[teiid-issues] [JBoss JIRA] Commented: (TEIID-616) Start time of the system could be incorrect in rare cases where processes are brought down and back up

[ https://jira.jboss.org/jira/browse/TEIID-616?page=com.atlassian.jira.plug... ] Steven Hawkins commented on TEIID-616: -------------------------------------- The question is what is system start time supposed to represent. The point of my first comment is that the previous notion of start time was tied to configuration state. The concern that Van is expressing is from a user perspective that the current approach may seem to understate our old notion of system start time in a rolling restart scenario where the configuration hasn't changed. Without considering configuration state you could track: eldest process start time as Larry suggests, propagate the original start time you could even use use the eldest or propagate the start time from all vms including host controllers Any one of these could be considered a correct system start time - and they could all be different than what we tracked previously. The next thing to note is that none of these actually capture if the cluster is viable (such as a queryservice or other essential service failure), just when it was started. Using it as a measure of availability would also be misleading. So the broader questions are what do we want the system start time to be and why is it meaningful? -- This message is automatically generated by JIRA. - If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa - For more information on JIRA, see: http://www.atlassian.com/software/jira