Vincent Lee
Vincent Lee
Yes, we use the etcd as the storage of the cluster meta data, such as topic partition number, replicator, the node placement strategy and some other options. @ploxiln There are...
For the requeue, we have 2 strategies currently, first is put back to the requeued channel and second is put back to the end of the file. So for deferred-publish...
Yeah, I think the first thing need to be done is: every message should be written to the disk file without involving the channel overflow to disk. All theses features...
After reviewed the changes, I noticed that WAL is added for each topic. In my redesigned solution, I tried to make the least changes. Since the topic disk file has...
@mreiferson channel cursor has the fileNum of the disk file and the offset of the disk file. And while a message read from file, we fill the offset and the...
1 second is very long and will cause too many errors in high TPS. While we restart the server normally we should expect the least error. Maybe it should handle...
Thank you and wish for great improvement.
能不能提供一下crash前后的一些服务端log. 看官方有个issue没解决, 和这个可能有关 https://github.com/facebook/rocksdb/issues/2560 #
是的, pd会定期平衡自动搞定的
确实如此, 可能当时有其他调用入口, 做了重复处理, 后面没有调整