[jira] [Updated] (HDFS-12612) DFSStripedOutputStream#close will throw if called a second time with a failed streamer
[
https://issues.apache.org/jira/browse/HDFS-12612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei (Eddy) Xu updated HDFS-12612:
-
Resolution: Fixed
Fix Version/s: 3.0.0
Status: Resolved (was: Patch Available)
Thanks a lot for the reviews [~xiaochen] and [~andrew.wang].
Committed to trunk and branch-3.0
> DFSStripedOutputStream#close will throw if called a second time with a failed
> streamer
> --
>
> Key: HDFS-12612
> URL: https://issues.apache.org/jira/browse/HDFS-12612
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
>Affects Versions: 3.0.0-beta1
>Reporter: Andrew Wang
>Assignee: Lei (Eddy) Xu
> Labels: hdfs-ec-3.0-must-do
> Fix For: 3.0.0
>
> Attachments: HDFS-12612.00.patch, HDFS-12612.01.patch,
> HDFS-12612.02.patch, HDFS-12612.03.patch
>
>
> Found while testing with Hive. We have a cluster with 2 DNs and the XOR-2-1
> policy. If you write a file and call close() twice, it throws this exception:
> {noformat}
> 17/10/04 16:02:14 WARN hdfs.DFSOutputStream: Cannot allocate parity
> block(index=2, policy=XOR-2-1-1024k). Not enough datanodes? Exclude nodes=[]
> ...
> Caused by: java.io.IOException: Failed to get parity block, index=2
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.allocateNewBlock(DFSStripedOutputStream.java:500)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.writeChunk(DFSStripedOutputStream.java:524)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> {noformat}
> This is because in DFSStripedOutputStream#closeImpl, if the stream is closed,
> we throw an exception if any of the striped streamers had an exception:
> {code}
> protected synchronized void closeImpl() throws IOException {
> if (isClosed()) {
> final MultipleIOException.Builder b = new MultipleIOException.Builder();
> for(int i = 0; i < streamers.size(); i++) {
> final StripedDataStreamer si = getStripedDataStreamer(i);
> try {
> si.getLastException().check(true);
> } catch (IOException e) {
> b.add(e);
> }
> }
> final IOException ioe = b.build();
> if (ioe != null) {
> throw ioe;
> }
> return;
> }
> {code}
> I think this is incorrect, since we only need to throw in this situation if
> we have too many failed streamers. close should also be idempotent, so it
> should throw the first time we call close if it's going to throw at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-12612) DFSStripedOutputStream#close will throw if called a second time with a failed streamer
[
https://issues.apache.org/jira/browse/HDFS-12612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei (Eddy) Xu updated HDFS-12612:
-
Attachment: HDFS-12612.03.patch
Fix checkstyle and findbug warnings.
> DFSStripedOutputStream#close will throw if called a second time with a failed
> streamer
> --
>
> Key: HDFS-12612
> URL: https://issues.apache.org/jira/browse/HDFS-12612
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
>Affects Versions: 3.0.0-beta1
>Reporter: Andrew Wang
>Assignee: Lei (Eddy) Xu
> Labels: hdfs-ec-3.0-must-do
> Attachments: HDFS-12612.00.patch, HDFS-12612.01.patch,
> HDFS-12612.02.patch, HDFS-12612.03.patch
>
>
> Found while testing with Hive. We have a cluster with 2 DNs and the XOR-2-1
> policy. If you write a file and call close() twice, it throws this exception:
> {noformat}
> 17/10/04 16:02:14 WARN hdfs.DFSOutputStream: Cannot allocate parity
> block(index=2, policy=XOR-2-1-1024k). Not enough datanodes? Exclude nodes=[]
> ...
> Caused by: java.io.IOException: Failed to get parity block, index=2
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.allocateNewBlock(DFSStripedOutputStream.java:500)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.writeChunk(DFSStripedOutputStream.java:524)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> {noformat}
> This is because in DFSStripedOutputStream#closeImpl, if the stream is closed,
> we throw an exception if any of the striped streamers had an exception:
> {code}
> protected synchronized void closeImpl() throws IOException {
> if (isClosed()) {
> final MultipleIOException.Builder b = new MultipleIOException.Builder();
> for(int i = 0; i < streamers.size(); i++) {
> final StripedDataStreamer si = getStripedDataStreamer(i);
> try {
> si.getLastException().check(true);
> } catch (IOException e) {
> b.add(e);
> }
> }
> final IOException ioe = b.build();
> if (ioe != null) {
> throw ioe;
> }
> return;
> }
> {code}
> I think this is incorrect, since we only need to throw in this situation if
> we have too many failed streamers. close should also be idempotent, so it
> should throw the first time we call close if it's going to throw at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-12612) DFSStripedOutputStream#close will throw if called a second time with a failed streamer
[
https://issues.apache.org/jira/browse/HDFS-12612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei (Eddy) Xu updated HDFS-12612:
-
Attachment: HDFS-12612.02.patch
Thanks for the suggestions [~andrew.wang]
Updated to fix findbugs and checkstyle warnings, and also extract the
{{LastException}} to a new class {{ExceptionLastSeen}} (choosing this name to
avoid findbugs warning).
> DFSStripedOutputStream#close will throw if called a second time with a failed
> streamer
> --
>
> Key: HDFS-12612
> URL: https://issues.apache.org/jira/browse/HDFS-12612
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
>Affects Versions: 3.0.0-beta1
>Reporter: Andrew Wang
>Assignee: Lei (Eddy) Xu
> Labels: hdfs-ec-3.0-must-do
> Attachments: HDFS-12612.00.patch, HDFS-12612.01.patch,
> HDFS-12612.02.patch
>
>
> Found while testing with Hive. We have a cluster with 2 DNs and the XOR-2-1
> policy. If you write a file and call close() twice, it throws this exception:
> {noformat}
> 17/10/04 16:02:14 WARN hdfs.DFSOutputStream: Cannot allocate parity
> block(index=2, policy=XOR-2-1-1024k). Not enough datanodes? Exclude nodes=[]
> ...
> Caused by: java.io.IOException: Failed to get parity block, index=2
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.allocateNewBlock(DFSStripedOutputStream.java:500)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.writeChunk(DFSStripedOutputStream.java:524)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> {noformat}
> This is because in DFSStripedOutputStream#closeImpl, if the stream is closed,
> we throw an exception if any of the striped streamers had an exception:
> {code}
> protected synchronized void closeImpl() throws IOException {
> if (isClosed()) {
> final MultipleIOException.Builder b = new MultipleIOException.Builder();
> for(int i = 0; i < streamers.size(); i++) {
> final StripedDataStreamer si = getStripedDataStreamer(i);
> try {
> si.getLastException().check(true);
> } catch (IOException e) {
> b.add(e);
> }
> }
> final IOException ioe = b.build();
> if (ioe != null) {
> throw ioe;
> }
> return;
> }
> {code}
> I think this is incorrect, since we only need to throw in this situation if
> we have too many failed streamers. close should also be idempotent, so it
> should throw the first time we call close if it's going to throw at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-12612) DFSStripedOutputStream#close will throw if called a second time with a failed streamer
[
https://issues.apache.org/jira/browse/HDFS-12612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei (Eddy) Xu updated HDFS-12612:
-
Attachment: HDFS-12612.01.patch
Discussed offline with Andrew. Now we consider if there are more than
{{dataUnits}} streamer success, then the {{OutputStream}} is success. Also
move the {{lastException}} to {{DFSStripedOutputStream}} that can be set by
{{abort()}}.
> DFSStripedOutputStream#close will throw if called a second time with a failed
> streamer
> --
>
> Key: HDFS-12612
> URL: https://issues.apache.org/jira/browse/HDFS-12612
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
>Affects Versions: 3.0.0-beta1
>Reporter: Andrew Wang
>Assignee: Lei (Eddy) Xu
> Labels: hdfs-ec-3.0-must-do
> Attachments: HDFS-12612.00.patch, HDFS-12612.01.patch
>
>
> Found while testing with Hive. We have a cluster with 2 DNs and the XOR-2-1
> policy. If you write a file and call close() twice, it throws this exception:
> {noformat}
> 17/10/04 16:02:14 WARN hdfs.DFSOutputStream: Cannot allocate parity
> block(index=2, policy=XOR-2-1-1024k). Not enough datanodes? Exclude nodes=[]
> ...
> Caused by: java.io.IOException: Failed to get parity block, index=2
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.allocateNewBlock(DFSStripedOutputStream.java:500)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.writeChunk(DFSStripedOutputStream.java:524)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> {noformat}
> This is because in DFSStripedOutputStream#closeImpl, if the stream is closed,
> we throw an exception if any of the striped streamers had an exception:
> {code}
> protected synchronized void closeImpl() throws IOException {
> if (isClosed()) {
> final MultipleIOException.Builder b = new MultipleIOException.Builder();
> for(int i = 0; i < streamers.size(); i++) {
> final StripedDataStreamer si = getStripedDataStreamer(i);
> try {
> si.getLastException().check(true);
> } catch (IOException e) {
> b.add(e);
> }
> }
> final IOException ioe = b.build();
> if (ioe != null) {
> throw ioe;
> }
> return;
> }
> {code}
> I think this is incorrect, since we only need to throw in this situation if
> we have too many failed streamers. close should also be idempotent, so it
> should throw the first time we call close if it's going to throw at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-12612) DFSStripedOutputStream#close will throw if called a second time with a failed streamer
[
https://issues.apache.org/jira/browse/HDFS-12612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei (Eddy) Xu updated HDFS-12612:
-
Status: Patch Available (was: Open)
> DFSStripedOutputStream#close will throw if called a second time with a failed
> streamer
> --
>
> Key: HDFS-12612
> URL: https://issues.apache.org/jira/browse/HDFS-12612
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
>Affects Versions: 3.0.0-beta1
>Reporter: Andrew Wang
>Assignee: Lei (Eddy) Xu
> Labels: hdfs-ec-3.0-must-do
> Attachments: HDFS-12612.00.patch
>
>
> Found while testing with Hive. We have a cluster with 2 DNs and the XOR-2-1
> policy. If you write a file and call close() twice, it throws this exception:
> {noformat}
> 17/10/04 16:02:14 WARN hdfs.DFSOutputStream: Cannot allocate parity
> block(index=2, policy=XOR-2-1-1024k). Not enough datanodes? Exclude nodes=[]
> ...
> Caused by: java.io.IOException: Failed to get parity block, index=2
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.allocateNewBlock(DFSStripedOutputStream.java:500)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.writeChunk(DFSStripedOutputStream.java:524)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> {noformat}
> This is because in DFSStripedOutputStream#closeImpl, if the stream is closed,
> we throw an exception if any of the striped streamers had an exception:
> {code}
> protected synchronized void closeImpl() throws IOException {
> if (isClosed()) {
> final MultipleIOException.Builder b = new MultipleIOException.Builder();
> for(int i = 0; i < streamers.size(); i++) {
> final StripedDataStreamer si = getStripedDataStreamer(i);
> try {
> si.getLastException().check(true);
> } catch (IOException e) {
> b.add(e);
> }
> }
> final IOException ioe = b.build();
> if (ioe != null) {
> throw ioe;
> }
> return;
> }
> {code}
> I think this is incorrect, since we only need to throw in this situation if
> we have too many failed streamers. close should also be idempotent, so it
> should throw the first time we call close if it's going to throw at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-12612) DFSStripedOutputStream#close will throw if called a second time with a failed streamer
[
https://issues.apache.org/jira/browse/HDFS-12612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei (Eddy) Xu updated HDFS-12612:
-
Attachment: HDFS-12612.00.patch
Add a test to verify the bug, and only logs remaining IOE from {{streams}}
after close() being called.
> DFSStripedOutputStream#close will throw if called a second time with a failed
> streamer
> --
>
> Key: HDFS-12612
> URL: https://issues.apache.org/jira/browse/HDFS-12612
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: erasure-coding
>Affects Versions: 3.0.0-beta1
>Reporter: Andrew Wang
>Assignee: Lei (Eddy) Xu
> Labels: hdfs-ec-3.0-must-do
> Attachments: HDFS-12612.00.patch
>
>
> Found while testing with Hive. We have a cluster with 2 DNs and the XOR-2-1
> policy. If you write a file and call close() twice, it throws this exception:
> {noformat}
> 17/10/04 16:02:14 WARN hdfs.DFSOutputStream: Cannot allocate parity
> block(index=2, policy=XOR-2-1-1024k). Not enough datanodes? Exclude nodes=[]
> ...
> Caused by: java.io.IOException: Failed to get parity block, index=2
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.allocateNewBlock(DFSStripedOutputStream.java:500)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> at
> org.apache.hadoop.hdfs.DFSStripedOutputStream.writeChunk(DFSStripedOutputStream.java:524)
> ~[hadoop-hdfs-client-3.0.0-alpha3-cdh6.x-SNAPSHOT.jar:?]
> {noformat}
> This is because in DFSStripedOutputStream#closeImpl, if the stream is closed,
> we throw an exception if any of the striped streamers had an exception:
> {code}
> protected synchronized void closeImpl() throws IOException {
> if (isClosed()) {
> final MultipleIOException.Builder b = new MultipleIOException.Builder();
> for(int i = 0; i < streamers.size(); i++) {
> final StripedDataStreamer si = getStripedDataStreamer(i);
> try {
> si.getLastException().check(true);
> } catch (IOException e) {
> b.add(e);
> }
> }
> final IOException ioe = b.build();
> if (ioe != null) {
> throw ioe;
> }
> return;
> }
> {code}
> I think this is incorrect, since we only need to throw in this situation if
> we have too many failed streamers. close should also be idempotent, so it
> should throw the first time we call close if it's going to throw at all.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
-
To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
