synapse-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "indika priyantha kumara (JIRA)" <j...@apache.org>
Subject [jira] Reopened: (SYNAPSE-490) Add a Field to the AlgorithmContext to Support Storing Algorithm Impl Specific Data
Date Tue, 23 Dec 2008 07:19:44 GMT

     [ https://issues.apache.org/jira/browse/SYNAPSE-490?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

indika priyantha kumara reopened SYNAPSE-490:
---------------------------------------------


Hi Hiranya 

Algorithm context is not for pass states (It was a mistake). Passing states only applicable
when there are a chain of responsibility for handling states and these states are independent
from algorithms. Purpose of the algorithm context is to hide the fact that states are kept
locally if synapse is in a non-cluster and states are kept in axis2 configuration context
for replication in a clustered environment. Because of that any algorithm implementer (not
user of algorithm) can use it to hold states without doing any thing to differentiate cluttered
states and non clustered states. Therefore, the way you have implemented getParameter and
setParameter is wrong. It MUST care differentiation between clustered and none clustered env.
Look at how other states are kept 

Thanks
Indika

> Add a Field to the AlgorithmContext to Support Storing Algorithm Impl Specific Data
> -----------------------------------------------------------------------------------
>
>                 Key: SYNAPSE-490
>                 URL: https://issues.apache.org/jira/browse/SYNAPSE-490
>             Project: Synapse
>          Issue Type: Improvement
>          Components: Endpoints
>    Affects Versions: NIGHTLY
>         Environment: Any
>            Reporter: Hiranya Jayathilaka
>            Assignee: Hiranya Jayathilaka
>            Priority: Minor
>             Fix For: NIGHTLY
>
>
> Currently AlgorithmContext can be used to store only endpoint index values. However this
may not be sufficient for custom load balance algorithm implementations. Such implementations
may need to store other impl specific data items. Therefore AlgorithmContext should have a
map of parameters and get-set methods to manipulate the parameters.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@synapse.apache.org
For additional commands, e-mail: dev-help@synapse.apache.org


Mime
View raw message