Iman Sahebi
Iman Sahebi
Hi, Currently when calling rebalance_disk=true, it rebalances all disks of all brokers. When #2145 merges, it resolves part of the problem that we can throttle kafka I/O to avoid cluttering...
Hi, When trying to rebalance disk with dry run true, in a large cluster, it takes several minutes in our case. This causes the CC to show a pending in...
Same as SlowBrokerFinder, this is useful for when a disk is slow and we want to automatically remove it from cluster.
Hi, we're using kafka cruise control in a big environment where the I/O rate of our disks are our bottleneck. With existing goals, we can not achieve this and we...
Hi, Currently when DISK_FAILURE occurs and triggers self healing, those dead replicas are moved outside of that broker. But when we have multiple disks on the same broker, it should...
Hi, As far as I know and saw in code and also in experience, cruise control does inter-broker and intra-broker rebalancing in two different ways and it's not possible to...
Hi, Currently when we want to run rebalance disk, e.g. via `/proposals?rebalance_disk=true&verbose=true`, it doesn't show which logdirs on the brokers are being moved. It just says that for example replica...
Hi, Currently when we call rebalancing of any kind, it starts instantly and continues until it finishes. But in reality, clusters have some times which are at their peak load...
Hi, Currently when we define DiskCapacity and when we want to call `/rebalance?rebalance_disk=true`, it takes all these disks into account. But it's always possible that a disk corrupts over time...
Hi, Currently, the progress of execution is calculated by the size of replicas that their movement is already finished. But suppose we have a large cluster that has replicas of...