[jbosscache-dev] State transfer tests in head

Manik Surtani manik at jboss.org
Tue Oct 31 18:17:16 EST 2006


This was actually from a series of IM and phone conversations I had  
with Brian around RegionManagers and Marshalling/Eviction Regions.

In the 1.x codebase, marshalling regions would queue calls made to  
regions marked as inactive, and then apply these calls when the  
region was activated.  Sort of a way to deal with calls that may come  
in to a cache before the cache has completed transferring state and  
fully joins the cluster.  We felt that with FLUSH in place, we'd have  
a better solution.  If this otherwise edge case starts becoming a  
concern, enabling FLUSH would be the correct solution.


--
Manik Surtani

Lead, JBoss Cache
JBoss, a division of Red Hat

Email: manik at jboss.org
Telephone: +44 7786 702 706
MSN: manik at surtani.org
Yahoo/AIM/Skype: maniksurtani


On 31 Oct 2006, at 17:12, Vladimir Blagojevic wrote:

> Ok thanks Brian. I do not remember "replacing queueing with FLUSH"  
> but I
> was not focused on that back then. I'll look.
>
>> -----Original Message-----
>> From: Brian Stansberry
>> Sent: Tuesday, October 31, 2006 12:08 PM
>> To: Vladimir Blagojevic; Manik Surtani
>> Subject: RE: [jbosscache-dev] State transfer tests in head
>>
>> If you look through e-mails I sent a couple months back when
>> we were attacking the siemens issue, there was one that had a
>> bunch of forum links.
>>
>> Vladimir Blagojevic wrote:
>>> Hey guys,
>>>
>>> I have this task on my plate now that Jgroups has been cleared up a
>>> bit. I also have to implemented partial state transfer as jgroups
>>> getState call rather than rpc invocation.
>>> I have already started coding on partial state transfer. Task that
>>> Manik mentions below is very much related since
>>> TreeCache.activateRegion is initiating partial state transfer.
>>>
>>> Would one of you, when you get a chance, give me a bit more
>> background
>>> pointers about .
>>> Anything will do: forum references, JIRA etc etc.
>>>
>>> Vladimir
>>>
>>>
>>>> -----Original Message-----
>>>> From: jbosscache-dev-bounces at lists.jboss.org
>>>> [mailto:jbosscache-dev-bounces at lists.jboss.org] On Behalf Of Manik
>>>> Surtani Sent: Friday, October 13, 2006 8:38 AM
>>>> To: jbosscache-dev at lists.jboss.org
>>>> Subject: [jbosscache-dev] State transfer tests in head
>>>>
>>>> Hi,
>>>>
>>>> As a part of refactoring the marshall.Region class to a generic
>>>> Region which can be used for eviction as well, Brian and I
>> discussed
>>>> removing queueing of calls to inactive regions, since this
>> would be
>>>> replaced with FLUSH.  As such, I've commented out a few unit tests
>>>> that test this queueing in VersionedTestsBase and have commented
>>>> accordingly. Vladimir, once we roll in FLUSH and have this
>> working as
>>>> expected, could you revisit these tests in state transfer?
>>>>
>>>> Thanks,
>>>> --
>>>> Manik Surtani
>>>>
>>>> Lead, JBoss Cache
>>>> JBoss, a division of Red Hat
>>>>
>>>> Email: manik at jboss.org
>>>> Telephone: +44 7786 702 706
>>>> MSN: manik at surtani.org
>>>> Yahoo/AIM/Skype: maniksurtani
>>>>
>>>>
>>>> _______________________________________________
>>>> jbosscache-dev mailing list
>>>> jbosscache-dev at lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/jbosscache-dev
>>
>>
>>
>> Brian Stansberry
>> Lead, AS Clustering
>> JBoss, a division of Red Hat
>> Ph: 510-396-3864
>> skype: bstansberry




More information about the jbosscache-dev mailing list