Thanks Stephen for the summary and lead the discussion. Problem 1&2, we definitely need such indicator, and move around the closed container replicas to make sure not only replica number meets the requirements, but also it meets the rack aware requirements. I'm not familar with Recon server function. It it cann't be done in Replication Manager, Recon server is also a good choice. Problem 5, the closed container placement policy is a little different than pipeline placement policy. The former considers rack awareness and datanode free disk space. The latter consider rack awareness and datanode pipeline load balance. Agree to unify the configuration and keep the configuration consistent.
Problem 4. thanks for bring storming four possible solutions for this problem. For the same mode, there are actually 2 sub-cases, one is the safe mode of a new cluster, another is the safe node of an existing stable cluster. Solution 1~3 seems work for safe node of existing cluster. While Solution 4 seems work for both sub-cases. Problem 3. It's inevitable to allocate new pipelines which doesn't meet rack tolerace policy. One way is we can close the pipeline once we find a better pipeline candidate sets. Another way is migrate one member of the pipeline from one datanode to another datanode. For example, we can let the 4th datanode join the pipeline, can learn the data, once it catches up, we can turn on this datanode, and turn off another beening replaced datanode. The second solution requires RATIS support. I'm not clear if it's feasible or how much workload will be. The first solution is quite straight-forward. Though close pipeline is a cost operation. But with multi-raft enabled, maybe we can offord this cost, for a higher data security guarantee. Bests, Sammi On Thu, Mar 12, 2020 at 9:19 PM Stephen O'Donnell <sodonn...@cloudera.com.invalid> wrote: > We had a discussion yesterday with some of the team related to network > topology and we came up with the following list of proposals which probably > need to be implemented to cover some edge cases and make the feature more > supportable. I am sharing them here to gather any further ideas, problems > and feedback before we attempt to fix these issues. > > > Problem 1: > > As of now, there is no tool to tell us if any containers are not replicated > on 2 racks. > > Solution: > > A feature should be added to Recon to check the replication and highlight > containers which are not on two racks. > > > Problem 2: > > If closed containers somehow end up on only 1 rack, there is no facility to > correct that. > > Solution: > > Replication Manager should be extended to check for both under replicated > and mis-replicated containers and it should work to correct them. It was > also suggested that if a container has only 2 replicas on 1 rack, the > cluster is rack aware, and no node is available from another rack, > replication manager should not schedule a 3rd copy on the same rack. It > should instead wait for a node on another rack to become available. > > Problem 3: > > If pipelines get created which are not rack tolerant, then they will be > long lived and will create containers which are not rack tolerant for a > long time. This can happen if nodes from another rack are not available > when pipelines are being created, or 1 rack of a 2 rack cluster is stopped. > > Solution: > > The existing pipeline scrubber should be extended to check for pipelines > which are not rack tolerant and also check if there are nodes available > from at least two racks. If so, it will destroy non-rack tolerant pipelines > in a controlled fashion. > > For a badly configured cluster, eg rack_1 has 10 nodes, rack_2 has 1 node, > we should never create non-rack tolerant pipelines even though it will > reduce the cluster throughput. That is, the fall back option when creating > pipelines should only be used when there is only 1 rack available. > > > Problem 4: > > With the existing design, pipelines start to be created as soon as 3 nodes > have registered with SCM. If 3 nodes from the same rack register first, the > system does not know the cluster is rack aware as yet (the current logic > checks the number of racks which have checked in) and so it will create a > non-rack tolerant pipeline. The solution to problem 3 can take care of > this, but it seems it would be better to try to prevent these bad pipelines > getting created to begin with. Additionally, with multi-raft, it would be > better to have most nodes registered before creating pipelines to spread > them out across the cluster more evenly. > > Solution: > > SCM already has a Safemode check. It is the ideal place to add a check like > this and we decided it would make sense to have some safe mode rules which > must pass before pipelines can start to be created. Several ideas were > discussed: > > 1. Wait for a static number of nodes to register. This is simple, but a > static configuration that must be changed as the cluster grows is not > ideal. This check already exists for exiting safemode, but it would need to > be changed slightly to block pipeline creation too. > > 2. Wait for the node count to stabilize. In this way, the safemode rule > would check the node count has not changed during some interval of time, > implying all nodes have registered. A negative is slowing down the startup > time, but due to (3) below this would not be a problem on an established > cluster. > > 3. Wait for some percentage of the total expected containers to be > reported, which would imply most of the expected nodes have registered. > This check is already present to exit safe mode, so we would need it to > block pipeline creation too. The one negative is that it may not work well > for clusters with a small number of nodes or few containers (ie new > clusters). It would also be possible for all containers to be reported with > only one third of the nodes registered in an extreme case. > > 4. Wait for at least 2 racks to be registered if the cluster is configured > as rack tolerant. This does help with ensuring the pipelines are spread > across all the nodes. > > This area needs some more exploration to figure out which of these ideas is > best. > > > Problem 5: > > The closed container replication policy is different from the pipeline > policy and it is possible to configure Replication Manager to use an > incompatible policy. > > Solution: > > It may not be possible or desirable to merge the closed container placement > policy with the pipeline policy, but we need to think about unifying the > configuration so it is not possible to set incompatible options. > > Thanks, > > Stephen. >
