On Thu, 16 May 2002, Jeremy Howard wrote: > I'm not 100% sure, but I think the change you've made to return if the > centry isn't found may mean the nconnections doesn't get updated > correctly in some cases. You probably want to update nconnections in the > case of the SERVICE_CONNECTION message regardless of whether you find > the centry.
nconnections won't get updated by messages from unknown children. This should not ever happen, and nconnections is only used for snmp accounting anyway so there are no weird side-effects. Still, I am seriously thinking about doing something to any unknown children master notices (with proper LOG_ERR logging, of course) because said children is messaging it. Such a thing (unknown children) can only happen if either the underlying OS is messed up, or if we have a bug or memory corruption in the linked list, AFAIK. Now, the question is: what to do? Send a TERM signal to the child? Add the child to the services list (where?)? Leave it alone (what we do now)? Send a STOP signal to keep it frozen for latter debugging? I would like to add a "Ominous events" counter to the SNMP MIB, which would be increased by all weirdness master notices (children dying in available state, children dying with non-OK exit status or due to signals, unknown messages or messages from unknown children, watchdog timeouts, etc). What do you people think about this idea? Obviously, it will only be done if CMU likes the idea, since the MIB is theirs... -- "One disk to rule them all, One disk to find them. One disk to bring them all and in the darkness grind them. In the Land of Redmond where the shadows lie." -- The Silicon Valley Tarot Henrique Holschuh
