[
https://issues.apache.org/jira/browse/HDFS-12200?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16200851#comment-16200851
]
Hanisha Koneru commented on HDFS-12200:
---------------------------------------
Hi [~yangjiandan]]
If I understand correctly, you want to add an option to disable rack lookup for
uncached hosts, right?
This patch would optimize rack lookup for an unusual architecture. Generally
HDFS and YARN are deployed on the same machines.
As such, I think the use case might be very limited.
Also, can you please update the Jira title to reflect that this is for adding
an option to disable rack lookup for uncached hosts.
Currently, when a DN registers, the DNS to Switch mapping is resolved during
the registration process itself. With this change and with
resolve-non-cached-host set to false, the rack resolution for new DN will be
skipped during registration. This might cause the rack new DN's rack resolution
to be incorrectly cached as default in the following 2 cases:
- a new DN is added, rack mapping script is updated, and the DN registers
before refreshNodes is called:
The rack will be resolved to default during DN registration. And since it has
already been resolved, it would not be updated during refreshNodes.
- a new DN is added, refreshNodes is called and then the rack mapping script is
updated:
In this case too, the mapping for the new DN would be updated with default
instead of the correct mapping.
> Optimize CachedDNSToSwitchMapping to avoid 100% cpu utilization
> ---------------------------------------------------------------
>
> Key: HDFS-12200
> URL: https://issues.apache.org/jira/browse/HDFS-12200
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: namenode
> Reporter: Jiandan Yang
> Assignee: Jiandan Yang
> Attachments: HDFS-12200-001.patch, HDFS-12200-002.patch,
> HDFS-12200-003.patch, cpu_ utilization.png, nn_thread_num.png
>
>
> 1. Background :
> Our hadoop cluster is disaggregated storage and compute, HDFS is deployed to
> 600+ machines, YARN is deployed to another machine pool.
> We found that sometimes NameNode cpu utilization rate of 90% or even 100%.
> The most serious is cpu utilization rate of 100% for a long time result in
> writing journalNode timeout, eventually leading to NameNode hang up. The
> reason is offline tasks running in a few hundred servers access HDFS at the
> same time, NameNode resolve rack of client machine, started several hundreds
> to two thousand sub-process.
> {code:java}
> "process reaper"#10864 daemon prio=10 os_prio=0 tid=0x00007fe270a31800
> nid=0x38d93 runnable [0x00007fcdc36fc000]
> java.lang.Thread.State: RUNNABLE
> at java.lang.UNIXProcess.waitForProcessExit(Native Method)
> at java.lang.UNIXProcess.lambda$initStreams$4(UNIXProcess.java:301)
> at java.lang.UNIXProcess$$Lambda$7/1447689627.run(Unknown Source)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1147)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:622)
> at java.lang.Thread.run(Thread.java:834
> {code}
> Our configuration as follows:
> {code:java}
> net.topology.node.switch.mapping.impl = ScriptBasedMapping,
> net.topology.script.file.name = 'a python script'
> {code}
> 2. Optimization
> In order to solve these two problems, we have optimized the
> CachedDNSToSwitchMapping
> (1) Added the DataNode IP list to the file of dfs.hosts configured. when
> NameNode starts it preloads DataNode rack information to the cache, get a
> batch of racks of hosts when running script once (the corresponding
> configuration is net.topology.script.number,the default value of 100)
> (2) Step (1) has ensured that the cache has all the DataNodes’ rack, so if
> the cache did not hit, then the host must be a client machine, then directly
> return /default-rack,
> (3) Each time you add new DataNodes you need to add the new DataNodes’ IP
> address to the file specified by dfs.hosts, and then run command of bin/hdfs
> dfsadmin -refreshNodes, it will put the newly added DataNodes’ rack into cache
> (4) Add new configuration items dfs.namenode.topology.resolve-non-cache-host,
> the value is false to open the above function, and the value is true to turn
> off the above functions, default value is true to keep compatibility
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
