Sounds good to me. Ismael
On Mon, Jul 24, 2017 at 6:46 PM, Colin McCabe <cmcc...@apache.org> wrote: > Hi all, > > I think maybe it makes sense to rename the "PreparingRebalance" consumer > group state to "Rebalancing." To me, "Preparing" implies that there > will be a later "rebalance" state that follows-- but there is not. > Since we're now exposing this state name publicly in these metrics, > perhaps it makes sense to do this rename now. Thoughts? > > best, > Colin > > > On Fri, Jul 21, 2017, at 13:52, Colin McCabe wrote: > > That's a good point. I revised the KIP to add metrics for all the group > > states. > > > > best, > > Colin > > > > > > On Fri, Jul 21, 2017, at 12:08, Guozhang Wang wrote: > > > Ah, that's right Jason. > > > > > > With that I can be convinced to add one metric per each state. > > > > > > Guozhang > > > > > > On Fri, Jul 21, 2017 at 11:44 AM, Jason Gustafson <ja...@confluent.io> > > > wrote: > > > > > > > > > > > > > "Dead" and "Empty" states are transient: groups usually only > leaves in > > > > this > > > > > state for a short while and then being deleted or transited to > other > > > > > states. > > > > > > > > > > > > This is not strictly true for the "Empty" state which we also use to > > > > represent simple groups which only use the coordinator to store > offsets. I > > > > think we may as well cover all the states if we're going to cover > any of > > > > them specifically. > > > > > > > > -Jason > > > > > > > > > > > > > > > > On Fri, Jul 21, 2017 at 9:45 AM, Guozhang Wang <wangg...@gmail.com> > wrote: > > > > > > > > > My two cents: > > > > > > > > > > "Dead" and "Empty" states are transient: groups usually only > leaves in > > > > this > > > > > state for a short while and then being deleted or transited to > other > > > > > states. > > > > > > > > > > Since we have the existing "*NumGroups*" metric, `*NumGroups - > > > > > **NumGroupsRebalancing > > > > > - **NumGroupsAwaitingSync`* should cover the above three, where > "Stable" > > > > > should be contributing most of the counts: If we have a bug that > causes > > > > the > > > > > num.Dead / Empty to keep increasing, then we would observe > `NumGroups` > > > > keep > > > > > increasing which should be sufficient for alerting. And trouble > shooting > > > > of > > > > > the issue could be relying on the log4j. > > > > > > > > > > *Guozhang* > > > > > > > > > > On Fri, Jul 21, 2017 at 7:19 AM, Ismael Juma <ism...@juma.me.uk> > wrote: > > > > > > > > > > > Thanks for the KIP, Colin. This will definitely be useful. One > > > > question: > > > > > > would it be useful to have a metric for for the number of groups > in > > > > each > > > > > > possible state? The KIP suggests "PreparingRebalance" and > > > > "AwaitingSync". > > > > > > That leaves "Stable", "Dead" and "Empty". Are those not useful? > > > > > > > > > > > > Ismael > > > > > > > > > > > > On Thu, Jul 20, 2017 at 6:52 PM, Colin McCabe < > cmcc...@apache.org> > > > > > wrote: > > > > > > > > > > > > > Hi all, > > > > > > > > > > > > > > I posted "KIP-180: Add a broker metric specifying the number of > > > > > consumer > > > > > > > group rebalances in progress" for discussion: > > > > > > > > > > > > > > https://cwiki.apache.org/confluence/display/KAFKA/KIP- > > > > > > > 180%3A+Add+a+broker+metric+specifying+the+number+of+ > > > > > > > consumer+group+rebalances+in+progress > > > > > > > > > > > > > > Check it out. > > > > > > > > > > > > > > cheers, > > > > > > > Colin > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > > -- Guozhang > > > > > > > > > > > > > > > > > > > > > -- > > > -- Guozhang >
