[jira] [Created] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
Yao Guangdong created HDFS-15186: Summary: Erasure Coding: Decommission may generate the parity block's content with all 0 in some case Key: HDFS-15186 URL: https://issues.apache.org/jira/browse/HDFS-15186 Project: Hadoop HDFS Issue Type: Bug Components: datanode, erasure-coding Affects Versions: 3.1.3, 3.2.1, 3.0.3 Reporter: Yao Guangdong Fix For: 3.0.4, 3.3.0, 3.1.4, 3.2.2 I can find some parity block's content with all 0 when i decommission some DataNode(more than 1) from a cluster. And the probability is very big(parts per thousand).This is a big problem.You can think that if we read data from the zero parity block or use the zero parity block to recover a block which can make us use the error data even we don't know it. There is some case in the below: B: Busy DataNode, D:Decommissioning DataNode, Others is normal. 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)], the DN may received reconstruct block command and the liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in the class StripedReconstructionInfo) length is 2. The targets's length is 2 which mean that the DataNode need recover 2 internal block in current code.But from the liveIndices we only can find 1 missing block, so the method StripedWriter#initTargetIndices will use 0 as the default recover block and don't care the indices 0 is in the sources indices or not. When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] use the ec algorithm.We can find that the indices [0] is in the both the sources indices and the targets indices in this case. The returned target buffer in the indices [6] is always 0 from the ec algorithm.So I think this is the ec algorithm's problem. Because it should more fault tolerance.I try to fixed it .But it is too hard. Because the case is too more. The second is another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm with a correct parameters. Which mean that remove the duplicate target indices 0 in this case.Finally, I fixed it in this way. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Fix Version/s: (was: 3.2.2) (was: 3.1.4) (was: 3.0.4) > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.2 >Reporter: Yao Guangdong >Priority: Critical > Fix For: 3.3.0 > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Affects Version/s: (was: 3.1.3) (was: 3.2.1) (was: 3.0.3) 3.0.2 > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.2 >Reporter: Yao Guangdong >Priority: Critical > Fix For: 3.0.4, 3.3.0, 3.1.4, 3.2.2 > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Affects Version/s: (was: 3.0.2) 3.0.3 3.2.1 3.1.3 > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Priority: Critical > Fix For: 3.3.0 > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Attachment: HDFS-15186.001.patch > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Priority: Critical > Fix For: 3.3.0 > > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17040794#comment-17040794 ] Yao Guangdong commented on HDFS-15186: -- Added the initial patch. Please check it once. [~weichiu] PTAL > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Priority: Critical > Fix For: 3.3.0 > > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17042472#comment-17042472 ] Yao Guangdong commented on HDFS-15186: -- [~ayushtkn], Yes, you are right. The major reason of the problem is from namenode.We also need to check the logic at HDFS-14768. But, as i said in the description, the case is too much, even some we don't know it. The data is very important in the production. We can't accept any data errors.So i think we could add this patch enhance the data security.And i also suggest that we need to increase the fault tolerance for ec algorithm. What do you think? > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17043133#comment-17043133 ] Yao Guangdong commented on HDFS-15186: -- [~ferhui], [~ayushtkn], [~gjhkael] Thanks for yours patient review. I agree with yours point that fix it in namenode side. I have a suspicion. We can copy blocks from others DN in 3 replica mode when we decommission DN and the decommissioning DN is busy. But, we only can copy blocks from the decommissioning DN in ec mode if we don't reconstruct it. The time we cost in decommission is 69 hours (100W / 4 / 3600 = 69 hours) if we have 100W blocks in one DN and the cost time we copy a block is one second and the hard limit is default 4. Which will make the speed of decommission very slow if we copy all blocks from decommission DN and we add decommissioning busy replica into live replica check? Is my comprehend right? > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17043286#comment-17043286 ] Yao Guangdong commented on HDFS-15186: -- [~ayushtkn], OK. Thanks for your reply. [~gjhkael] had been fixed it in namenode side by HDFS-14768 . I think this is duplicated. You can close it. > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17043286#comment-17043286 ] Yao Guangdong edited comment on HDFS-15186 at 2/24/20 10:06 AM: [~ayushtkn], OK. Thanks for your advice. [~gjhkael] had been fixed it in namenode side by HDFS-14768 . I think this is duplicated. You can close it. was (Author: yaoguangdong): [~ayushtkn], OK. Thanks for your reply. [~gjhkael] had been fixed it in namenode side by HDFS-14768 . I think this is duplicated. You can close it. > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17043360#comment-17043360 ] Yao Guangdong commented on HDFS-15186: -- [~ayushtkn], Thank you very much. I will fix it as soon as possible. :D > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Attachment: HDFS-15186.002.patch > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17044085#comment-17044085 ] Yao Guangdong commented on HDFS-15186: -- [~ferhui], [~ayushtkn], [~gjhkael], [~marvelrock], I have been fixed it in namenode side and add a new patch HDFS-15186.002.patch. Could you have time to review it. Thanks. > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Comment Edited] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17043133#comment-17043133 ] Yao Guangdong edited comment on HDFS-15186 at 2/25/20 3:08 AM: --- [~ferhui], [~ayushtkn], [~gjhkael] Thanks for yours patient review. I agree with yours point that fix it in namenode side. I have a suspicion. We can copy blocks from others DN in 3 replica mode when we decommission DN and the decommissioning DN is busy. But, we only can copy blocks from the decommissioning DN in ec mode if we don't reconstruct it. The time we cost in decommission is 69 hours (1m / 4 / 3600 = 69 hours) if we have 1 million blocks in one DN and the cost time we copy a block is one second and the hard limit is default 4. Which will make the speed of decommission very slow if we copy all blocks from decommission DN and we add decommissioning busy replica into live replica check? Is my comprehend right? was (Author: yaoguangdong): [~ferhui], [~ayushtkn], [~gjhkael] Thanks for yours patient review. I agree with yours point that fix it in namenode side. I have a suspicion. We can copy blocks from others DN in 3 replica mode when we decommission DN and the decommissioning DN is busy. But, we only can copy blocks from the decommissioning DN in ec mode if we don't reconstruct it. The time we cost in decommission is 69 hours (100W / 4 / 3600 = 69 hours) if we have 100W blocks in one DN and the cost time we copy a block is one second and the hard limit is default 4. Which will make the speed of decommission very slow if we copy all blocks from decommission DN and we add decommissioning busy replica into live replica check? Is my comprehend right? > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Attachment: HDFS-15186.003.patch > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch, > HDFS-15186.003.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17044148#comment-17044148 ] Yao Guangdong commented on HDFS-15186: -- [~ferhui], Thanks for your review and advice. I have been fixed it. Please check it again. > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch, > HDFS-15186.003.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Attachment: HDFS-15186.004.patch > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch, > HDFS-15186.003.patch, HDFS-15186.004.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17045196#comment-17045196 ] Yao Guangdong commented on HDFS-15186: -- [~ayushtkn], Thanks for your reply. That's weird that I reproduce the issue every time when I use the latest code test it today. Finally, I added the following code from HDFS-14768 to ensure the busy DN shouldn't be reconstructed. Then I added a new path HDFS-15186.004.patch. Please check it again. Thanks. {code:java} //7. Busy DN shouldn't be reconstructed DatanodeStorageInfo[] newDnStorageInfos = bm.getStorages(firstBlock); Assert.assertEquals("Busy DN shouldn't be reconstructed", dnStorageInfos[busyDNIndex].getStorageID(), newDnStorageInfos[busyDNIndex].getStorageID());{code} > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch, > HDFS-15186.003.patch, HDFS-15186.004.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Updated] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yao Guangdong updated HDFS-15186: - Attachment: HDFS-15186.005.patch > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch, > HDFS-15186.003.patch, HDFS-15186.004.patch, HDFS-15186.005.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15186) Erasure Coding: Decommission may generate the parity block's content with all 0 in some case
[ https://issues.apache.org/jira/browse/HDFS-15186?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17046060#comment-17046060 ] Yao Guangdong commented on HDFS-15186: -- [~ayushtkn], Thanks for your remind. The checkstyle warnings have been fixed and updated. Please check it again. Thanks. > Erasure Coding: Decommission may generate the parity block's content with all > 0 in some case > > > Key: HDFS-15186 > URL: https://issues.apache.org/jira/browse/HDFS-15186 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Affects Versions: 3.0.3, 3.2.1, 3.1.3 >Reporter: Yao Guangdong >Assignee: Yao Guangdong >Priority: Critical > Attachments: HDFS-15186.001.patch, HDFS-15186.002.patch, > HDFS-15186.003.patch, HDFS-15186.004.patch, HDFS-15186.005.patch > > > I can find some parity block's content with all 0 when i decommission some > DataNode(more than 1) from a cluster. And the probability is very big(parts > per thousand).This is a big problem.You can think that if we read data from > the zero parity block or use the zero parity block to recover a block which > can make us use the error data even we don't know it. > There is some case in the below: > B: Busy DataNode, > D:Decommissioning DataNode, > Others is normal. > 1.Group indices is [0, 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > 2.Group indices is [0(B,D), 1, 2, 3, 4, 5, 6(B,D), 7, 8(D)]. > > In the first case when the block group indices is [0, 1, 2, 3, 4, 5, 6(B,D), > 7, 8(D)], the DN may received reconstruct block command and the > liveIndices=[0, 1, 2, 3, 4, 5, 7, 8] and the targets's(the field which in > the class StripedReconstructionInfo) length is 2. > The targets's length is 2 which mean that the DataNode need recover 2 > internal block in current code.But from the liveIndices we only can find 1 > missing block, so the method StripedWriter#initTargetIndices will use 0 as > the default recover block and don't care the indices 0 is in the sources > indices or not. > When they use sources indices [0, 1, 2, 3, 4, 5] to recover indices [6, 0] > use the ec algorithm.We can find that the indices [0] is in the both the > sources indices and the targets indices in this case. The returned target > buffer in the indices [6] is always 0 from the ec algorithm.So I think this > is the ec algorithm's problem. Because it should more fault tolerance.I try > to fixed it .But it is too hard. Because the case is too more. The second is > another case in the example above(use sources indices [1, 2, 3, 4, 5, 7] to > recover indices [0, 6, 0]). So I changed my mind.Invoke the ec algorithm > with a correct parameters. Which mean that remove the duplicate target > indices 0 in this case.Finally, I fixed it in this way. > -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org
[jira] [Commented] (HDFS-15240) Erasure Coding: dirty buffer causes reconstruction block error
[ https://issues.apache.org/jira/browse/HDFS-15240?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17067569#comment-17067569 ] Yao Guangdong commented on HDFS-15240: -- [~weichiu] We have the same problem. PTAL. > Erasure Coding: dirty buffer causes reconstruction block error > -- > > Key: HDFS-15240 > URL: https://issues.apache.org/jira/browse/HDFS-15240 > Project: Hadoop HDFS > Issue Type: Bug > Components: datanode, erasure-coding >Reporter: HuangTao >Assignee: HuangTao >Priority: Major > Attachments: HDFS-15240.001.patch > > > When read some lzo files we found some blocks were broken. > I read back all internal blocks(b0-b8) of the block group(RS-6-3-1024k), and > choose 6(b0-b5) blocks to decode other 3(b6', b7', b8') blocks. And find the > longest common sequenece(LCS) between b6' and b6(b7'/b7 and b8'/b8). > After selecting 6 blocks of the block group in combinations one time and > iterating through all cases, I find one case that the length of LCS is the > block length - 64KB, 64KB is just the length of ByteBuffer used by > StripedBlockReader. So the corrupt reconstruction block is made by a dirty > buffer. > The following log snippet is my check program output > {code:java} > decode from [0, 2, 3, 4, 5, 7] -> [1, 6, 8] > Check Block(1) first 131072 bytes longest common substring length 4 > Check Block(6) first 131072 bytes longest common substring length 4 > Check Block(8) first 131072 bytes longest common substring length 4 > decode from [0, 2, 3, 4, 5, 6] -> [1, 7, 8] > Check Block(1) first 131072 bytes longest common substring length 65536 > CHECK AGAIN: Block(1) all 27262976 bytes longest common substring length > 27197440 # this one > Check Block(7) first 131072 bytes longest common substring length 4 > Check Block(8) first 131072 bytes longest common substring length 4{code} > Now I know the dirty buffer causes reconstruction block error, but how does > the dirty buffer come about? > After digging into the code and DN log, I found this following DN log is the > root reason. > {code:java} > [INFO] [stripedRead-1017] : Interrupted while waiting for IO on channel > java.nio.channels.SocketChannel[connected local=/:52586 > remote=/:50010]. 18 millis timeout left. > [WARN] [StripedBlockReconstruction-199] : Failed to reconstruct striped > block: BP-714356632--1519726836856:blk_-YY_3472979393 > java.lang.NullPointerException > at > org.apache.hadoop.hdfs.util.StripedBlockUtil.getNextCompletedStripedRead(StripedBlockUtil.java:314) > at > org.apache.hadoop.hdfs.server.datanode.erasurecode.StripedReader.doReadMinimumSources(StripedReader.java:308) > at > org.apache.hadoop.hdfs.server.datanode.erasurecode.StripedReader.readMinimumSources(StripedReader.java:269) > at > org.apache.hadoop.hdfs.server.datanode.erasurecode.StripedBlockReconstructor.reconstruct(StripedBlockReconstructor.java:94) > at > org.apache.hadoop.hdfs.server.datanode.erasurecode.StripedBlockReconstructor.run(StripedBlockReconstructor.java:60) > at > java.base/java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:515) > at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:264) > at > java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1128) > at > java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:628) > at java.base/java.lang.Thread.run(Thread.java:834) {code} > Reading from DN may timeout(hold by a future(F)) and output the INFO log, but > the futures that contains the future(F) is cleared, > {code:java} > return new StripingChunkReadResult(futures.remove(future), > StripingChunkReadResult.CANCELLED); {code} > futures.remove(future) cause NPE. So the EC reconstruction is failed. In the > finally phase, the code snippet in *getStripedReader().close()* > {code:java} > reconstructor.freeBuffer(reader.getReadBuffer()); > reader.freeReadBuffer(); > reader.closeBlockReader(); {code} > free buffer firstly, but the StripedBlockReader still holds the buffer and > write it. -- This message was sent by Atlassian Jira (v8.3.4#803005) - To unsubscribe, e-mail: hdfs-issues-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-issues-h...@hadoop.apache.org