[DISCUSS] Optimize the invoke retry of async repsonder when some nameservices are slow

Hello, everyone, so sorry to bother you and I would like to discuss how to optimize the invoke retry logic of async repsonder when some nameservices are slow. Currently, all nameservices share one async responder thread pool which is in class AsyncRpcProtocolPBUtil. Think below situation, we

[DISCUSS] Optimize the invoke retry of async repsonder when some nameservices are slow

invoke method (using async repsonder thread here) and target nameserivce has poor performance. Best Regards, - Zhang Haobo Replied Message | From | Xiaoqiao He | | Date | 04/23/2025 20:57 | | To | Steve Loughran | | Cc | Zhanghaobo , common-...@hadoop.apache.org , Hdfs-dev | | Subject

Re: Can not run HDFS balancer cause metrics already exists

try to delete /system/balancer.id and search some error or warn logs in namenode. Replied Message | From | Sébastien Rebecchi | | Date | 3/9/2025 23:08 | | To | Zhanghaobo | | Cc | hadoop-user-maillist, hdfs-dev | | Subject | Re: Can not run HDFS balancer cause metrics already exists

Re: Can not run HDFS balancer cause metrics already exists

got it, you can use it as a service and see what will happen. Replied Message | From | Sébastien Rebecchi | | Date | 03/09/2025 22:22 | | To | Zhanghaobo | | Cc | u...@hadoop.apache.org、hdfs-dev@hadoop.apache.org | | Subject | Re: Can not run HDFS balancer cause metrics already exists

Re: Can not run HDFS balancer cause metrics already exists

Hi, @Sébastien Rebecchi Don't know more details about how you start balancer, did you use -asService? Replied Message | From | Sébastien Rebecchi | | Date | 3/9/2025 18:03 | | To | , | | Subject | Re: Can not run HDFS balancer cause metrics already exists | Hello Could anyone

Re: [VOTE] Request to merge branch HDFS-17531 into trunk.

+1 - Zhang Haobo Replied Message | From | Xiaoqiao He | | Date | 02/14/2025 19:38 | | To | Ayush Saxena | | Cc | Hui Fei , jian zhang , Hdfs-dev , , | | Subject | Re: [VOTE] Request to merge branch HDFS-17531 into trunk. | +1 Best Regards, - He Xiaoqiao On Fri, Feb 14, 2025 at

Re: [DISCUSS] Request to merge branch HDFS-17531 into trunk.

@Hui Fei Hi, Sir: For the first opinion, I have create an umbrella JIRA https://issues.apache.org/jira/browse/HDFS-17716 and move non-core JIRA under it. Best Wishes Haobo Zhang Replied Message | From | Hui Fei | | Date | 01/22/2025 17:37 | | To | jian zhang | | Cc | Hdfs-dev ,

Re: [VOTE] Merge HDFS-17384 to trunk

+1 (binding) It has been running stably for several months on prod clusters Replied Message | From | Hui Fei | | Date | 01/13/2025 17:14 | | To | ZanderXu | | Cc | Hdfs-dev , , Ayush Saxena , Xiaoqiao He , slfan1989 , Zhanghaobo , , | | Subject | Re: [VOTE] Merge HDFS-17384 to trunk

Discussion about NameNode Fine-grained locking

Thanks for your great work! +1 for merging phase 1 codes. My product clusters have been running phase 1 codes for several months, it looks good. Hope to push this feature forward. | | 张浩博 | | hfutzhan...@163.com | Replied Message | From | haiyang hu | | Date | 12/31/2024 23:08

Re: [Discuss] RBF: Aynchronous router RPC.

@zhangjian This is a great job, Thanks for working on this. I have a question here. Should we let AsyncHandler thread pool have the ability of isolation? I’m worried about async handler thread will be exhausted when there are some nameservices which have bad performance. | | 张浩博 | | hfutzhan.

[jira] [Created] (HDFS-16452) Msync RPC should send to Acitve Namenode directly

zhanghaobo created HDFS-16452: - Summary: Msync RPC should send to Acitve Namenode directly Key: HDFS-16452 URL: https://issues.apache.org/jira/browse/HDFS-16452 Project: Hadoop HDFS Issue Type

[jira] [Created] (HDFS-16425) Add debug log when requeue observer read call

zhanghaobo created HDFS-16425: - Summary: Add debug log when requeue observer read call Key: HDFS-16425 URL: https://issues.apache.org/jira/browse/HDFS-16425 Project: Hadoop HDFS Issue Type: New

[jira] [Created] (HDFS-16424) Add a new rpcMetric called requeueTime

zhanghaobo created HDFS-16424: - Summary: Add a new rpcMetric called requeueTime Key: HDFS-16424 URL: https://issues.apache.org/jira/browse/HDFS-16424 Project: Hadoop HDFS Issue Type: Improvement

[jira] [Created] (HDFS-16382) Router getContentSummary rpc compute sub-directory repeatedly

zhanghaobo created HDFS-16382: - Summary: Router getContentSummary rpc compute sub-directory repeatedly Key: HDFS-16382 URL: https://issues.apache.org/jira/browse/HDFS-16382 Project: Hadoop HDFS

[jira] [Created] (HDFS-16372) RefreshNodes should not use System.out but System.err when exceptions occur

zhanghaobo created HDFS-16372: - Summary: RefreshNodes should not use System.out but System.err when exceptions occur Key: HDFS-16372 URL: https://issues.apache.org/jira/browse/HDFS-16372 Project: Hadoop

[jira] [Created] (HDFS-16368) DFSAdmin supports refresh topology info without restarting namenode

zhanghaobo created HDFS-16368: - Summary: DFSAdmin supports refresh topology info without restarting namenode Key: HDFS-16368 URL: https://issues.apache.org/jira/browse/HDFS-16368 Project: Hadoop HDFS

[jira] [Created] (HDFS-16226) Generate locatedBlocks can use bi-search to find curBlk

zhanghaobo created HDFS-16226: - Summary: Generate locatedBlocks can use bi-search to find curBlk Key: HDFS-16226 URL: https://issues.apache.org/jira/browse/HDFS-16226 Project: Hadoop HDFS Issue