Couchbase操作员问题

Couchbase操作员（enterprise 1.2）舵图出现问题

对于新的couchbase集群，以下问题总是发生

couchbase操作符couchbase-operator-0.1.2 1.2

time="2019-06-25T04:22:14Z" level=error msg="failed to reconcile: unable to get buckets from cluster: [Client error `unmarshal json response`: json: cannot unmarsh
al number 897.8978978978979 into Go struct field BucketBasicStats.diskFetches of type int], [Client error `unmarshal json response`: json: cannot unmarshal number
897.8978978978979 into Go struct field BucketBasicStats.diskFetches of type int], [Client error `unmarshal json response`: json: cannot unmarshal number 897.897897
8978979 into Go struct field BucketBasicStats.diskFetches of type int]" cluster-name=couchbase-cluster-couchbase-cluster module=cluster
time="2019-06-25T04:22:22Z" level=error msg="failed to reconcile: unable to get buckets from cluster: [Client error `unmarshal json response`: json: cannot unmarsh
al number 575.7622377622378 into Go struct field BucketBasicStats.diskFetches of type int], [Client error `unmarshal json response`: json: cannot unmarshal number
575.7622377622378 into Go struct field BucketBasicStats.diskFetches of type int], [Client error `unmarshal json response`: json: cannot unmarshal number 575.762237
7622378 into Go struct field BucketBasicStats.diskFetches of type int]" cluster-name=couchbase-cluster-couchbase-cluster module=cluster

---

更新：这是现在修复

哦，那是个好主意！是的，看起来diskFetches实际上是每秒的速率，而不是客户端代码假定的整数计数器

目前我们无能为力，但我已经标记为一个关键补丁，很快将出现在1.2.1中。为您提供信息，操作员在执行任何存储桶操作（如果操作失败）之前确保集群大小正确、平衡且健康，因此您的数据是安全的

我建议您在集群定义中将spec.disableBucketManagement设置为true，并使用UI或客户端SDK创建bucket，这将完全避免这种情况，直到我们得到热修复