RE: http://jira.jboss.com/jira/browse/JBCACHE-847
by Ben Wang
Yes, I am hoping to include it in 1.4. It is corner case but it is an important one, IMO. Furthermore, it doesn't look like that difficult to fix so I will try to get it done tomorrow.
Cheers,
-Ben
-----Original Message-----
From: Manik Surtani [mailto:manik@jboss.org]
Sent: Wednesday, November 15, 2006 10:35 PM
To: Ben Wang
Cc: jbosscache-dev(a)lists.jboss.org
Subject: http://jira.jboss.com/jira/browse/JBCACHE-847
Ben,
Does this need to be in 1.4.1 or can it be deferred? I'm hoping to tag 1.4.1.CR1 over the weekend before I fly out to Berlin.
Cheers,
--
Manik Surtani
Lead, JBoss Cache
JBoss, a division of Red Hat
Email: manik(a)jboss.org
Telephone: +44 7786 702 706
MSN: manik(a)surtani.org
Yahoo/AIM/Skype: maniksurtani
18 years, 3 months
http://jira.jboss.com/jira/browse/JBCACHE-847
by Manik Surtani
Ben,
Does this need to be in 1.4.1 or can it be deferred? I'm hoping to
tag 1.4.1.CR1 over the weekend before I fly out to Berlin.
Cheers,
--
Manik Surtani
Lead, JBoss Cache
JBoss, a division of Red Hat
Email: manik(a)jboss.org
Telephone: +44 7786 702 706
MSN: manik(a)surtani.org
Yahoo/AIM/Skype: maniksurtani
18 years, 3 months
RE: [jbosscache-dev] Eviction configuration stuff
by Brian Stansberry
jbosscache-dev-bounces(a)lists.jboss.org wrote:
> Brian,
>
> Just trying to understand the eviction config pojos in HEAD,
>
> I see a o.j.c.c.EvictionConfig, which I presume is a
> top-level configuration for evictions. Within this, I see a
> collection of o.j.c.c.EvictionRegionConfig which is a
> regionalised set of configs, and each one of these has a
> reference to o.j.c.e.EvictionPolicyConfig which is
> policy-specific. I also see a o.j.c.e.EvictionConfigurator interface.
>
> 1) Is there a reason why EvictionPolicyConfig is in o.j.c.e
> instead of o.j.c.c along with the other configuration classes?
Just because some of this kind of stuff was in o.j.c.e before. Moving
them makes more sense. I think I ran out of energy at that point. :-(
> 2) Can we lose o.j.c.e.EvictionConfigurator? Nothing
> implements this interface and it is deprecated.
>
It's just there for backward compatibility with any old custom eviction
policies people have written. I'm perfectly fine with getting rid of it;
it's ugly.
- Brian
18 years, 3 months
Eviction configuration stuff
by Manik Surtani
Brian,
Just trying to understand the eviction config pojos in HEAD,
I see a o.j.c.c.EvictionConfig, which I presume is a top-level
configuration for evictions. Within this, I see a collection of
o.j.c.c.EvictionRegionConfig which is a regionalised set of configs,
and each one of these has a reference to o.j.c.e.EvictionPolicyConfig
which is policy-specific. I also see a o.j.c.e.EvictionConfigurator
interface.
1) Is there a reason why EvictionPolicyConfig is in o.j.c.e instead
of o.j.c.c along with the other configuration classes?
2) Can we lose o.j.c.e.EvictionConfigurator? Nothing implements
this interface and it is deprecated.
Cheers,
--
Manik Surtani
Lead, JBoss Cache
JBoss, a division of Red Hat
Email: manik(a)jboss.org
Telephone: +44 7786 702 706
MSN: manik(a)surtani.org
Yahoo/AIM/Skype: maniksurtani
18 years, 3 months
RE: [jbosscache-dev] 2.0.0 ALPHA and release sched
by Brian Stansberry
jbosscache-dev-bounces(a)lists.jboss.org wrote:
> Hi,
>
> I hope to tag 2.0.0.ALPHA on Thu evening EST.
>
> Could we make sure that all non-alpha changes do not go into
> HEAD between now and then?
>
> Vlad, I believe you still need to check in your FLUSH and
> state transfer stuff.
Yes, please do this ASAP. I need it to test http session replication.
> Brian, is there anything specific that you've
> still got to do?
The only thing I'd probably do is any tidying up related to the JMX
integration stuff.
>
> Ben, if there is anything specific on CC that is failing wrt.
> PojoCache, could you look into these?
>
> I will focus on the current breakage (due to my merging the
> eviction Region/RegionManagers with the generic
> Region/RegionManager constructs), I should have these
> finished by tomorrow though and can focus on any other breakages.
>
> The plan is that once we have this basic functionality in
> place, I can tag the alpha, push it through the release
> process and put it into the AS5 beta1 build.
>
> After that we can look at stabilising for a JBC beta1 by
> mid-Dec, performance tuning and refactoring for a beta2, and
> then however many CRs it takes to release 2.0.0 after that by Q1'07.
>
> Thoughts?
If we're going to integrate 2.0 in AS 4.2, they were shooting for a
2.0.0.GA at year end. I realize that's a tall order. IMHO, if its not
doable we can deal with it. Integrating 2.0.0 is better, but we'll be
supporting 4.2 for a long time, so I'd rather have a stable 1.4.1 than a
buggy, rushed 2.0.0.
18 years, 3 months
Void return from put(key, value)
by Brian Stansberry
Looks like Node and Cache have void returns for put(key, value) instead
of the returning the previous value. I'm going to assume that was
inadvertent and restore the old behavior; please let me know if I'm
wrong.
Brian Stansberry
Lead, AS Clustering
JBoss, a division of Red Hat
Ph: 510-396-3864
skype: bstansberry
IT executives: Red Hat still #1 for value
http://www.redhat.com/promo/vendor/
18 years, 3 months
RE: [jbosscache-dev] 2.0.0 ALPHA and release sched
by Vladimir Blagojevic
Hey Brian,
> > Vlad, I believe you still need to check in your FLUSH and state
> > transfer stuff.
>
> Yes, please do this ASAP. I need it to test http session replication.
Partial state transfer is in JBC HEAD now. On my local machine I did not
notice any effect on having channel blocking turned and having FLUSH in
stack. I ran all-functional-xknown with and without my changes and both
runs produced passing rate of 92%. I will continue to polish up this
stuff after Thursday. Brian, your review would be much appreciated sa
well.
18 years, 3 months
2.0.0 ALPHA and release sched
by Manik Surtani
Hi,
I hope to tag 2.0.0.ALPHA on Thu evening EST.
Could we make sure that all non-alpha changes do not go into HEAD
between now and then?
Vlad, I believe you still need to check in your FLUSH and state
transfer stuff. Brian, is there anything specific that you've still
got to do?
Ben, if there is anything specific on CC that is failing wrt.
PojoCache, could you look into these?
I will focus on the current breakage (due to my merging the eviction
Region/RegionManagers with the generic Region/RegionManager
constructs), I should have these finished by tomorrow though and can
focus on any other breakages.
The plan is that once we have this basic functionality in place, I
can tag the alpha, push it through the release process and put it
into the AS5 beta1 build.
After that we can look at stabilising for a JBC beta1 by mid-Dec,
performance tuning and refactoring for a beta2, and then however many
CRs it takes to release 2.0.0 after that by Q1'07.
Thoughts?
--
Manik Surtani
Lead, JBoss Cache
JBoss, a division of Red Hat
Email: manik(a)jboss.org
Telephone: +44 7786 702 706
MSN: manik(a)surtani.org
Yahoo/AIM/Skype: maniksurtani
18 years, 3 months
Re: HEAD
by Manik Surtani
Real soon. Thu evening. :-)
--
Manik Surtani
Lead, JBoss Cache
JBoss, a division of Red Hat
Email: manik(a)jboss.org
Telephone: +44 7786 702 706
MSN: manik(a)surtani.org
Yahoo/AIM/Skype: maniksurtani
On 14 Nov 2006, at 17:49, Dimitris Andreadis wrote:
> Ok, to conclude, we are expecting (really soon :) a tagged JBC Alpha
> release for JBoss5 Beta1, right?
>
>> -----Original Message-----
>> From: Manik Surtani [mailto:manik@jboss.org]
>> Sent: Tuesday, November 14, 2006 7:43 PM
>> To: Brian Stansberry
>> Cc: jbosscache-dev(a)lists.jboss.org; Dimitris Andreadis
>> Subject: Re: HEAD
>>
>>
>>
>>
>> On 14 Nov 2006, at 17:06, Brian Stansberry wrote:
>>
>>> Manik Surtani wrote:
>>>> Brian,
>>>>
>>>> I've started work on merging the Region/RegionManagers for
>> eviction.
>>>> The interfaces and API are now in place, but it breaks in a few
>>>> places, stuff that's still WIP.
>>>>
>>>> Hopefully at least you can go ahead with your integration, I will
>>>> have it all functional by tomorrow though. And then the rest of
>>>> tomorrow and thu to stabilise as much as I can before I release on
>>>> Fri.
>>>
>>> This should be OK. If there are delays past tomorrow, if you can
>>> please try to structure your approach to have either eviction or
>>> marshalling working so I can test either EJB3 (eviction) or FIELD
>>> session repl (marshalling). My preference would be to have the
>>> marshalling region stuff working earlier than eviction. Without
>>> marshalling I can't test the PojoCache integration, which is more
>>> likely to have complex issues that need time to resolve.
>>
>> Marshalling already works. It's just the eviction bits.
>>
>>>
>>>>
>>>> Do you think we can get away with a .DR2 or Alpha instead
>> of a Beta?
>>>> Because this certainly isn't Beta quality. :-)
>>>>
>>>
>>> We just discussed this on our weekly AS call. :-) Alpha should be
>>> fine (Dimitris, if I misunderstood, please correct me.)
>>
>> Wahey!
>>
>>>
>>> - Brian
>>
>>
>>
18 years, 3 months
RE: HEAD
by Brian Stansberry
Manik Surtani wrote:
> Brian,
>
> I've started work on merging the Region/RegionManagers for eviction.
> The interfaces and API are now in place, but it breaks in a
> few places, stuff that's still WIP.
>
> Hopefully at least you can go ahead with your integration, I
> will have it all functional by tomorrow though. And then the
> rest of tomorrow and thu to stabilise as much as I can before I
> release on Fri.
This should be OK. If there are delays past tomorrow, if you can please
try to structure your approach to have either eviction or marshalling
working so I can test either EJB3 (eviction) or FIELD session repl
(marshalling). My preference would be to have the marshalling region
stuff working earlier than eviction. Without marshalling I can't test
the PojoCache integration, which is more likely to have complex issues
that need time to resolve.
>
> Do you think we can get away with a .DR2 or Alpha instead of a Beta?
> Because this certainly isn't Beta quality. :-)
>
We just discussed this on our weekly AS call. :-) Alpha should be fine
(Dimitris, if I misunderstood, please correct me.)
- Brian
18 years, 3 months
