I'm not so sure about the conclusion that because the events of interest are local you can safely use userData. IIUC under heavy load local events may be aggregated to mitigate the load and in that process 'localness' is lost...
Can anyone more knowledgeable address that point? Thanks, -Rob -----Original Message----- From: Tommaso Teofili [mailto:[email protected]] Sent: Tuesday, October 07, 2014 2:32 AM To: [email protected] Subject: Re: ObservationManager's userData in Oak Hi Michael, thanks for the information, despite the given limitations I think it could still work for my use case, as the distinction I need to make by means of userData should always be on local events. Regards, Tommaso 2014-10-03 17:10 GMT+02:00 Michael Dürig <[email protected]>: > > Hi Tommaso, > > Unfortunately Oak has only limited support for user data as it cannot > preserve commit boundaries in all cases (e.g. commits from other > cluster nodes). See also > http://jackrabbit.apache.org/oak/docs/differences.html# > Observation > > Michael > > > > On 3.10.14 5:00 , Tommaso Teofili wrote: > >> Hi all, >> >> in the context of Sling [1][2] I am evaluating the possibility of >> using the ObservationManager#setUserData [3] as a way to signal which >> Sessions generated certain events in order to filter some having >> specific userData. >> >> I don't have the whole background about how userData is supported in >> Oak, could someone give some information? >> More specifically: is userData always guaranteed to be provided on >> the event if set by the related Session? >> >> Thanks in advance for any help, >> Tommaso >> >> [1] : https://issues.apache.org/jira/browse/SLING-3993 >> [2] : https://issues.apache.org/jira/browse/SLING-3983 >> [3] : >> http://www.day.com/maven/jsr170/javadocs/jcr-2.0/javax/jcr/observatio >> n/ >> ObservationManager.html#setUserData(java.lang.String) >> >>
