On 24-Apr-15 2:34 PM, Yogesh Kumar wrote: > Hi OpenSaf Team, > > I have some queries regarding the configuration . > > > 1) The configuration of SG/SU/chassis blade done via modelling ,can we > change this configuration at run time. > For e.g. > Let's assume we have a 6 slot chassis where currently I have two controller > nodes and 4 payload nodes .The application runs on payload 1 and payload 2 > Is it possible I remove the component from payload 1 and start running the > application on payload 3.To perform this whether changes can be done in xml > at run time > How does cluster update the xml and apply it to all nodes. Whether > nodes needs to be restarted or what are the impacts and ways to do this. > This can be done by dynamically creating a SU3 (in lock-in state) and a component on payload 3 and then unlock-in this SU. After this SU on payload 1 can be removed on SU1 in locked-in state. After this command "immdump <file name>" will give the updated system configuration.
Note: Components in SU3 should have same capability as in SU1, so that when Su1 is locked for deletion, SU3 will become assigned for the same service which SU1 was serving. > > 2) Let's assume we have only 4 blades 2 controllers two payload nodes > and the application is running on the payload nodes .In future I add two more > blades to this 6 slot chassis and I need to run separate applications(than > the one currently running) into newly added blades .For this we might have to > create new SG/SU etc. .Can this be done without affecting the running cluster > and whether the changes in xml will be enough to support this. > For this in the beginning also imm.xml can be prepared with 2 con and 5 payloads. Dynamicallly also AMF and CLM nodes can be created. And after creating nodes, application also be hosted on the newly created nodes. > > 3) The application component in case needs to report some fault > conditions to AMF ,how is it achieved .There can be a case where some > critical condition has happened at application and we need to inform AMF for > doing switchover, is there some notification mechanism provided .Is there a > sample example which we can refer for the same .I assume even HPI faults > raised via platform needs to be filtered and sent to AMF to take corrective > actions, whether same functionality can be done at application level and how > is it done. > Is it controller switchover or the intention is switchover of application services from that node? In AMF, saAmfComponentErrorReport_() can be used by a component to reprot some error to AMF. This API has "recovery policy" as one of the argument. When this API is invoked the the component or any other monitoring application then AMF declares the component faulty and take corrective action. Thanks PRaveen > > Best Regards, > Yogesh Kumar > ------------------------------------------------------------------------------ > One dashboard for servers and applications across Physical-Virtual-Cloud > Widest out-of-the-box monitoring support with 50+ applications > Performance metrics, stats and reports that give you Actionable Insights > Deep dive visibility with transaction tracing using APM Insight. > http://ad.doubleclick.net/ddm/clk/290420510;117567292;y > _______________________________________________ > Opensaf-devel mailing list > Opensaf-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/opensaf-devel > ------------------------------------------------------------------------------ One dashboard for servers and applications across Physical-Virtual-Cloud Widest out-of-the-box monitoring support with 50+ applications Performance metrics, stats and reports that give you Actionable Insights Deep dive visibility with transaction tracing using APM Insight. http://ad.doubleclick.net/ddm/clk/290420510;117567292;y _______________________________________________ Opensaf-devel mailing list Opensaf-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/opensaf-devel
