[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Sean Busbey updated HBASE-14497: Fix Version/s: 1.2.7 > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun >Priority: Major > Fix For: 2.0.0, 1.3.0, 0.98.16, 1.2.7 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1-v6.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v4.patch, > HBASE-14497-master-v5.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v7.6.3#76005)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Resolution: Fixed Fix Version/s: 0.98.16 Status: Resolved (was: Patch Available) That test was from hdfs test suite - not related. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0, 1.3.0, 0.98.16 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1-v6.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v4.patch, > HBASE-14497-master-v5.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Anonymous updated HBASE-14497: -- Status: Patch Available (was: Reopened) > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 0.98.14, 2.0.0, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0, 1.3.0 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1-v6.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v4.patch, > HBASE-14497-master-v5.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-0.98-v6.patch Patch for 0.98 branch. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0, 1.3.0 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1-v6.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v4.patch, > HBASE-14497-master-v5.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Resolution: Fixed Status: Resolved (was: Patch Available) > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0, 1.3.0 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98.patch, HBASE-14497-branch-1-v2.patch, > HBASE-14497-branch-1-v3.patch, HBASE-14497-branch-1-v6.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master-v5.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Fix Version/s: 1.3.0 > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0, 1.3.0 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98.patch, HBASE-14497-branch-1-v2.patch, > HBASE-14497-branch-1-v3.patch, HBASE-14497-branch-1-v6.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master-v5.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Attachment: 14497-branch-1-v6.patch > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0 > > Attachments: 14497-branch-1-v6.patch, 14497-master-v6.patch, > HBASE-14497-0.98.patch, HBASE-14497-branch-1-v2.patch, > HBASE-14497-branch-1-v3.patch, HBASE-14497-branch-1-v6.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master-v5.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-branch-1-v6.patch Re-post patch for branch-1 to QA run. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0 > > Attachments: 14497-master-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1-v6.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v4.patch, > HBASE-14497-master-v5.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Hadoop Flags: Reviewed Fix Version/s: 2.0.0 > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun >Assignee: Yerui Sun > Fix For: 2.0.0 > > Attachments: 14497-master-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master-v5.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-branch-1-v3.patch Here's the v3 patch for branch-1. Updated based on master v6 patch. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: 14497-master-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1-v3.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master-v5.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Attachment: 14497-master-v6.patch Patch v6 closes region at the end of testReverseScanner_StackOverflow Please prepare patch for branch-1 > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: 14497-master-v6.patch, HBASE-14497-0.98.patch, > HBASE-14497-branch-1-v2.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v4.patch, > HBASE-14497-master-v5.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-master-v5.patch A strictly line length rules... Just wrap long comment lines based on master branch v4 patch. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1-v2.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master-v5.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-master-v4.patch Here's the v4 patch for master branch. Also fix the same problem on MemstoreScanner. [~yuzhih...@gmail.com],[~zjushch], would you please review this? Thanks. I'll post the patches for branch-1 and 0.98 later, after Hadoop QA run and passed this patch. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1-v2.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v4.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-branch-1-v2.patch Attach patch for branch-1 for QA run. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1-v2.patch, > HBASE-14497-branch-1.patch, HBASE-14497-master-v2.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-master-v3.patch re-attach patch for QA run again. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master-v3.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-master-v3.patch Rename recusiveSeek to keepingSeek. Fix zombie tests. Wrap long code lines. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master-v3.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-master-v2.patch Sorry for the compiling error. I have fix it and repost patch for master branch. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master-v2.patch, HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Ted Yu updated HBASE-14497: --- Status: Patch Available (was: Open) > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 0.98.14, 2.0.0, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-master.patch HBASE-14497-branch-1.patch The patches for branch-1 and master branch. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch, HBASE-14497-branch-1.patch, > HBASE-14497-master.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
[jira] [Updated] (HBASE-14497) Reverse Scan threw StackOverflow caused by readPt checking
[ https://issues.apache.org/jira/browse/HBASE-14497?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Yerui Sun updated HBASE-14497: -- Attachment: HBASE-14497-0.98.patch Here's the patch for 0.98 branch. Including a unit test that reproduced the error. > Reverse Scan threw StackOverflow caused by readPt checking > -- > > Key: HBASE-14497 > URL: https://issues.apache.org/jira/browse/HBASE-14497 > Project: HBase > Issue Type: Bug >Affects Versions: 2.0.0, 0.98.14, 1.3.0 >Reporter: Yerui Sun > Attachments: HBASE-14497-0.98.patch > > > I met stack overflow error in StoreFileScanner.seekToPreviousRow using > reversed scan. I searched and founded HBASE-14155, but it seems to be a > different reason. > The seekToPreviousRow will fetch the row which closest before, and compare > mvcc to the readPt, which acquired when scanner created. If the row's mvcc is > bigger than readPt, an recursive call of seekToPreviousRow will invoked, to > find the next closest before row. > Considering we created a scanner for reversed scan, and some data with > smaller rows was written and flushed, before calling scanner next. When > seekToPreviousRow was invoked, it would call itself recursively, until all > rows which written after scanner created were iterated. The depth of > recursive calling stack depends on the count of rows, the stack overflow > error will be threw if the count of rows is large, like 1. -- This message was sent by Atlassian JIRA (v6.3.4#6332)
