[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13526874#comment-13526874 ] Jonathan Hsieh commented on HBASE-6353: --- nit: There is a enable 'test-table' command missing after the restore_snapshot 'test-snapshot' command the the test script > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Fix For: hbase-6055 > > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch, > snapshot-offline-test-demo.shell > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * clone_snapshot > ** Create a new table based on the snapshot content > * list_snapshots [filter] > ** Show the list of snapshots taken > * delete_snapshot > ** Remove a specified snapshot > * rename_snapshot > ** Rename a snapshot to a new name > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > This allows a full and quick rollback to a previous snapshot. > h6. Clone from Snapshot > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Add/Remove data from this table without affecting the original one or > the snapshot. > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13492409#comment-13492409 ] Jesse Yates commented on HBASE-6353: yeah, this should be good for the moment. I'll roll it in. > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch, > snapshot-offline-test-demo.shell > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * clone_snapshot > ** Create a new table based on the snapshot content > * list_snapshots [filter] > ** Show the list of snapshots taken > * delete_snapshot > ** Remove a specified snapshot > * rename_snapshot > ** Rename a snapshot to a new name > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > This allows a full and quick rollback to a previous snapshot. > h6. Clone from Snapshot > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Add/Remove data from this table without affecting the original one or > the snapshot. > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[
https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13492019#comment-13492019
]
Hadoop QA commented on HBASE-6353:
--
{color:red}-1 overall{color}. Here are the results of testing the latest
attachment
http://issues.apache.org/jira/secure/attachment/12551976/snapshot-offline-test-demo.shell
against trunk revision .
{color:green}+1 @author{color}. The patch does not contain any @author
tags.
{color:red}-1 tests included{color}. The patch doesn't appear to include
any new or modified tests.
Please justify why no new tests are needed for this
patch.
Also please list what manual steps were performed to
verify this patch.
{color:red}-1 patch{color}. The patch command could not apply the patch.
Console output:
https://builds.apache.org/job/PreCommit-HBASE-Build/3244//console
This message is automatically generated.
> Snapshots shell
> ---
>
> Key: HBASE-6353
> URL: https://issues.apache.org/jira/browse/HBASE-6353
> Project: HBase
> Issue Type: Sub-task
> Components: shell
>Reporter: Matteo Bertozzi
>Assignee: Matteo Bertozzi
> Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch,
> snapshot-offline-test-demo.shell
>
>
> h6. hbase shell with snapshot commands
> * snapshot
> ** Take a snapshot of the specified name with the specified name
> * restore_snapshot
> ** Restore specified snapshot on the original table
> * clone_snapshot
> ** Create a new table based on the snapshot content
> * list_snapshots [filter]
> ** Show the list of snapshots taken
> * delete_snapshot
> ** Remove a specified snapshot
> * rename_snapshot
> ** Rename a snapshot to a new name
> h6. Restore Table
> Given a "snapshot name" restore override the original table with the snapshot
> content.
> Before restoring a new snapshot of the table is taken, just to avoid bad
> situations.
> This allows a full and quick rollback to a previous snapshot.
> h6. Clone from Snapshot
> Given a "snapshot name" a new table is created with the content of the
> specified snapshot.
> This operation allows:
> * To have an old version of the table in parallel with the current one.
> ** Look at snapshot side-by-side with the "current" before making the
> decision whether to roll back or not
> * To Add/Remove data from this table without affecting the original one or
> the snapshot.
> * To Restore only "individual items" (only some small range of data was lost
> from "current")
> ** MR job that scan the cloned table and update the data in the original
> one. (Partial restore of the data)
> h6. Open points
> * Add snapshot type option on take snapshot command (global, timestamp)?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[
https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13491869#comment-13491869
]
Hadoop QA commented on HBASE-6353:
--
{color:red}-1 overall{color}. Here are the results of testing the latest
attachment
http://issues.apache.org/jira/secure/attachment/12551976/snapshot-offline-test-demo.shell
against trunk revision .
{color:green}+1 @author{color}. The patch does not contain any @author
tags.
{color:red}-1 tests included{color}. The patch doesn't appear to include
any new or modified tests.
Please justify why no new tests are needed for this
patch.
Also please list what manual steps were performed to
verify this patch.
{color:red}-1 patch{color}. The patch command could not apply the patch.
Console output:
https://builds.apache.org/job/PreCommit-HBASE-Build/3236//console
This message is automatically generated.
> Snapshots shell
> ---
>
> Key: HBASE-6353
> URL: https://issues.apache.org/jira/browse/HBASE-6353
> Project: HBase
> Issue Type: Sub-task
> Components: shell
>Reporter: Matteo Bertozzi
>Assignee: Matteo Bertozzi
> Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch,
> snapshot-offline-test-demo.shell
>
>
> h6. hbase shell with snapshot commands
> * snapshot
> ** Take a snapshot of the specified name with the specified name
> * restore_snapshot
> ** Restore specified snapshot on the original table
> * clone_snapshot
> ** Create a new table based on the snapshot content
> * list_snapshots [filter]
> ** Show the list of snapshots taken
> * delete_snapshot
> ** Remove a specified snapshot
> * rename_snapshot
> ** Rename a snapshot to a new name
> h6. Restore Table
> Given a "snapshot name" restore override the original table with the snapshot
> content.
> Before restoring a new snapshot of the table is taken, just to avoid bad
> situations.
> This allows a full and quick rollback to a previous snapshot.
> h6. Clone from Snapshot
> Given a "snapshot name" a new table is created with the content of the
> specified snapshot.
> This operation allows:
> * To have an old version of the table in parallel with the current one.
> ** Look at snapshot side-by-side with the "current" before making the
> decision whether to roll back or not
> * To Add/Remove data from this table without affecting the original one or
> the snapshot.
> * To Restore only "individual items" (only some small range of data was lost
> from "current")
> ** MR job that scan the cloned table and update the data in the original
> one. (Partial restore of the data)
> h6. Open points
> * Add snapshot type option on take snapshot command (global, timestamp)?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[
https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13491862#comment-13491862
]
Hadoop QA commented on HBASE-6353:
--
{color:red}-1 overall{color}. Here are the results of testing the latest
attachment
http://issues.apache.org/jira/secure/attachment/12551976/snapshot-offline-test-demo.shell
against trunk revision .
{color:green}+1 @author{color}. The patch does not contain any @author
tags.
{color:red}-1 tests included{color}. The patch doesn't appear to include
any new or modified tests.
Please justify why no new tests are needed for this
patch.
Also please list what manual steps were performed to
verify this patch.
{color:red}-1 patch{color}. The patch command could not apply the patch.
Console output:
https://builds.apache.org/job/PreCommit-HBASE-Build/3235//console
This message is automatically generated.
> Snapshots shell
> ---
>
> Key: HBASE-6353
> URL: https://issues.apache.org/jira/browse/HBASE-6353
> Project: HBase
> Issue Type: Sub-task
> Components: shell
>Reporter: Matteo Bertozzi
>Assignee: Matteo Bertozzi
> Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch,
> snapshot-offline-test-demo.shell
>
>
> h6. hbase shell with snapshot commands
> * snapshot
> ** Take a snapshot of the specified name with the specified name
> * restore_snapshot
> ** Restore specified snapshot on the original table
> * clone_snapshot
> ** Create a new table based on the snapshot content
> * list_snapshots [filter]
> ** Show the list of snapshots taken
> * delete_snapshot
> ** Remove a specified snapshot
> * rename_snapshot
> ** Rename a snapshot to a new name
> h6. Restore Table
> Given a "snapshot name" restore override the original table with the snapshot
> content.
> Before restoring a new snapshot of the table is taken, just to avoid bad
> situations.
> This allows a full and quick rollback to a previous snapshot.
> h6. Clone from Snapshot
> Given a "snapshot name" a new table is created with the content of the
> specified snapshot.
> This operation allows:
> * To have an old version of the table in parallel with the current one.
> ** Look at snapshot side-by-side with the "current" before making the
> decision whether to roll back or not
> * To Add/Remove data from this table without affecting the original one or
> the snapshot.
> * To Restore only "individual items" (only some small range of data was lost
> from "current")
> ** MR job that scan the cloned table and update the data in the original
> one. (Partial restore of the data)
> h6. Open points
> * Add snapshot type option on take snapshot command (global, timestamp)?
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13491858#comment-13491858 ] Jonathan Hsieh commented on HBASE-6353: --- Jesse, where do we stand on this? Do we need more or is this good to go? > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch, > snapshot-offline-test-demo.shell > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * clone_snapshot > ** Create a new table based on the snapshot content > * list_snapshots [filter] > ** Show the list of snapshots taken > * delete_snapshot > ** Remove a specified snapshot > * rename_snapshot > ** Rename a snapshot to a new name > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > This allows a full and quick rollback to a previous snapshot. > h6. Clone from Snapshot > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Add/Remove data from this table without affecting the original one or > the snapshot. > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487925#comment-13487925 ] Jonathan Hsieh commented on HBASE-6353: --- I don't think the ruby stuff fits into a unit test nicely (or the effort to do so would be a huge project unto itself). I agree with the testing concern but feel that the shell script / hbase script and documentation on how to run it here in the jira would be sufficient. > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487917#comment-13487917 ] Jesse Yates commented on HBASE-6353: [~jmhsieh] doesn't mean we shouldn't test it :) Any way the tests can be pulled into real tests? > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487911#comment-13487911 ] Jonathan Hsieh commented on HBASE-6353: --- Hey Matteo, there was a test script that executed the commands attached to the review. Attach a copy of it to this jira so that other folks can try it? Jesse -- do you think that is sufficient? > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487912#comment-13487912 ] Jonathan Hsieh commented on HBASE-6353: --- (Most of the shell code in hbase isn't really tested sadly). > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487893#comment-13487893 ] Jesse Yates commented on HBASE-6353: Last minute realization - do we need to add some tests for this? I don't think they would be very telling - it would mostly be mocks of the underlying objects (e.g. a mock HBaseAdmin) and that they get a call (e.g. admin.snapshot) - but figured it would be prudent to ask what everyone else thought > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch, HBASE-6353-v1.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487409#comment-13487409 ] Jesse Yates commented on HBASE-6353: Yeah. Matteo - do you want to attach the latest version and I'll roll it in? > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13487401#comment-13487401 ] Jonathan Hsieh commented on HBASE-6353: --- Hey Jesse, I believe this got all +1's on the review. Can we commit this to the snapshot branch so this isn't hanging around anymore? https://reviews.apache.org/r/7583/ > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: Sub-task > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13410689#comment-13410689 ] Jesse Yates commented on HBASE-6353: general comments: * 'take snapshot' should have an type field (maybe -t option since timestamp is the default). * we should probably slow down on this until the restore operations are in the java code first. Makes the shell super easy and generally how people roll in stuff (also helps keep stuff logically separate). bq. What do you think about having mount and restore as separate feature? I feel like it should track the outline from HBASE-6230 in terms of naming. See my latest comments (https://issues.apache.org/jira/browse/HBASE-6230?focusedCommentId=13410682&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-13410682) for thoughts on how the operations could also be named and a little bit of thought on how they could work. > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: New Feature > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13409204#comment-13409204 ] Matteo Bertozzi commented on HBASE-6353: This patch is just an initial draft to describe and discuss the snapshot operations. @Lars Yes, the patch should be just the ruby files for the shell. There's also the protobuf + HMasterInterface + HBaseAdmin since you need to be able to run it, but once we've HBASE-6055 all this code will not be here. @Jesse yes restore code will not be here, and I agree to add restore to the table class What do you think about having mount and restore as separate feature? From a code point of view are almost the same, but for the user is a different operation/use case. > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: New Feature > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13409162#comment-13409162 ] Lars Hofhansl commented on HBASE-6353: -- To be fair, without the generated protobuf stuff the patch is "only" 47k. Still big, and no tests. A feature like this should have tests. > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: New Feature > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HBASE-6353) Snapshots shell
[ https://issues.apache.org/jira/browse/HBASE-6353?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13409062#comment-13409062 ] Jesse Yates commented on HBASE-6353: Woah, that patch is massive :) Do you think you can break it into two jiras? The shell stuff should be here, but not the restore functionality which you seem to be adding. Also, those tickets should probably be dependencies, rather than relates (so you don't need to include all the restore/snapshot code here). Can you add the features such that you can also do it from a table reference (see the shell's Table class for adding the admin functions to a table). Good stuff Matteo. > Snapshots shell > --- > > Key: HBASE-6353 > URL: https://issues.apache.org/jira/browse/HBASE-6353 > Project: HBase > Issue Type: New Feature > Components: shell >Reporter: Matteo Bertozzi >Assignee: Matteo Bertozzi > Attachments: HBASE-6353-v0.patch > > > h6. hbase shell with snapshot commands > * snapshot > ** Take a snapshot of the specified name with the specified name > * restore_snapshot > ** Restore specified snapshot on the original table > * mount_snapshot [readonly] > ** Load the snapshot data as specified table (optional readonly flag) > * list_snapshots [filter] > ** Show a list of snapshots > * delete_snapshot > ** Remove a specified snapshot > h6. Restore Table > Given a "snapshot name" restore override the original table with the snapshot > content. > Before restoring a new snapshot of the table is taken, just to avoid bad > situations. > (If the table is not disabled we can keep serving reads) > This allows a full and quick rollback to a previous snapshot. > h6. Mount Table (Aka Clone Table) > Given a "snapshot name" a new table is created with the content of the > specified snapshot. > This operation allows: > * To have an old version of the table in parallel with the current one. > ** Look at snapshot side-by-side with the "current" before making the > decision whether to roll back or not > * To Restore only "individual items" (only some small range of data was lost > from "current") > ** MR job that scan the cloned table and update the data in the original > one. (Partial restore of the data) > * if the table is not marked as read-only > ** To Add/Remove data from this table without affecting the original one or > the snapshot. > h6. Open points > * Add snapshot type option on take snapshot command (global, timestamp)? > * Keep separate the "restore" from "mount"? -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa For more information on JIRA, see: http://www.atlassian.com/software/jira
