-
```
Checked in memcached-1.4.15.
assoc_maintenance_thread() (assoc.c) has a piece of code with:
slabs_rebalancer_resume();
....
slabs_rebalancer_pause();
The problem is that slabs_rebalancer_resume…
-
It is pretty easy to replicate excessive Go memory usage in a 3 node cluster by pausing / resuming one of the nodes. Basic setup was to create a 3 node cluster using `roachdemo`, apply load using `blo…
-
We currently provide the following APIs to users who want to create distributed tables.
- `master_create_distributed_table(table_name, distribution_column, distribution_method)` where the `distributio…
-
Two questions about scaling out:
# Scenario 1
I have one master, and two workers. Worker 2 is under heavy load for some reason, it turns out we need to add a third worker and move the troubled shard…
synic updated
8 years ago
-
```
Checked in memcached-1.4.15.
assoc_maintenance_thread() (assoc.c) has a piece of code with:
slabs_rebalancer_resume();
....
slabs_rebalancer_pause();
The problem is that slabs_rebalancer_resume…
-
由于很早前rebalancer速度太慢,加了几个虚拟机机器,然后将datanode中磁盘使用少的机器以nfs形式挂载到虚拟机上,再做rebalancer,后来磁盘使用率够了,将nfs umount了,重新启动这几个datanode,报错如下:
```
2016-05-06 11:23:54,378 FATAL org.apache.hadoop.hdfs.server.datanode.Data…
-
```
Checked in memcached-1.4.15.
assoc_maintenance_thread() (assoc.c) has a piece of code with:
slabs_rebalancer_resume();
....
slabs_rebalancer_pause();
The problem is that slabs_rebalancer_resume…
-
On the beta cluster, we have a range that for some reason failed to split for a long time (for reasons that are still unclear). This cluster is currently configured to split at 6MB; this range had man…
-
```
Checked in memcached-1.4.15.
assoc_maintenance_thread() (assoc.c) has a piece of code with:
slabs_rebalancer_resume();
....
slabs_rebalancer_pause();
The problem is that slabs_rebalancer_resume…
-
```
Checked in memcached-1.4.15.
assoc_maintenance_thread() (assoc.c) has a piece of code with:
slabs_rebalancer_resume();
....
slabs_rebalancer_pause();
The problem is that slabs_rebalancer_resume…