Hi Andras, You can also demo health check failure by causing a infinite loop in your application.
This AVA down fast detection is done for only the registered process of The component. So if you want to do training have a component with 2 processes One the registered process which doesn't do health check and the other process Should start health check. Then when the none registered process is killed The health check failure will only cause the recovery to happen. Regards Sugadeesh > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] On Behalf Of > Mathivanan Np-G19859 > Sent: Monday, November 05, 2007 4:11 PM > To: Hans Feldt; shankar reddy > Cc: [email protected] > Subject: Re: [Users] Component health checking in OpenSAF > > > The "AvA Down" message is originated from MDS when a > component(mds dest) goes down. > > Regards, > Mathi. > > > > -----Original Message----- > > From: [EMAIL PROTECTED] > > [mailto:[EMAIL PROTECTED] On Behalf Of Hans Feldt > > Sent: Monday, November 05, 2007 3:37 PM > > To: shankar reddy > > Cc: [email protected] > > Subject: Re: [Users] Component health checking in OpenSAF > > > > shankar reddy wrote: > > > Andras, > > > Yes. It is the expected behaviour in OpenSAF, because when the > > > application is killed then SCAP (AMF) process receives AVA > > down (Agent > > > Down) message. Then AMF restarts the component if it is the > > > restartable component. This another mechanism apart from > > HealthCheck. > > > > Is this "AVA Down" message originated from TIPC? > > > > Regards, > > Hans > > > > > > > > Thanks > > > Shankar > > > > > > > > > */Andras Kovi <[EMAIL PROTECTED]>/* wrote: > > > > > > > > > Hi All, > > > > > > I have created a simple application that runs on a > > controller node > > > and consists of only one component. The component healthchecks > > > interval is two seconds. Everything works fine, I get > the HCs in > > > time. But when I kill the component process directly by > > issuing a > > > pkill, OpenSAF restarts the component immediately as if > > there was > > > some other mean that monitored the process, other than > > the HC. Is > > > this true? As far as I know only passive monitoring is > > entitled for > > > such functionality in the AMF but I did not start > > anything like that. > > > > > > Ok, most probably this is a very good supportive > > functionality, but > > > it can cause some problems, if you are creating educative > > > demonstrations. > > > > > > Regards, > > > Andras > > > _______________________________________________ > > > Users mailing list > > > [email protected] > > > http://list.opensaf.org/maillist/listinfo/users > > > > > > > > > __________________________________________________ > > > Do You Yahoo!? > > > Tired of spam? Yahoo! Mail has the best spam protection around > > > http://mail.yahoo.com > > > > > > > > > > > > ---------------------------------------------------------------------- > > > -- > > > > > > _______________________________________________ > > > Users mailing list > > > [email protected] > > > http://list.opensaf.org/maillist/listinfo/users > > > > _______________________________________________ > > Users mailing list > > [email protected] > > http://list.opensaf.org/maillist/listinfo/users > > > _______________________________________________ > Users mailing list > [email protected] > http://list.opensaf.org/maillist/listinfo/users > _______________________________________________ Users mailing list [email protected] http://list.opensaf.org/maillist/listinfo/users
