Hi, Colin, Thanks for the update. A few more follow up comments.
100. FailedReplicaRecord: Since this is reported by each broker independently, perhaps we could use a more concise representation that has a top level broker field, an array of topics, which has an array of partitions. 200. Sounds good. If we remove the broker-side fencing logic, do we plan to still keep FENCED in broker state? Do we plan to expose the new states through the existing BrokerState metric and if so, what are the values for the new states? 201. This may be fine too. Could we document what happens when the broker.id/controller.id in metadata.properties don't match the broker config when the broker starts up? 204. There is still "The highest metadata offset which the broker has not reached" referenced under BrokerRegistration. 206. Is that separate step needed given KIP-516? With KIP-516 ( https://cwiki.apache.org/confluence/display/KAFKA/KIP-516%3A+Topic+Identifiers#KIP516:TopicIdentifiers-LeaderAndIsr), we don't need to wait for the topic data to be removed from all brokers before removing the topic metadata. The combination of unmatching topicId or the missing topicId from the metadata is enough for the broker to clean up deleted topics asynchronously. Jun On Tue, Dec 8, 2020 at 5:27 PM Colin McCabe <cmcc...@apache.org> wrote: > On Thu, Dec 3, 2020, at 16:37, Jun Rao wrote: > > Hi, Colin, > > > > Thanks for the updated KIP. A few more comments below. > > > > Hi Jun, > > Thanks again for the reviews. > > > 80.2 For deprecated configs, we need to include zookeeper.* and > > broker.id.generation.enable. > > > > Added. > > > 83.1 If a broker is down, does the controller keep the previously > > registered broker epoch forever? If not, how long does the controller > keep > > it? What does the controller do when receiving a broker heartbeat request > > with an unfound broker epoch? > > > > Yes, the controller keeps the previous registration forever. > > Broker heartbeat requests with an incorrect broker epoch will be rejected > with STALE_BROKER_EPOCH. > > > 100. Have you figured out if we need to add a new record type for > reporting > > partitions on failed disks? > > > > I added FailedReplicaRecord to reflect the case where a JBOD directory has > failed, leading to failed replicas. > > > 102. For debugging purposes, sometimes it's useful to read the metadata > > topic using tools like console-consumer. Should we support that and if > so, > > how? > > > > For now, we have the ability to read the metadata logs with the dump-logs > tool. I think we will come up with some other tools in the future as we > get experience. > > > 200. "brokers which are fenced will not appear in MetadataResponses. The > > broker will not respond to these requests-- instead, it will simply > > disconnect." If the controller is partitioned off from the brokers, this > > design will cause every broker to stop accepting new client requests. In > > contrast, if ZK is partitioned off, the existing behavior is that the > > brokers can continue to work based on the last known metadata. So, I am > not > > sure if we should change the existing behavior because of the bigger > impact > > in the new one. Another option is to keep the existing behavior and > expose > > a metric for fenced brokers so that the operator could be alerted. > > > > I'm skeptical about how well running without ZK currently works. However, > I will move the broker-side fencing into a follow-up KIP. This KIP is > already pretty large and there is no hard dependency on this. There may > also be other ways of accomplishing the positive effects of what > broker-side fencing, so more discussion is needed. > > > 201. I read Ron's comment, but I am still not sure the benefit of keeping > > broker.id and controller.id in meta.properties. It seems that we are > just > > duplicating the same info in two places and have the additional burden of > > making sure the values in the two places are consistent. > > > > I think the reasoning is that having broker.id protects us against > accidentally bringing up a broker with a disk from a different broker. I > don't feel strongly about this but it seemed simpler to keep it. > > > 202. controller.connect.security.protocol: Is this needed since > > controller.listener.names and listener.security.protocol.map imply the > > security protocol already? > > > > You're right, this isn't needed. I'll remove it. > > > 203. registration.heartbeat.interval.ms: It defaults to 2k. ZK uses 1/3 > of > > the session timeout for heartbeat. So, given the default 18k for > > registration.lease.timeout.ms, should we default > > registration.heartbeat.interval.ms to 6k? > > > > 6 seconds seems like a pretty long time between heartbeats. It might be > useful to know when a broker is missing heartbeats, with less time than > that. I provisionally set it to 3 seconds (we can always change later...) > > I also changed the name of these configurations to " > broker.heartbeat.interval.ms" and "broker.registration.timeout.ms" to try > to clarify them a bit. > > > 204. "The highest metadata offset which the broker has not reached." It > > seems this should be "has reached". > > > > I changed this to "one more than the highest metadata offset which the > broker has reached." > > > 205. UnfenceBrokerRecord and UnregisterBrokerRecord: To me, they seem to > be > > the same. Do we need both? > > > > Unregistration means that the broker has been removed from the cluster. > That is different than unfencing, which marks the broker as active. > > > 206. TopicRecord: The Deleting field is used to indicate that the topic > is > > being deleted. I am wondering if this is really needed since RemoveTopic > > already indicates the same thing. > > > > RemoveTopic is the last step, that scrubs all metadata about the topic. > In order to get to that last step, the topic data needs to removed from all > brokers (after each broker notices that the topic is being deleted). > > best, > Colin > > > Jun > > > > On Wed, Dec 2, 2020 at 2:50 PM Colin McCabe <cmcc...@apache.org> wrote: > > > > > On Wed, Dec 2, 2020, at 14:07, Ron Dagostino wrote: > > > > Hi Colin. Thanks for the updates. It's now clear to me that brokers > > > > keep their broker epoch for the life of their JVM -- they register > > > > once, get their broker epoch in the response, and then never > > > > re-register again. Brokers may get fenced, but they keep the same > > > > broker epoch for the life of their JVM. The incarnation ID is also > > > > kept for the life of the JVM but is generated by the broker itself > > > > upon startup, and the combination of the two allows the Controller to > > > > act idempotently if any previously-sent registration response gets > > > > lost. Makes sense. > > > > > > > > > > Thanks, Ron. That's a good summary. > > > > > > > One thing I wonder about is if it might be helpful for the broker to > > > > send the Cluster ID as determined from its meta.properties file in > its > > > > registration request. Does it even make sense for the broker to > > > > successfully register and enter the Fenced state if it has the wrong > > > > Cluster ID? > > > > > > Yeah, that's a good idea. Let's have the broker pass its cluster ID in > > > the registration RPC, and then registration can fail if the broker is > > > configured for the wrong cluster. > > > > > > > The nextMetadatOffset value that the broker communicates > > > > in its registration request only has meaning within the correct > > > > cluster, so it feels to me that the Controller should have some way > to > > > > perform this sanity check. There is currently (pre-KIP 500) a check > > > > in the broker to make sure its configured cluster ID matches the one > > > > stored in ZooKeeper, and we will have to perform this validation > > > > somewhere in the KIP-500 world. If the Controller doesn't do it > > > > within the registration request then the broker will have to make a > > > > metadata request to the Controller, retrieve the Cluster ID, and > > > > perform the check itself. It feels to me that it might be better for > > > > the Controller to just do it, and then the broker doesn't have to > > > > worry about it anymore once it successfully registers. > > > > > > > > I also have a question about the broker.id value and > meta.properties. > > > > The KIP now says "In version 0 of meta.properties, there is a > > > > broker.id field. Version 1 does not have this field. It is no > longer > > > > needed because we no longer support dynamic broker id assignment." > > > > But then there is an example version 1 meta.properties file that > shows > > > > the broker.id value. I actually wonder if maybe the broker.id value > > > > would be good to keep in the version 1 meta.properties file because > it > > > > currently (pre-KIP 500, version 0) acts as a sanity check to make > sure > > > > the broker is using the correct log directory. Similarly with the > > > > controller.id value on controllers -- it would allow the same type > of > > > > sanity check for quorum controllers. > > > > > > > > > > That's a good point. I will add broker.id back, and also add > > > controller.id as a possibility. > > > > > > cheers, > > > Colin > > > > > > > > > > > On Mon, Nov 30, 2020 at 7:41 PM Colin McCabe <cmcc...@apache.org> > wrote: > > > > > > > > > > On Fri, Oct 23, 2020, at 16:10, Jun Rao wrote: > > > > > > Hi, Colin, > > > > > > > > > > > > Thanks for the reply. A few more comments. > > > > > > > > > > Hi Jun, > > > > > > > > > > Thanks again for the reply. Sorry for the long hiatus. I was on > > > vacation for a while. > > > > > > > > > > > > > > > > > 55. There is still text that favors new broker registration. > "When a > > > broker > > > > > > first starts up, when it is in the INITIAL state, it will always > > > "win" > > > > > > broker ID conflicts. However, once it is granted a lease, it > > > transitions > > > > > > out of the INITIAL state. Thereafter, it may lose subsequent > > > conflicts if > > > > > > its broker epoch is stale. (See KIP-380 for some background on > > > broker > > > > > > epoch.) The reason for favoring new processes is to accommodate > the > > > common > > > > > > case where a process is killed with kill -9 and then restarted. > We > > > want it > > > > > > to be able to reclaim its old ID quickly in this case." > > > > > > > > > > > > > > > > Thanks for the reminder. I have clarified the language here. > > > Hopefully now it is clear that we don't allow quick re-use of broker > IDs. > > > > > > > > > > > 80.1 Sounds good. Could you document that listeners is a required > > > config > > > > > > now? It would also be useful to annotate other required configs. > For > > > > > > example, controller.connect should be required. > > > > > > > > > > > > > > > > I added a note specifying that these are required. > > > > > > > > > > > 80.2 Could you list all deprecated existing configs? Another one > is > > > > > > control.plane.listener.name since the controller no longer sends > > > > > > LeaderAndIsr, UpdateMetadata and StopReplica requests. > > > > > > > > > > > > > > > > I added a section specifying some deprecated configs. > > > > > > > > > > > 83.1 It seems that the broker can transition from FENCED to > RUNNING > > > without > > > > > > registering for a new broker epoch. I am not sure how this works. > > > Once the > > > > > > controller fences a broker, there is no need for the controller > to > > > keep the > > > > > > boker epoch around. So, if the fenced broker's heartbeat request > > > with the > > > > > > existing broker epoch will be rejected, leading the broker back > to > > > the > > > > > > FENCED state again. > > > > > > > > > > > > > > > > The broker epoch refers to the broker registration. So we DO keep > the > > > broker epoch around even while the broker is fenced. > > > > > > > > > > The broker epoch changes only when there is a new broker > > > registration. Fencing or unfencing the broker doesn't change the > broker > > > epoch. > > > > > > > > > > > 83.5 Good point on KIP-590. Then should we expose the controller > for > > > > > > debugging purposes? If not, we should deprecate the controllerID > > > field in > > > > > > MetadataResponse? > > > > > > > > > > > > > > > > I think it's OK to expose it for now, with the proviso that it > won't > > > be reachable by clients. > > > > > > > > > > > 90. We rejected the shared ID with just one reason "This is not a > > > good idea > > > > > > because NetworkClient assumes a single ID space. So if there is > > > both a > > > > > > controller 1 and a broker 1, we don't have a way of picking the > > > "right" > > > > > > one." This doesn't seem to be a strong reason. For example, we > could > > > > > > address the NetworkClient issue with the node type as you pointed > > > out or > > > > > > using the negative value of a broker ID as the controller ID. > > > > > > > > > > > > > > > > It would require a lot of code changes to support multiple types of > > > node IDs. It's not clear to me that the end result would be better -- > I > > > tend to think it would be worse, since it would be more complex. In a > > > similar vein, using negative numbers seems dangerous, since we use > > > negatives or -1 as "special values" in many places. For example, -1 > often > > > represents "no such node." > > > > > > > > > > One important thing to keep in mind is that we want to be able to > > > transition from a broker and a controller being co-located to them no > > > longer being co-located. This is much easier to do when they have > separate > > > IDs. > > > > > > > > > > > 100. In KIP-589 > > > > > > < > > > > https://cwiki.apache.org/confluence/display/KAFKA/KIP-589+Add+API+to+update+Replica+state+in+Controller > > > >, > > > > > > the broker reports all offline replicas due to a disk failure to > the > > > > > > controller. It seems this information needs to be persisted to > the > > > > > > metadata > > > > > > log. Do we have a corresponding record for that? > > > > > > > > > > > > > > > > Hmm, I have to look into this a little bit more. We may need a new > > > record type. > > > > > > > > > > > 101. Currently, StopReplica request has 2 modes, without deletion > > > and with > > > > > > deletion. The former is used for controlled shutdown and handling > > > disk > > > > > > failure, and causes the follower to stop. The latter is for topic > > > deletion > > > > > > and partition reassignment, and causes the replica to be deleted. > > > Since we > > > > > > are deprecating StopReplica, could we document what triggers the > > > stopping > > > > > > of a follower and the deleting of a replica now? > > > > > > > > > > > > > > > > RemoveTopic triggers deletion. In general the functionality of > > > StopReplica is subsumed by the metadata records. > > > > > > > > > > > 102. Should we include the metadata topic in the > MetadataResponse? > > > If so, > > > > > > when it will be included and what will the metadata response look > > > like? > > > > > > > > > > > > > > > > No, it won't be included in the metadata response sent back from > the > > > brokers. > > > > > > > > > > > 103. "The active controller assigns the broker a new broker > epoch, > > > based on > > > > > > the latest committed offset in the log." This seems inaccurate > since > > > the > > > > > > latest committed offset doesn't always advance on every log > append. > > > > > > > > > > > > > > > > Given that the new broker epoch won't be visible until the commit > has > > > happened, I have changed this to "the next available offset in the log" > > > > > > > > > > > 104. REGISTERING(1) : It says "Otherwise, the broker moves into > the > > > FENCED > > > > > > state.". It seems this should be RUNNING? > > > > > > > > > > > > 105. RUNNING: Should we require the broker to catch up to the > > > metadata log > > > > > > to get into this state? > > > > > > > > > > For 104 and 105, these sections have been reworked. > > > > > > > > > > best, > > > > > Colin > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > > > > > Jun > > > > > > > > > > > > > > > > > > > > > > > > On Fri, Oct 23, 2020 at 1:20 PM Colin McCabe <cmcc...@apache.org > > > > > wrote: > > > > > > > > > > > > > On Wed, Oct 21, 2020, at 05:51, Tom Bentley wrote: > > > > > > > > Hi Colin, > > > > > > > > > > > > > > > > On Mon, Oct 19, 2020, at 08:59, Ron Dagostino wrote: > > > > > > > > > > Hi Colin. Thanks for the hard work on this KIP. > > > > > > > > > > > > > > > > > > > > I have some questions about what happens to a broker > when it > > > becomes > > > > > > > > > > fenced (e.g. because it can't send a heartbeat request to > > > keep its > > > > > > > > > > lease). The KIP says "When a broker is fenced, it cannot > > > process any > > > > > > > > > > client requests. This prevents brokers which are not > > > receiving > > > > > > > > > > metadata updates or that are not receiving and processing > > > them fast > > > > > > > > > > enough from causing issues to clients." And in the > > > description of the > > > > > > > > > > FENCED(4) state it likewise says "While in this state, > the > > > broker > > > > > > > does > > > > > > > > > > not respond to client requests." It makes sense that a > > > fenced broker > > > > > > > > > > should not accept producer requests -- I assume any such > > > requests > > > > > > > > > > would result in NotLeaderOrFollowerException. But what > > > about KIP-392 > > > > > > > > > > (fetch from follower) consumer requests? It is > conceivable > > > that > > > > > > > these > > > > > > > > > > could continue. Related to that, would a fenced broker > > > continue to > > > > > > > > > > fetch data for partitions where it thinks it is a > follower? > > > Even if > > > > > > > > > > it rejects consumer requests it might still continue to > > > fetch as a > > > > > > > > > > follower. Might it be helpful to clarify both decisions > > > here? > > > > > > > > > > > > > > > > > > Hi Ron, > > > > > > > > > > > > > > > > > > Good question. I think a fenced broker should continue to > > > fetch on > > > > > > > > > partitions it was already fetching before it was fenced, > > > unless it > > > > > > > hits a > > > > > > > > > problem. At that point it won't be able to continue, > since it > > > doesn't > > > > > > > have > > > > > > > > > the new metadata. For example, it won't know about > leadership > > > changes > > > > > > > in > > > > > > > > > the partitions it's fetching. The rationale for > continuing to > > > fetch > > > > > > > is to > > > > > > > > > try to avoid disruptions as much as possible. > > > > > > > > > > > > > > > > > > I don't think fenced brokers should accept client requests. > > > The issue > > > > > > > is > > > > > > > > > that the fenced broker may or may not have any data it is > > > supposed to > > > > > > > > > have. It may or may not have applied any configuration > > > changes, etc. > > > > > > > that > > > > > > > > > it is supposed to have applied. So it could get pretty > > > confusing, and > > > > > > > also > > > > > > > > > potentially waste the client's time. > > > > > > > > > > > > > > > > > > > > > > > > > > When fenced, how would the broker reply to a client which did > > > make a > > > > > > > > request? > > > > > > > > > > > > > > > > > > > > > > Hi Tom, > > > > > > > > > > > > > > The broker will respond with a retryable error in that case. > Once > > > the > > > > > > > client has re-fetched its metadata, it will no longer see the > > > fenced broker > > > > > > > as part of the cluster. I added a note to the KIP. > > > > > > > > > > > > > > best, > > > > > > > Colin > > > > > > > > > > > > > > > > > > > > > > > Thanks, > > > > > > > > > > > > > > > > Tom > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >
