning:
>> >>
>> >> forwardEvictionAndExpirationEvents()
>> >>
>> >>
>> >> $0.02,
>> >> Anthony
>> >>
>> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <ag...@pivotal.io>
>> >> wrote:
>> >>
>> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any alternate
>> >> names?).
>> >>
>> >>
>> >>
>> >
>>
>
>
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Posted by William Markito <wm...@pivotal.io>.
+1
On Fri, Jun 3, 2016 at 3:32 PM, Dan Smith <ds...@pivotal.io> wrote:
> +1 for splitting it up and getting rid of the Eviction option for now.
>
> I think I was the one who suggested setIgnoreExpiration originally.
> Ignore/Forward, I guess it's not really a big deal. But since you are
> setting a property on the queue, I think it should be setForwardExpiration,
> not just forwardExpiration. I believe the current property is
> setIgnoreXXXX.
>
> -Dan
>
> On Fri, Jun 3, 2016 at 12:25 PM, Anilkumar Gingade <ag...@pivotal.io>
> wrote:
>
> > Yes...Jason, thanks for correction...
> >
> > [corrected]
> > I will wait for other feedbacks/comments, if there is no objection i
> > will change
> > it to "forwardExpirationDestroy" with default value set to false.
> >
> > -Anil.
> >
> >
> >
> > On Fri, Jun 3, 2016 at 12:01 PM, Jason Huynh <jh...@pivotal.io> wrote:
> >
> > > *I will wait for other feedbacks/comments, if there is no objection i
> > > willchange it to "forwardExpirationDestroy" with default value set to
> > > true.*
> > >
> > > think you meant to say default value to set false :-)
> > >
> > > On Fri, Jun 3, 2016 at 11:47 AM, Anilkumar Gingade <
> agingade@pivotal.io>
> > > wrote:
> > >
> > > > Thanks Darrel...
> > > >
> > > > The name change was based on the review comments:
> > > > https://reviews.apache.org/r/46243/
> > > >
> > > > I am fine with either one, ignoreXXX or forwardXXX.
> > > >
> > > > I will wait for other feedbacks/comments, if there is no objection i
> > will
> > > > change it to "forwardExpirationDestroy" with default value set to
> true.
> > > >
> > > > Thanks,
> > > > -Anil.
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > > > On Thu, Jun 2, 2016 at 2:14 PM, Darrel Schneider <
> > dschneider@pivotal.io>
> > > > wrote:
> > > >
> > > > > When did forwardXXX become ignoreXXX? I read through the email
> thread
> > > and
> > > > > couldn't find why that happened. It is best for the default on a
> > > boolean
> > > > > property to be false. That was the case when it was forwardXXX. But
> > now
> > > > > that it has changed to ignoreXXX the default has become true. I'd
> > vote
> > > > for
> > > > > it being named something whose default can be false.
> > > > >
> > > > >
> > > > > On Thu, Jun 2, 2016 at 11:10 AM, Anilkumar Gingade <
> > > agingade@pivotal.io>
> > > > > wrote:
> > > > >
> > > > > > Hi Team,
> > > > > >
> > > > > > As proposed here, we added support to propagate eviction and
> > > expiration
> > > > > > (destroy) operation to AsyncEventQueue using single
> flag/attribute
> > > > > > "ignoreEvictionAndExpiration" by default which is true (to keep
> the
> > > > same
> > > > > > behavior) and one could set (false) to receive
> eviction/expiration
> > > > > event...
> > > > > >
> > > > > > But we come across a product issue, GEODE-1472, that cause data
> > > > > > inconsistency (with eviction destroy)....For this reason we are
> > > > planning
> > > > > to
> > > > > > break the "ignoreEvictionAndExpiration" attribute to eviction and
> > > > > > expiration specific:
> > > > > > "ignoreEvictionDestroy", "ignoreExpirationDestroy"...
> > > > > >
> > > > > > Currently we are planning to support "ignoreExpirationDestroy",
> > and
> > > > add
> > > > > > "ignoreEvictionDestroy" once GEODE-1472 is fixed...
> > > > > >
> > > > > > Looking for comments on this...
> > > > > >
> > > > > > Thanks,
> > > > > > -Anil.
> > > > > >
> > > > > > https://issues.apache.org/jira/browse/GEODE-1472
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <
> > > > agingade@pivotal.io>
> > > > > > wrote:
> > > > > >
> > > > > > > Kirk, We could not think of any such requirement...And with
> this
> > > > > > > application will get all the update operation and can take
> > > > appropriate
> > > > > > > action (use or ignore)...
> > > > > > >
> > > > > > > -Anil.
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io>
> > > wrote:
> > > > > > >
> > > > > > >> Would any user ever have a reason to enable forwarding of one
> > type
> > > > but
> > > > > > not
> > > > > > >> the other? If so then I would separate them as
> > > > forwardEvictionEvents()
> > > > > > and
> > > > > > >> forwardExpirationEvents().
> > > > > > >>
> > > > > > >>
> > > > > > >> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io>
> > > > wrote:
> > > > > > >>
> > > > > > >> > +1 for being more explicit with the "And" conjunction
> > > > > > >> >
> > > > > > >> > -Kirk
> > > > > > >> >
> > > > > > >> >
> > > > > > >> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <
> > > abaker@pivotal.io
> > > > >
> > > > > > >> wrote:
> > > > > > >> >
> > > > > > >> >> I’d prefer to insert a conjunction to clarify the meaning:
> > > > > > >> >>
> > > > > > >> >> forwardEvictionAndExpirationEvents()
> > > > > > >> >>
> > > > > > >> >>
> > > > > > >> >> $0.02,
> > > > > > >> >> Anthony
> > > > > > >> >>
> > > > > > >> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <
> > > > > agingade@pivotal.io>
> > > > > > >> >> wrote:
> > > > > > >> >>
> > > > > > >> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any
> > > > alternate
> > > > > > >> >> names?).
> > > > > > >> >>
> > > > > > >> >>
> > > > > > >> >>
> > > > > > >> >
> > > > > > >>
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
--
~/William
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Posted by Dan Smith <ds...@pivotal.io>.
+1 for splitting it up and getting rid of the Eviction option for now.
I think I was the one who suggested setIgnoreExpiration originally.
Ignore/Forward, I guess it's not really a big deal. But since you are
setting a property on the queue, I think it should be setForwardExpiration,
not just forwardExpiration. I believe the current property is setIgnoreXXXX.
-Dan
On Fri, Jun 3, 2016 at 12:25 PM, Anilkumar Gingade <ag...@pivotal.io>
wrote:
> Yes...Jason, thanks for correction...
>
> [corrected]
> I will wait for other feedbacks/comments, if there is no objection i
> will change
> it to "forwardExpirationDestroy" with default value set to false.
>
> -Anil.
>
>
>
> On Fri, Jun 3, 2016 at 12:01 PM, Jason Huynh <jh...@pivotal.io> wrote:
>
> > *I will wait for other feedbacks/comments, if there is no objection i
> > willchange it to "forwardExpirationDestroy" with default value set to
> > true.*
> >
> > think you meant to say default value to set false :-)
> >
> > On Fri, Jun 3, 2016 at 11:47 AM, Anilkumar Gingade <ag...@pivotal.io>
> > wrote:
> >
> > > Thanks Darrel...
> > >
> > > The name change was based on the review comments:
> > > https://reviews.apache.org/r/46243/
> > >
> > > I am fine with either one, ignoreXXX or forwardXXX.
> > >
> > > I will wait for other feedbacks/comments, if there is no objection i
> will
> > > change it to "forwardExpirationDestroy" with default value set to true.
> > >
> > > Thanks,
> > > -Anil.
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > >
> > > On Thu, Jun 2, 2016 at 2:14 PM, Darrel Schneider <
> dschneider@pivotal.io>
> > > wrote:
> > >
> > > > When did forwardXXX become ignoreXXX? I read through the email thread
> > and
> > > > couldn't find why that happened. It is best for the default on a
> > boolean
> > > > property to be false. That was the case when it was forwardXXX. But
> now
> > > > that it has changed to ignoreXXX the default has become true. I'd
> vote
> > > for
> > > > it being named something whose default can be false.
> > > >
> > > >
> > > > On Thu, Jun 2, 2016 at 11:10 AM, Anilkumar Gingade <
> > agingade@pivotal.io>
> > > > wrote:
> > > >
> > > > > Hi Team,
> > > > >
> > > > > As proposed here, we added support to propagate eviction and
> > expiration
> > > > > (destroy) operation to AsyncEventQueue using single flag/attribute
> > > > > "ignoreEvictionAndExpiration" by default which is true (to keep the
> > > same
> > > > > behavior) and one could set (false) to receive eviction/expiration
> > > > event...
> > > > >
> > > > > But we come across a product issue, GEODE-1472, that cause data
> > > > > inconsistency (with eviction destroy)....For this reason we are
> > > planning
> > > > to
> > > > > break the "ignoreEvictionAndExpiration" attribute to eviction and
> > > > > expiration specific:
> > > > > "ignoreEvictionDestroy", "ignoreExpirationDestroy"...
> > > > >
> > > > > Currently we are planning to support "ignoreExpirationDestroy",
> and
> > > add
> > > > > "ignoreEvictionDestroy" once GEODE-1472 is fixed...
> > > > >
> > > > > Looking for comments on this...
> > > > >
> > > > > Thanks,
> > > > > -Anil.
> > > > >
> > > > > https://issues.apache.org/jira/browse/GEODE-1472
> > > > >
> > > > >
> > > > >
> > > > >
> > > > > On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <
> > > agingade@pivotal.io>
> > > > > wrote:
> > > > >
> > > > > > Kirk, We could not think of any such requirement...And with this
> > > > > > application will get all the update operation and can take
> > > appropriate
> > > > > > action (use or ignore)...
> > > > > >
> > > > > > -Anil.
> > > > > >
> > > > > >
> > > > > >
> > > > > > On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io>
> > wrote:
> > > > > >
> > > > > >> Would any user ever have a reason to enable forwarding of one
> type
> > > but
> > > > > not
> > > > > >> the other? If so then I would separate them as
> > > forwardEvictionEvents()
> > > > > and
> > > > > >> forwardExpirationEvents().
> > > > > >>
> > > > > >>
> > > > > >> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io>
> > > wrote:
> > > > > >>
> > > > > >> > +1 for being more explicit with the "And" conjunction
> > > > > >> >
> > > > > >> > -Kirk
> > > > > >> >
> > > > > >> >
> > > > > >> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <
> > abaker@pivotal.io
> > > >
> > > > > >> wrote:
> > > > > >> >
> > > > > >> >> I’d prefer to insert a conjunction to clarify the meaning:
> > > > > >> >>
> > > > > >> >> forwardEvictionAndExpirationEvents()
> > > > > >> >>
> > > > > >> >>
> > > > > >> >> $0.02,
> > > > > >> >> Anthony
> > > > > >> >>
> > > > > >> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <
> > > > agingade@pivotal.io>
> > > > > >> >> wrote:
> > > > > >> >>
> > > > > >> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any
> > > alternate
> > > > > >> >> names?).
> > > > > >> >>
> > > > > >> >>
> > > > > >> >>
> > > > > >> >
> > > > > >>
> > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
>
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Posted by Anilkumar Gingade <ag...@pivotal.io>.
Yes...Jason, thanks for correction...
[corrected]
I will wait for other feedbacks/comments, if there is no objection i
will change
it to "forwardExpirationDestroy" with default value set to false.
-Anil.
On Fri, Jun 3, 2016 at 12:01 PM, Jason Huynh <jh...@pivotal.io> wrote:
> *I will wait for other feedbacks/comments, if there is no objection i
> willchange it to "forwardExpirationDestroy" with default value set to
> true.*
>
> think you meant to say default value to set false :-)
>
> On Fri, Jun 3, 2016 at 11:47 AM, Anilkumar Gingade <ag...@pivotal.io>
> wrote:
>
> > Thanks Darrel...
> >
> > The name change was based on the review comments:
> > https://reviews.apache.org/r/46243/
> >
> > I am fine with either one, ignoreXXX or forwardXXX.
> >
> > I will wait for other feedbacks/comments, if there is no objection i will
> > change it to "forwardExpirationDestroy" with default value set to true.
> >
> > Thanks,
> > -Anil.
> >
> >
> >
> >
> >
> >
> >
> >
> > On Thu, Jun 2, 2016 at 2:14 PM, Darrel Schneider <ds...@pivotal.io>
> > wrote:
> >
> > > When did forwardXXX become ignoreXXX? I read through the email thread
> and
> > > couldn't find why that happened. It is best for the default on a
> boolean
> > > property to be false. That was the case when it was forwardXXX. But now
> > > that it has changed to ignoreXXX the default has become true. I'd vote
> > for
> > > it being named something whose default can be false.
> > >
> > >
> > > On Thu, Jun 2, 2016 at 11:10 AM, Anilkumar Gingade <
> agingade@pivotal.io>
> > > wrote:
> > >
> > > > Hi Team,
> > > >
> > > > As proposed here, we added support to propagate eviction and
> expiration
> > > > (destroy) operation to AsyncEventQueue using single flag/attribute
> > > > "ignoreEvictionAndExpiration" by default which is true (to keep the
> > same
> > > > behavior) and one could set (false) to receive eviction/expiration
> > > event...
> > > >
> > > > But we come across a product issue, GEODE-1472, that cause data
> > > > inconsistency (with eviction destroy)....For this reason we are
> > planning
> > > to
> > > > break the "ignoreEvictionAndExpiration" attribute to eviction and
> > > > expiration specific:
> > > > "ignoreEvictionDestroy", "ignoreExpirationDestroy"...
> > > >
> > > > Currently we are planning to support "ignoreExpirationDestroy", and
> > add
> > > > "ignoreEvictionDestroy" once GEODE-1472 is fixed...
> > > >
> > > > Looking for comments on this...
> > > >
> > > > Thanks,
> > > > -Anil.
> > > >
> > > > https://issues.apache.org/jira/browse/GEODE-1472
> > > >
> > > >
> > > >
> > > >
> > > > On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <
> > agingade@pivotal.io>
> > > > wrote:
> > > >
> > > > > Kirk, We could not think of any such requirement...And with this
> > > > > application will get all the update operation and can take
> > appropriate
> > > > > action (use or ignore)...
> > > > >
> > > > > -Anil.
> > > > >
> > > > >
> > > > >
> > > > > On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io>
> wrote:
> > > > >
> > > > >> Would any user ever have a reason to enable forwarding of one type
> > but
> > > > not
> > > > >> the other? If so then I would separate them as
> > forwardEvictionEvents()
> > > > and
> > > > >> forwardExpirationEvents().
> > > > >>
> > > > >>
> > > > >> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io>
> > wrote:
> > > > >>
> > > > >> > +1 for being more explicit with the "And" conjunction
> > > > >> >
> > > > >> > -Kirk
> > > > >> >
> > > > >> >
> > > > >> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <
> abaker@pivotal.io
> > >
> > > > >> wrote:
> > > > >> >
> > > > >> >> I’d prefer to insert a conjunction to clarify the meaning:
> > > > >> >>
> > > > >> >> forwardEvictionAndExpirationEvents()
> > > > >> >>
> > > > >> >>
> > > > >> >> $0.02,
> > > > >> >> Anthony
> > > > >> >>
> > > > >> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <
> > > agingade@pivotal.io>
> > > > >> >> wrote:
> > > > >> >>
> > > > >> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any
> > alternate
> > > > >> >> names?).
> > > > >> >>
> > > > >> >>
> > > > >> >>
> > > > >> >
> > > > >>
> > > > >
> > > > >
> > > >
> > >
> >
>
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Posted by Jason Huynh <jh...@pivotal.io>.
*I will wait for other feedbacks/comments, if there is no objection i
willchange it to "forwardExpirationDestroy" with default value set to true.*
think you meant to say default value to set false :-)
On Fri, Jun 3, 2016 at 11:47 AM, Anilkumar Gingade <ag...@pivotal.io>
wrote:
> Thanks Darrel...
>
> The name change was based on the review comments:
> https://reviews.apache.org/r/46243/
>
> I am fine with either one, ignoreXXX or forwardXXX.
>
> I will wait for other feedbacks/comments, if there is no objection i will
> change it to "forwardExpirationDestroy" with default value set to true.
>
> Thanks,
> -Anil.
>
>
>
>
>
>
>
>
> On Thu, Jun 2, 2016 at 2:14 PM, Darrel Schneider <ds...@pivotal.io>
> wrote:
>
> > When did forwardXXX become ignoreXXX? I read through the email thread and
> > couldn't find why that happened. It is best for the default on a boolean
> > property to be false. That was the case when it was forwardXXX. But now
> > that it has changed to ignoreXXX the default has become true. I'd vote
> for
> > it being named something whose default can be false.
> >
> >
> > On Thu, Jun 2, 2016 at 11:10 AM, Anilkumar Gingade <ag...@pivotal.io>
> > wrote:
> >
> > > Hi Team,
> > >
> > > As proposed here, we added support to propagate eviction and expiration
> > > (destroy) operation to AsyncEventQueue using single flag/attribute
> > > "ignoreEvictionAndExpiration" by default which is true (to keep the
> same
> > > behavior) and one could set (false) to receive eviction/expiration
> > event...
> > >
> > > But we come across a product issue, GEODE-1472, that cause data
> > > inconsistency (with eviction destroy)....For this reason we are
> planning
> > to
> > > break the "ignoreEvictionAndExpiration" attribute to eviction and
> > > expiration specific:
> > > "ignoreEvictionDestroy", "ignoreExpirationDestroy"...
> > >
> > > Currently we are planning to support "ignoreExpirationDestroy", and
> add
> > > "ignoreEvictionDestroy" once GEODE-1472 is fixed...
> > >
> > > Looking for comments on this...
> > >
> > > Thanks,
> > > -Anil.
> > >
> > > https://issues.apache.org/jira/browse/GEODE-1472
> > >
> > >
> > >
> > >
> > > On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <
> agingade@pivotal.io>
> > > wrote:
> > >
> > > > Kirk, We could not think of any such requirement...And with this
> > > > application will get all the update operation and can take
> appropriate
> > > > action (use or ignore)...
> > > >
> > > > -Anil.
> > > >
> > > >
> > > >
> > > > On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io> wrote:
> > > >
> > > >> Would any user ever have a reason to enable forwarding of one type
> but
> > > not
> > > >> the other? If so then I would separate them as
> forwardEvictionEvents()
> > > and
> > > >> forwardExpirationEvents().
> > > >>
> > > >>
> > > >> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io>
> wrote:
> > > >>
> > > >> > +1 for being more explicit with the "And" conjunction
> > > >> >
> > > >> > -Kirk
> > > >> >
> > > >> >
> > > >> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <abaker@pivotal.io
> >
> > > >> wrote:
> > > >> >
> > > >> >> I’d prefer to insert a conjunction to clarify the meaning:
> > > >> >>
> > > >> >> forwardEvictionAndExpirationEvents()
> > > >> >>
> > > >> >>
> > > >> >> $0.02,
> > > >> >> Anthony
> > > >> >>
> > > >> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <
> > agingade@pivotal.io>
> > > >> >> wrote:
> > > >> >>
> > > >> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any
> alternate
> > > >> >> names?).
> > > >> >>
> > > >> >>
> > > >> >>
> > > >> >
> > > >>
> > > >
> > > >
> > >
> >
>
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Posted by Anilkumar Gingade <ag...@pivotal.io>.
Thanks Darrel...
The name change was based on the review comments:
https://reviews.apache.org/r/46243/
I am fine with either one, ignoreXXX or forwardXXX.
I will wait for other feedbacks/comments, if there is no objection i will
change it to "forwardExpirationDestroy" with default value set to true.
Thanks,
-Anil.
On Thu, Jun 2, 2016 at 2:14 PM, Darrel Schneider <ds...@pivotal.io>
wrote:
> When did forwardXXX become ignoreXXX? I read through the email thread and
> couldn't find why that happened. It is best for the default on a boolean
> property to be false. That was the case when it was forwardXXX. But now
> that it has changed to ignoreXXX the default has become true. I'd vote for
> it being named something whose default can be false.
>
>
> On Thu, Jun 2, 2016 at 11:10 AM, Anilkumar Gingade <ag...@pivotal.io>
> wrote:
>
> > Hi Team,
> >
> > As proposed here, we added support to propagate eviction and expiration
> > (destroy) operation to AsyncEventQueue using single flag/attribute
> > "ignoreEvictionAndExpiration" by default which is true (to keep the same
> > behavior) and one could set (false) to receive eviction/expiration
> event...
> >
> > But we come across a product issue, GEODE-1472, that cause data
> > inconsistency (with eviction destroy)....For this reason we are planning
> to
> > break the "ignoreEvictionAndExpiration" attribute to eviction and
> > expiration specific:
> > "ignoreEvictionDestroy", "ignoreExpirationDestroy"...
> >
> > Currently we are planning to support "ignoreExpirationDestroy", and add
> > "ignoreEvictionDestroy" once GEODE-1472 is fixed...
> >
> > Looking for comments on this...
> >
> > Thanks,
> > -Anil.
> >
> > https://issues.apache.org/jira/browse/GEODE-1472
> >
> >
> >
> >
> > On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <ag...@pivotal.io>
> > wrote:
> >
> > > Kirk, We could not think of any such requirement...And with this
> > > application will get all the update operation and can take appropriate
> > > action (use or ignore)...
> > >
> > > -Anil.
> > >
> > >
> > >
> > > On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io> wrote:
> > >
> > >> Would any user ever have a reason to enable forwarding of one type but
> > not
> > >> the other? If so then I would separate them as forwardEvictionEvents()
> > and
> > >> forwardExpirationEvents().
> > >>
> > >>
> > >> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io> wrote:
> > >>
> > >> > +1 for being more explicit with the "And" conjunction
> > >> >
> > >> > -Kirk
> > >> >
> > >> >
> > >> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <ab...@pivotal.io>
> > >> wrote:
> > >> >
> > >> >> I’d prefer to insert a conjunction to clarify the meaning:
> > >> >>
> > >> >> forwardEvictionAndExpirationEvents()
> > >> >>
> > >> >>
> > >> >> $0.02,
> > >> >> Anthony
> > >> >>
> > >> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <
> agingade@pivotal.io>
> > >> >> wrote:
> > >> >>
> > >> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any alternate
> > >> >> names?).
> > >> >>
> > >> >>
> > >> >>
> > >> >
> > >>
> > >
> > >
> >
>
Re: Proposal to allow eviction and expiration operations/events with AsyncEventQueue.
Posted by Darrel Schneider <ds...@pivotal.io>.
When did forwardXXX become ignoreXXX? I read through the email thread and
couldn't find why that happened. It is best for the default on a boolean
property to be false. That was the case when it was forwardXXX. But now
that it has changed to ignoreXXX the default has become true. I'd vote for
it being named something whose default can be false.
On Thu, Jun 2, 2016 at 11:10 AM, Anilkumar Gingade <ag...@pivotal.io>
wrote:
> Hi Team,
>
> As proposed here, we added support to propagate eviction and expiration
> (destroy) operation to AsyncEventQueue using single flag/attribute
> "ignoreEvictionAndExpiration" by default which is true (to keep the same
> behavior) and one could set (false) to receive eviction/expiration event...
>
> But we come across a product issue, GEODE-1472, that cause data
> inconsistency (with eviction destroy)....For this reason we are planning to
> break the "ignoreEvictionAndExpiration" attribute to eviction and
> expiration specific:
> "ignoreEvictionDestroy", "ignoreExpirationDestroy"...
>
> Currently we are planning to support "ignoreExpirationDestroy", and add
> "ignoreEvictionDestroy" once GEODE-1472 is fixed...
>
> Looking for comments on this...
>
> Thanks,
> -Anil.
>
> https://issues.apache.org/jira/browse/GEODE-1472
>
>
>
>
> On Tue, Apr 12, 2016 at 6:04 PM, Anilkumar Gingade <ag...@pivotal.io>
> wrote:
>
> > Kirk, We could not think of any such requirement...And with this
> > application will get all the update operation and can take appropriate
> > action (use or ignore)...
> >
> > -Anil.
> >
> >
> >
> > On Tue, Apr 12, 2016 at 5:46 PM, Kirk Lund <kl...@pivotal.io> wrote:
> >
> >> Would any user ever have a reason to enable forwarding of one type but
> not
> >> the other? If so then I would separate them as forwardEvictionEvents()
> and
> >> forwardExpirationEvents().
> >>
> >>
> >> On Tue, Apr 12, 2016 at 5:44 PM, Kirk Lund <kl...@pivotal.io> wrote:
> >>
> >> > +1 for being more explicit with the "And" conjunction
> >> >
> >> > -Kirk
> >> >
> >> >
> >> > On Tue, Apr 12, 2016 at 5:25 PM, Anthony Baker <ab...@pivotal.io>
> >> wrote:
> >> >
> >> >> I’d prefer to insert a conjunction to clarify the meaning:
> >> >>
> >> >> forwardEvictionAndExpirationEvents()
> >> >>
> >> >>
> >> >> $0.02,
> >> >> Anthony
> >> >>
> >> >> On Apr 12, 2016, at 5:11 PM, Anilkumar Gingade <ag...@pivotal.io>
> >> >> wrote:
> >> >>
> >> >> *New attribute:* "forwardEvictionExpirationEvents()" (Any alternate
> >> >> names?).
> >> >>
> >> >>
> >> >>
> >> >
> >>
> >
> >
>