[
https://issues.apache.org/jira/browse/HDFS-8578?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14591256#comment-14591256
]
Vinayakumar B commented on HDFS-8578:
-------------------------------------
bq. OK, so your idea is that that will allow us to parallelize upgrades between
different block pools. Fair enough.
Not exactly. {{DataStorage#addStorageLocations(..)}} is called per blockpool
only. And two different blockpools initialization is synchronized in
Datanode.java itself. {code}synchronized (this) {
storage.recoverTransitionRead(this, nsInfo, dataDirs, startOpt);
}{code}
My change was because for the same blockpool, each directory is being loaded
separately in . {{DataStorage#addStorageLocations(..)}}.
{code}List<File> bpDataDirs = new ArrayList<File>();
bpDataDirs.add(BlockPoolSliceStorage.getBpRoot(bpid, new File(root,
STORAGE_DIR_CURRENT)));
...
...
bpStorage.recoverTransitionRead(datanode, nsInfo, bpDataDirs, startOpt);
{code}
So in {{BlockPoolSliceStorage#loadBpStorageDirectories(... Collection<File>
dataDirs, ...)}}, {{dataDirs}} will have only one directory during startup. So
adding parallelism there will not change anything.
bq. I don't think this is true. The DN will not start up if more than
dfs.datanode.failed.volumes.tolerated volumes have failed, as per this code:
Yes!! You are right. I just checked only
{{DataStorage#addStorageLocations(..)}}, had totally forgot about
{{dfs.datanode.failed.volumes.tolerated}}.
> On upgrade, Datanode should process all storage/data dirs in parallel
> ---------------------------------------------------------------------
>
> Key: HDFS-8578
> URL: https://issues.apache.org/jira/browse/HDFS-8578
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: datanode
> Reporter: Raju Bairishetti
> Priority: Critical
> Attachments: HDFS-8578-01.patch, HDFS-8578-02.patch
>
>
> Right now, during upgrades datanode is processing all the storage dirs
> sequentially. Assume it takes ~20 mins to process a single storage dir then
> datanode which has ~10 disks will take around 3hours to come up.
> *BlockPoolSliceStorage.java*
> {code}
> for (int idx = 0; idx < getNumStorageDirs(); idx++) {
> doTransition(datanode, getStorageDir(idx), nsInfo, startOpt);
> assert getCTime() == nsInfo.getCTime()
> : "Data-node and name-node CTimes must be the same.";
> }
> {code}
> It would save lots of time during major upgrades if datanode process all
> storagedirs/disks parallelly.
> Can we make datanode to process all storage dirs parallelly?
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
