[jira] [Commented] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179569#comment-17179569
]
Zheng Wang commented on HBASE-24894:
The ut failed since "Valid numbers are [0-23]", then we can not set hour 23 as
offpeak, seems it is a bug?
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179569#comment-17179569
]
Zheng Wang edited comment on HBASE-24894 at 8/18/20, 12:03 PM:
---
The ut failed due to "Valid numbers are [0-23]", then we can not set hour 23
as offpeak, seems it is a bug?
was (Author: filtertip):
The ut failed since "Valid numbers are [0-23]", then we can not set hour 23 as
offpeak, seems it is a bug?
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24709) Support MoveCostFunction use a lower multiplier in offpeak hours
[ https://issues.apache.org/jira/browse/HBASE-24709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179481#comment-17179481 ] Zheng Wang commented on HBASE-24709: The flaky test fixed in HBASE-24894, FYI. [~bharathv] [~ndimiduk] > Support MoveCostFunction use a lower multiplier in offpeak hours > > > Key: HBASE-24709 > URL: https://issues.apache.org/jira/browse/HBASE-24709 > Project: HBase > Issue Type: Improvement > Components: Balancer >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > Attachments: HBASE-24709-jenkins-log > > > Currently the default multiplier of MoveCostFunction is 7, it should be lower > in offpeak hours, thus balancer could do more things. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179463#comment-17179463
]
Zheng Wang edited comment on HBASE-24894 at 8/18/20, 8:53 AM:
--
Oh, the reason is isOffPeakHour does not include the endhour, so if currentHour
is 23, it will fail.
{code:java}
@Override
public boolean isOffPeakHour(int targetHour) {
if (startHour <= endHour) {
return startHour <= targetHour && targetHour < endHour;
}
return targetHour < endHour || startHour <= targetHour;
}
{code}
We can just change this value to 24:
{code:java}
conf.setInt("hbase.offpeak.end.hour",23);{code}
was (Author: filtertip):
Oh, the reason is isOffPeakHour do not include the endhour, so if currentHour
is 23, it will fail.
{code:java}
@Override
public boolean isOffPeakHour(int targetHour) {
if (startHour <= endHour) {
return startHour <= targetHour && targetHour < endHour;
}
return targetHour < endHour || startHour <= targetHour;
}
{code}
We can just change this value to 24:
{code:java}
conf.setInt("hbase.offpeak.end.hour",23);{code}
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179467#comment-17179467
]
Zheng Wang commented on HBASE-24894:
Reviewed that changing almost 2 hours, so sad..
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179463#comment-17179463
]
Zheng Wang edited comment on HBASE-24894 at 8/18/20, 8:43 AM:
--
Oh, the reason is isOffPeakHour do not include the endhour, so if currentHour
is 23, it will fail.
{code:java}
@Override
public boolean isOffPeakHour(int targetHour) {
if (startHour <= endHour) {
return startHour <= targetHour && targetHour < endHour;
}
return targetHour < endHour || startHour <= targetHour;
}
{code}
We can just change this value to 24:
{code:java}
conf.setInt("hbase.offpeak.end.hour",23);{code}
was (Author: filtertip):
Oh, the reason is isOffPeakHour do not include the endhour, so if currentHour
is 23, it will fail.
{code:java}
@Override
public boolean isOffPeakHour(int targetHour) {
if (startHour <= endHour) {
return startHour <= targetHour && targetHour < endHour;
}
return targetHour < endHour || startHour <= targetHour;
}
{code}
We can just change this value to 24:
conf.setInt("hbase.offpeak.end.hour",23);
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179465#comment-17179465
]
Zheng Wang edited comment on HBASE-24894 at 8/18/20, 8:38 AM:
--
Let me fix it, and duplicate HBASE-24848.
was (Author: filtertip):
Let me fix it, and please help to close HBASE-24848. [~vjasani]
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Resolved] (HBASE-24848) Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier
[ https://issues.apache.org/jira/browse/HBASE-24848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang resolved HBASE-24848. Resolution: Duplicate > Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier > - > > Key: HBASE-24848 > URL: https://issues.apache.org/jira/browse/HBASE-24848 > Project: HBase > Issue Type: Improvement > Components: test >Reporter: Zheng Wang >Priority: Major > Attachments: HBASE-24709-jenkins-log > > > This test added by HBASE-24709, and it failed on jdk7 reported by Hudson for > branch-1, but cant reproduce in my local env, so create this jira to keep > tracing. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Assigned] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang reassigned HBASE-24894:
--
Assignee: Zheng Wang
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Assignee: Zheng Wang
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179465#comment-17179465
]
Zheng Wang edited comment on HBASE-24894 at 8/18/20, 8:34 AM:
--
Let me fix it, and please help to close HBASE-24848. [~vjasani]
was (Author: filtertip):
Let me fix it, and close HBASE-24848.
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179465#comment-17179465
]
Zheng Wang edited comment on HBASE-24894 at 8/18/20, 8:32 AM:
--
Let me fix it, and close HBASE-24848.
was (Author: filtertip):
Let me fix it in HBASE-24848.
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24848) Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier
[ https://issues.apache.org/jira/browse/HBASE-24848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24848: --- Summary: Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier (was: Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier for branch-1) > Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier > - > > Key: HBASE-24848 > URL: https://issues.apache.org/jira/browse/HBASE-24848 > Project: HBase > Issue Type: Improvement > Components: test >Affects Versions: 1.7.0 >Reporter: Zheng Wang >Priority: Major > Attachments: HBASE-24709-jenkins-log > > > This test added by HBASE-24709, and it failed on jdk7 reported by Hudson for > branch-1, but cant reproduce in my local env, so create this jira to keep > tracing. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24848) Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier
[ https://issues.apache.org/jira/browse/HBASE-24848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24848: --- Affects Version/s: (was: 1.7.0) > Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier > - > > Key: HBASE-24848 > URL: https://issues.apache.org/jira/browse/HBASE-24848 > Project: HBase > Issue Type: Improvement > Components: test >Reporter: Zheng Wang >Priority: Major > Attachments: HBASE-24709-jenkins-log > > > This test added by HBASE-24709, and it failed on jdk7 reported by Hudson for > branch-1, but cant reproduce in my local env, so create this jira to keep > tracing. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179465#comment-17179465
]
Zheng Wang commented on HBASE-24894:
Let me fix it in HBASE-24848.
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179463#comment-17179463
]
Zheng Wang commented on HBASE-24894:
Oh, the reason is isOffPeakHour do not include the endhour, so if currentHour
is 23, it will fail.
{code:java}
@Override
public boolean isOffPeakHour(int targetHour) {
if (startHour <= endHour) {
return startHour <= targetHour && targetHour < endHour;
}
return targetHour < endHour || startHour <= targetHour;
}
{code}
We can just change this value to 24:
conf.setInt("hbase.offpeak.end.hour",23);
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24894) [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
[
https://issues.apache.org/jira/browse/HBASE-24894?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17179326#comment-17179326
]
Zheng Wang commented on HBASE-24894:
Will dig more today. [~ndimiduk][~vjasani]
> [Flakey Test] TestStochasticLoadBalancer.testMoveCostMultiplier
> ---
>
> Key: HBASE-24894
> URL: https://issues.apache.org/jira/browse/HBASE-24894
> Project: HBase
> Issue Type: Test
> Components: Balancer, master, test
>Affects Versions: 3.0.0-alpha-1, 2.3.0
>Reporter: Nick Dimiduk
>Priority: Major
>
> I've noticed this test has gotten flakey on a couple PRs. The only recent
> change to the class under test is HBASE-24709.
> Failure looks like:
> {noformat}
> [ERROR] Failures:
> [ERROR]
> org.apache.hadoop.hbase.master.balancer.TestStochasticLoadBalancer.testMoveCostMultiplier
> [ERROR] Run 1: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 2: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> [ERROR] Run 3: TestStochasticLoadBalancer.testMoveCostMultiplier:304
> expected:<3.0> but was:<7.0>
> {noformat}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24890) The command regioninfo is not available
[ https://issues.apache.org/jira/browse/HBASE-24890?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24890: --- Description: In hbase shell mode: #input regioninfo '1588230740' #output ERROR: wrong number of arguments (1 for 0) was: input: regioninfo '1588230740' output: ERROR: wrong number of arguments (1 for 0) > The command regioninfo is not available > --- > > Key: HBASE-24890 > URL: https://issues.apache.org/jira/browse/HBASE-24890 > Project: HBase > Issue Type: Bug > Components: shell >Affects Versions: 3.0.0-alpha-1 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > In hbase shell mode: > #input > regioninfo '1588230740' > #output > ERROR: wrong number of arguments (1 for 0) -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24890) The command regioninfo is not available
Zheng Wang created HBASE-24890: -- Summary: The command regioninfo is not available Key: HBASE-24890 URL: https://issues.apache.org/jira/browse/HBASE-24890 Project: HBase Issue Type: Bug Components: shell Affects Versions: 3.0.0-alpha-1 Reporter: Zheng Wang Assignee: Zheng Wang input: regioninfo '1588230740' output: ERROR: wrong number of arguments (1 for 0) -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24875) Remove the force param for unassign since it dose not take effect any more
[ https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24875: --- Description: Currently unassign region in fact only close it, so not need force param any more.(was: The semantic of unassign not correct and the force param does't take effct since the implementation changed at server side) > Remove the force param for unassign since it dose not take effect any more > -- > > Key: HBASE-24875 > URL: https://issues.apache.org/jira/browse/HBASE-24875 > Project: HBase > Issue Type: Improvement > Components: Client >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > Currently unassign region in fact only close it, so not need force param any > more. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24875) Remove the force param for unassign since it dose not take effect any more
[ https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24875: --- Summary: Remove the force param for unassign since it dose not take effect any more (was: The semantic of unassign changed and the force param does't take effect) > Remove the force param for unassign since it dose not take effect any more > -- > > Key: HBASE-24875 > URL: https://issues.apache.org/jira/browse/HBASE-24875 > Project: HBase > Issue Type: Improvement > Components: Client >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > The semantic of unassign not correct and the force param does't take effct > since the implementation changed at server side -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24875) The semantic of unassign changed and the force param does't take effect
[ https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24875: --- Summary: The semantic of unassign changed and the force param does't take effect (was: The semantic of unassign changed and the force param does't take effct) > The semantic of unassign changed and the force param does't take effect > --- > > Key: HBASE-24875 > URL: https://issues.apache.org/jira/browse/HBASE-24875 > Project: HBase > Issue Type: Improvement > Components: Client >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > The semantic of unassign not correct and the force param does't take effct > since the implementation changed at server side -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24875) The semantic of unassign changed and the force param does't take effct
[ https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24875: --- Summary: The semantic of unassign changed and the force param does't take effct (was: The semantic of unassign not correct and the force param does't take effct since the implementation changed at server side) > The semantic of unassign changed and the force param does't take effct > -- > > Key: HBASE-24875 > URL: https://issues.apache.org/jira/browse/HBASE-24875 > Project: HBase > Issue Type: Improvement > Components: Client >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > The semantic of unassign not correct and the force param does't take effct > since the implementation changed at server side -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24875) The semantic of unassign not correct and the force param does't take effct since the implementation changed at server side
[
https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24875:
---
Summary: The semantic of unassign not correct and the force param does't
take effct since the implementation changed at server side (was: The describe
and param for unassign not correct since the implementation changed at server
side)
> The semantic of unassign not correct and the force param does't take effct
> since the implementation changed at server side
> --
>
> Key: HBASE-24875
> URL: https://issues.apache.org/jira/browse/HBASE-24875
> Project: HBase
> Issue Type: Improvement
> Components: Client
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> The description of unassign in Admin.java shows below, it is not true, in
> fact, we just close the region now, also we do not need the force param any
> more.
> {code:java}
> /**
> * Unassign a region from current hosting regionserver. Region will then be
> assigned to a
> * regionserver chosen at random. Region could be reassigned back to the
> same server. Use {@link
> * #move(byte[], ServerName)} if you want to control the region movement.
> *
> * @param regionName Region to unassign. Will clear any existing RegionPlan
> if one found.
> * @param force If true, force unassign (Will remove region from
> regions-in-transition too if
> * present. If results in double assignment use hbck -fix to resolve. To be
> used by experts).
> * @throws IOException if a remote or network exception occurs
> */
> void unassign(byte[] regionName, boolean force)
> throws IOException;
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24875) The semantic of unassign not correct and the force param does't take effct since the implementation changed at server side
[
https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24875:
---
Description: The semantic of unassign not correct and the force param
does't take effct since the implementation changed at server side (was: The
description of unassign in Admin.java shows below, it is not true, in fact, we
just close the region now, also we do not need the force param any more.
{code:java}
/**
* Unassign a region from current hosting regionserver. Region will then be
assigned to a
* regionserver chosen at random. Region could be reassigned back to the same
server. Use {@link
* #move(byte[], ServerName)} if you want to control the region movement.
*
* @param regionName Region to unassign. Will clear any existing RegionPlan if
one found.
* @param force If true, force unassign (Will remove region from
regions-in-transition too if
* present. If results in double assignment use hbck -fix to resolve. To be
used by experts).
* @throws IOException if a remote or network exception occurs
*/
void unassign(byte[] regionName, boolean force)
throws IOException;
{code})
> The semantic of unassign not correct and the force param does't take effct
> since the implementation changed at server side
> --
>
> Key: HBASE-24875
> URL: https://issues.apache.org/jira/browse/HBASE-24875
> Project: HBase
> Issue Type: Improvement
> Components: Client
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> The semantic of unassign not correct and the force param does't take effct
> since the implementation changed at server side
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24854) Correct the help content of assign and unassign commands in hbase shell
[
https://issues.apache.org/jira/browse/HBASE-24854?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17176923#comment-17176923
]
Zheng Wang commented on HBASE-24854:
Thanks for all the comments.
> Correct the help content of assign and unassign commands in hbase shell
> ---
>
> Key: HBASE-24854
> URL: https://issues.apache.org/jira/browse/HBASE-24854
> Project: HBase
> Issue Type: Improvement
> Components: shell
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Minor
> Fix For: 3.0.0-alpha-1, 2.4.0, 2.2.7, 2.3.2
>
>
> Help content in assign.rb:
> Assign a region. Use with caution. *If region already assigned,*
> *this command will do a force reassign.* For experts only.
>
> Help content in unassign.rb:
> Unassign a region. *Unassign will close region in current location and then*
> *reopen it again.* Pass 'true' to force the unassignment ('force' will clear
> all in-memory state in master before the reassign. If results in
> double assignment use hbck -fix to resolve. To be used by experts).
>
> The bold part are not right now, so update them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24875) The describe and param for unassign not correct since the implementation changed at server side
[
https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24875:
---
Description:
The description of unassign in Admin.java shows below, it is not true, in fact,
we just close the region now, also we do not need the force param any more.
{code:java}
/**
* Unassign a region from current hosting regionserver. Region will then be
assigned to a
* regionserver chosen at random. Region could be reassigned back to the same
server. Use {@link
* #move(byte[], ServerName)} if you want to control the region movement.
*
* @param regionName Region to unassign. Will clear any existing RegionPlan if
one found.
* @param force If true, force unassign (Will remove region from
regions-in-transition too if
* present. If results in double assignment use hbck -fix to resolve. To be
used by experts).
* @throws IOException if a remote or network exception occurs
*/
void unassign(byte[] regionName, boolean force)
throws IOException;
{code}
was:
The describe of unassign in Admin.java shows below, it is not true, in fact, we
just close the region now, also we do not need the force param any more.
{code:java}
/**
* Unassign a region from current hosting regionserver. Region will then be
assigned to a
* regionserver chosen at random. Region could be reassigned back to the same
server. Use {@link
* #move(byte[], ServerName)} if you want to control the region movement.
*
* @param regionName Region to unassign. Will clear any existing RegionPlan if
one found.
* @param force If true, force unassign (Will remove region from
regions-in-transition too if
* present. If results in double assignment use hbck -fix to resolve. To be
used by experts).
* @throws IOException if a remote or network exception occurs
*/
void unassign(byte[] regionName, boolean force)
throws IOException;
{code}
> The describe and param for unassign not correct since the implementation
> changed at server side
> ---
>
> Key: HBASE-24875
> URL: https://issues.apache.org/jira/browse/HBASE-24875
> Project: HBase
> Issue Type: Improvement
> Components: Client
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> The description of unassign in Admin.java shows below, it is not true, in
> fact, we just close the region now, also we do not need the force param any
> more.
> {code:java}
> /**
> * Unassign a region from current hosting regionserver. Region will then be
> assigned to a
> * regionserver chosen at random. Region could be reassigned back to the
> same server. Use {@link
> * #move(byte[], ServerName)} if you want to control the region movement.
> *
> * @param regionName Region to unassign. Will clear any existing RegionPlan
> if one found.
> * @param force If true, force unassign (Will remove region from
> regions-in-transition too if
> * present. If results in double assignment use hbck -fix to resolve. To be
> used by experts).
> * @throws IOException if a remote or network exception occurs
> */
> void unassign(byte[] regionName, boolean force)
> throws IOException;
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24875) The describe and param for unassign not correct since the implementation changed at server side
[
https://issues.apache.org/jira/browse/HBASE-24875?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24875:
---
Summary: The describe and param for unassign not correct since the
implementation changed at server side (was: The describe and param for
unassign not correct since the implement changed at server side)
> The describe and param for unassign not correct since the implementation
> changed at server side
> ---
>
> Key: HBASE-24875
> URL: https://issues.apache.org/jira/browse/HBASE-24875
> Project: HBase
> Issue Type: Improvement
> Components: Client
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> The describe of unassign in Admin.java shows below, it is not true, in fact,
> we just close the region now, also we do not need the force param any more.
> {code:java}
> /**
> * Unassign a region from current hosting regionserver. Region will then be
> assigned to a
> * regionserver chosen at random. Region could be reassigned back to the
> same server. Use {@link
> * #move(byte[], ServerName)} if you want to control the region movement.
> *
> * @param regionName Region to unassign. Will clear any existing RegionPlan
> if one found.
> * @param force If true, force unassign (Will remove region from
> regions-in-transition too if
> * present. If results in double assignment use hbck -fix to resolve. To be
> used by experts).
> * @throws IOException if a remote or network exception occurs
> */
> void unassign(byte[] regionName, boolean force)
> throws IOException;
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (HBASE-24875) The describe and param for unassign not correct since the implement changed at server side
Zheng Wang created HBASE-24875:
--
Summary: The describe and param for unassign not correct since the
implement changed at server side
Key: HBASE-24875
URL: https://issues.apache.org/jira/browse/HBASE-24875
Project: HBase
Issue Type: Improvement
Components: Client
Reporter: Zheng Wang
Assignee: Zheng Wang
The describe of unassign in Admin.java shows below, it is not true, in fact, we
just close the region now, also we do not need the force param any more.
{code:java}
/**
* Unassign a region from current hosting regionserver. Region will then be
assigned to a
* regionserver chosen at random. Region could be reassigned back to the same
server. Use {@link
* #move(byte[], ServerName)} if you want to control the region movement.
*
* @param regionName Region to unassign. Will clear any existing RegionPlan if
one found.
* @param force If true, force unassign (Will remove region from
regions-in-transition too if
* present. If results in double assignment use hbck -fix to resolve. To be
used by experts).
* @throws IOException if a remote or network exception occurs
*/
void unassign(byte[] regionName, boolean force)
throws IOException;
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24854) Correct the help content of assign and unassign commands in hbase shell
[
https://issues.apache.org/jira/browse/HBASE-24854?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24854:
---
Description:
Help content in assign.rb:
Assign a region. Use with caution. *If region already assigned,*
*this command will do a force reassign.* For experts only.
Help content in unassign.rb:
Unassign a region. *Unassign will close region in current location and then*
*reopen it again.* Pass 'true' to force the unassignment ('force' will clear
all in-memory state in master before the reassign. If results in
double assignment use hbck -fix to resolve. To be used by experts).
The bold part are not right now, so update them.
was:
Help content in assign.rb:
Assign a region. Use with caution. *If region already assigned,*
*this command will do a force reassign.* For experts only.
Help content in unassign.rb:
Unassign a region. *Unassign will close region in current location and then*
*reopen it again.* Pass 'true' to force the unassignment ('force' will clear
all in-memory state in master before the reassign. If results in
double assignment use hbck -fix to resolve. To be used by experts).
The bold part are not real now, so update them.
> Correct the help content of assign and unassign commands in hbase shell
> ---
>
> Key: HBASE-24854
> URL: https://issues.apache.org/jira/browse/HBASE-24854
> Project: HBase
> Issue Type: Improvement
> Components: shell
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Minor
>
> Help content in assign.rb:
> Assign a region. Use with caution. *If region already assigned,*
> *this command will do a force reassign.* For experts only.
>
> Help content in unassign.rb:
> Unassign a region. *Unassign will close region in current location and then*
> *reopen it again.* Pass 'true' to force the unassignment ('force' will clear
> all in-memory state in master before the reassign. If results in
> double assignment use hbck -fix to resolve. To be used by experts).
>
> The bold part are not right now, so update them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24854) Correct the help content of assign and unassign commands in hbase shell
[
https://issues.apache.org/jira/browse/HBASE-24854?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24854:
---
Description:
Help content in assign.rb:
Assign a region. Use with caution. *If region already assigned,*
*this command will do a force reassign.* For experts only.
Help content in unassign.rb:
Unassign a region. *Unassign will close region in current location and then*
*reopen it again.* Pass 'true' to force the unassignment ('force' will clear
all in-memory state in master before the reassign. If results in
double assignment use hbck -fix to resolve. To be used by experts).
The bold part are not real now, so update them.
was:
In assign.rb:
Assign a region. Use with caution. *If region already assigned,*
*this command will do a force reassign.* For experts only.
In unassign.rb:
Unassign a region. *Unassign will close region in current location and then*
*reopen it again.* Pass 'true' to force the unassignment ('force' will clear
all in-memory state in master before the reassign. If results in
double assignment use hbck -fix to resolve. To be used by experts).
The bold part are not real now, so update them.
> Correct the help content of assign and unassign commands in hbase shell
> ---
>
> Key: HBASE-24854
> URL: https://issues.apache.org/jira/browse/HBASE-24854
> Project: HBase
> Issue Type: Improvement
> Components: shell
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Minor
>
> Help content in assign.rb:
> Assign a region. Use with caution. *If region already assigned,*
> *this command will do a force reassign.* For experts only.
>
> Help content in unassign.rb:
> Unassign a region. *Unassign will close region in current location and then*
> *reopen it again.* Pass 'true' to force the unassignment ('force' will clear
> all in-memory state in master before the reassign. If results in
> double assignment use hbck -fix to resolve. To be used by experts).
>
> The bold part are not real now, so update them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24854) Correct the help content of assign and unassign commands in hbase shell
[
https://issues.apache.org/jira/browse/HBASE-24854?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24854:
---
Description:
In assign.rb:
Assign a region. Use with caution. *If region already assigned,*
*this command will do a force reassign.* For experts only.
In unassign.rb:
Unassign a region. *Unassign will close region in current location and then*
*reopen it again.* Pass 'true' to force the unassignment ('force' will clear
all in-memory state in master before the reassign. If results in
double assignment use hbck -fix to resolve. To be used by experts).
The bold part are not real now, so update them.
> Correct the help content of assign and unassign commands in hbase shell
> ---
>
> Key: HBASE-24854
> URL: https://issues.apache.org/jira/browse/HBASE-24854
> Project: HBase
> Issue Type: Improvement
> Components: shell
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Minor
>
> In assign.rb:
> Assign a region. Use with caution. *If region already assigned,*
> *this command will do a force reassign.* For experts only.
> In unassign.rb:
> Unassign a region. *Unassign will close region in current location and then*
> *reopen it again.* Pass 'true' to force the unassignment ('force' will clear
> all in-memory state in master before the reassign. If results in
> double assignment use hbck -fix to resolve. To be used by experts).
> The bold part are not real now, so update them.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (HBASE-24854) Correct the help content of assign and unassign commands in hbase shell
Zheng Wang created HBASE-24854: -- Summary: Correct the help content of assign and unassign commands in hbase shell Key: HBASE-24854 URL: https://issues.apache.org/jira/browse/HBASE-24854 Project: HBase Issue Type: Improvement Components: shell Reporter: Zheng Wang Assignee: Zheng Wang -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24853) The archiveLogFile should be called before decreasing totalLogSize in AbstractFSWAL.cleanOldLogs
Zheng Wang created HBASE-24853: -- Summary: The archiveLogFile should be called before decreasing totalLogSize in AbstractFSWAL.cleanOldLogs Key: HBASE-24853 URL: https://issues.apache.org/jira/browse/HBASE-24853 Project: HBase Issue Type: Improvement Components: wal Reporter: Zheng Wang Assignee: Zheng Wang As discussed in HBASE-24822, if archiveLogFile failed to rename, then the totalLogSize will not correct. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24709) Support MoveCostFunction use a lower multiplier in offpeak hours
[ https://issues.apache.org/jira/browse/HBASE-24709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17175170#comment-17175170 ] Zheng Wang commented on HBASE-24709: Filed a new Jira HBASE-24848. [~ndimiduk] [~bharathv] > Support MoveCostFunction use a lower multiplier in offpeak hours > > > Key: HBASE-24709 > URL: https://issues.apache.org/jira/browse/HBASE-24709 > Project: HBase > Issue Type: Improvement > Components: Balancer >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > Attachments: HBASE-24709-jenkins-log > > > Currently the default multiplier of MoveCostFunction is 7, it should be lower > in offpeak hours, thus balancer could do more things. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24848) Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier for branch-1
[ https://issues.apache.org/jira/browse/HBASE-24848?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24848: --- Attachment: HBASE-24709-jenkins-log > Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier for branch-1 > -- > > Key: HBASE-24848 > URL: https://issues.apache.org/jira/browse/HBASE-24848 > Project: HBase > Issue Type: Improvement > Components: test >Affects Versions: 1.7.0 >Reporter: Zheng Wang >Priority: Major > Attachments: HBASE-24709-jenkins-log > > > This test added by HBASE-24709, and it failed on jdk7 reported by Hudson for > branch-1, but cant reproduce in my local env, so create this jira to keep > tracing. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24848) Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier for branch-1
Zheng Wang created HBASE-24848: -- Summary: Flakey test TestStochasticLoadBalancer.testMoveCostMultiplier for branch-1 Key: HBASE-24848 URL: https://issues.apache.org/jira/browse/HBASE-24848 Project: HBase Issue Type: Improvement Components: test Affects Versions: 1.7.0 Reporter: Zheng Wang Attachments: HBASE-24709-jenkins-log This test added by HBASE-24709, and it failed on jdk7 reported by Hudson for branch-1, but cant reproduce in my local env, so create this jira to keep tracing. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24694) Support flush a single column family of table
[ https://issues.apache.org/jira/browse/HBASE-24694?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17174230#comment-17174230 ] Zheng Wang commented on HBASE-24694: Thanks for your time. [~wchevreuil] > Support flush a single column family of table > - > > Key: HBASE-24694 > URL: https://issues.apache.org/jira/browse/HBASE-24694 > Project: HBase > Issue Type: New Feature >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > This is follow-on work of HBASE-24404, do it as a seprate issue could make it > easier to reveiw. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172789#comment-17172789
]
Zheng Wang edited comment on HBASE-24822 at 8/10/20, 9:07 AM:
--
It could be improved in future, such as archive a specific log file, archive
all log files, etc.
was (Author: filtertip):
It could be improved in future, such as archive a specify log file, archive all
log files, etc.
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24694) Support flush a single column family of table
[ https://issues.apache.org/jira/browse/HBASE-24694?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17174190#comment-17174190 ] Zheng Wang commented on HBASE-24694: Pushed a new PR to branch-2. [~wchevreuil] > Support flush a single column family of table > - > > Key: HBASE-24694 > URL: https://issues.apache.org/jira/browse/HBASE-24694 > Project: HBase > Issue Type: New Feature >Affects Versions: 3.0.0-alpha-1 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > This is follow-on work of HBASE-24404, do it as a seprate issue could make it > easier to reveiw. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24840) Avoid shows closing region task when create table
[ https://issues.apache.org/jira/browse/HBASE-24840?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24840: --- Attachment: HBASE-24840-beforepatch.png HBASE-24840-afterpatch.png > Avoid shows closing region task when create table > - > > Key: HBASE-24840 > URL: https://issues.apache.org/jira/browse/HBASE-24840 > Project: HBase > Issue Type: Improvement > Components: monitoring >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24840-afterpatch.png, HBASE-24840-beforepatch.png > > > Currently when we create region dir, we new a HRegion and then close it, this > step is before the real initialize, so there will exist a closing region task > at first, it is strange, and will confuse users. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24840) Avoid shows closing region task when create table
Zheng Wang created HBASE-24840: -- Summary: Avoid shows closing region task when create table Key: HBASE-24840 URL: https://issues.apache.org/jira/browse/HBASE-24840 Project: HBase Issue Type: Improvement Components: monitoring Reporter: Zheng Wang Assignee: Zheng Wang Attachments: HBASE-24840-afterpatch.png, HBASE-24840-beforepatch.png Currently when we create region dir, we new a HRegion and then close it, this step is before the real initialize, so there will exist a closing region task at first, it is strange, and will confuse users. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24694) Support flush a single column family of table
[ https://issues.apache.org/jira/browse/HBASE-24694?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17173060#comment-17173060 ] Zheng Wang commented on HBASE-24694: No prob, will do it tomorrow. Thanks. [~wchevreuil] > Support flush a single column family of table > - > > Key: HBASE-24694 > URL: https://issues.apache.org/jira/browse/HBASE-24694 > Project: HBase > Issue Type: New Feature >Affects Versions: 3.0.0-alpha-1 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > This is follow-on work of HBASE-24404, do it as a seprate issue could make it > easier to reveiw. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17173059#comment-17173059 ] Zheng Wang commented on HBASE-24404: Thanks for improving the release note. [~wchevreuil] > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24826) Add some comments for processlist in hbase shell
[ https://issues.apache.org/jira/browse/HBASE-24826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172792#comment-17172792 ] Zheng Wang edited comment on HBASE-24826 at 8/7/20, 3:15 AM: - Thanks. [~stack] [~bitoffdev] was (Author: filtertip): Thanks. [~stack] > Add some comments for processlist in hbase shell > > > Key: HBASE-24826 > URL: https://issues.apache.org/jira/browse/HBASE-24826 > Project: HBase > Issue Type: Improvement > Components: shell >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is not easy to understand the difference of the options for processlist, > so maybe some comments is necessary. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24826) Add some comments for processlist in hbase shell
[ https://issues.apache.org/jira/browse/HBASE-24826?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172792#comment-17172792 ] Zheng Wang commented on HBASE-24826: Thanks. [~stack] > Add some comments for processlist in hbase shell > > > Key: HBASE-24826 > URL: https://issues.apache.org/jira/browse/HBASE-24826 > Project: HBase > Issue Type: Improvement > Components: shell >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is not easy to understand the difference of the options for processlist, > so maybe some comments is necessary. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172789#comment-17172789
]
Zheng Wang edited comment on HBASE-24822 at 8/7/20, 2:59 AM:
-
It could be improved in future, such as archive a specify log file, archive all
log files, etc.
was (Author: filtertip):
It could be improve in future, such as archive a specify log file, archive all
log files, etc.
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172789#comment-17172789
]
Zheng Wang commented on HBASE-24822:
It could be improve in future, such as archive a specify log file, archive all
log files, etc.
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172781#comment-17172781
]
Zheng Wang commented on HBASE-24822:
At that time, I wanted to verify on running mode that if the correct stores of
regions are flushed when we force archive earliest wal, but currently it only
could be triggered by logRoller(if reach maxLogs), so it's a little
inconvenient for me.
Thanks.[~stack]
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172518#comment-17172518
]
Zheng Wang edited comment on HBASE-24822 at 8/7/20, 1:48 AM:
-
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it would be useful for folks as a common tool.
[~stack]
was (Author: filtertip):
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it would be useful for folks. [~stack]
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172518#comment-17172518
]
Zheng Wang edited comment on HBASE-24822 at 8/6/20, 4:30 PM:
-
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it would be useful for folks. [~stack]
was (Author: filtertip):
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it will be
useful for folks. [~stack]
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172518#comment-17172518
]
Zheng Wang edited comment on HBASE-24822 at 8/6/20, 4:29 PM:
-
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it will be
useful for folks. [~stack]
was (Author: filtertip):
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it will be
useful for folks.
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Commented] (HBASE-24822) Add a command to support archive the earliest log file manually
[
https://issues.apache.org/jira/browse/HBASE-24822?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17172518#comment-17172518
]
Zheng Wang commented on HBASE-24822:
I tried to do this when was implementing HBASE-24382, but could not find a
suitable command, so i think it will be
useful for folks.
> Add a command to support archive the earliest log file manually
> ---
>
> Key: HBASE-24822
> URL: https://issues.apache.org/jira/browse/HBASE-24822
> Project: HBase
> Issue Type: New Feature
> Components: wal
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Major
>
> This new command named wal_archive, belong to the group of "tools".
> It is a sync call, if the data related to earliest log file not all
> persisted, they will be flushed immediately.
> Usage:
> {code:java}
> wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24826) Add some comments for processlist in hbase shell
[ https://issues.apache.org/jira/browse/HBASE-24826?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24826: --- Description: It is not easy to understand the difference of the options for processlist, so maybe some comments is necessary. was: It is not easy to understand the different of the options for processlist, so maybe some comments is necessary. > Add some comments for processlist in hbase shell > > > Key: HBASE-24826 > URL: https://issues.apache.org/jira/browse/HBASE-24826 > Project: HBase > Issue Type: Improvement > Components: shell >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > > It is not easy to understand the difference of the options for processlist, > so maybe some comments is necessary. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24826) Add some comments for processlist in hbase shell
[ https://issues.apache.org/jira/browse/HBASE-24826?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24826: --- Summary: Add some comments for processlist in hbase shell (was: Add some comments for command of processlist) > Add some comments for processlist in hbase shell > > > Key: HBASE-24826 > URL: https://issues.apache.org/jira/browse/HBASE-24826 > Project: HBase > Issue Type: Improvement > Components: shell >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > > It is not easy to understand the different of the options for processlist, so > maybe some comments is necessary. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24826) Add some comments for command of processlist
Zheng Wang created HBASE-24826: -- Summary: Add some comments for command of processlist Key: HBASE-24826 URL: https://issues.apache.org/jira/browse/HBASE-24826 Project: HBase Issue Type: Improvement Components: shell Reporter: Zheng Wang Assignee: Zheng Wang It is not easy to understand the different of the options for processlist, so maybe some comments is necessary. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24821) Simplify the logic of getRegionInfo in TestFlushFromClient to reduce redundancy code
[
https://issues.apache.org/jira/browse/HBASE-24821?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17171555#comment-17171555
]
Zheng Wang commented on HBASE-24821:
Thanks. [~vjasani]
> Simplify the logic of getRegionInfo in TestFlushFromClient to reduce
> redundancy code
>
>
> Key: HBASE-24821
> URL: https://issues.apache.org/jira/browse/HBASE-24821
> Project: HBase
> Issue Type: Improvement
> Components: test
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Minor
> Fix For: 3.0.0-alpha-1, 2.3.1, 2.4.0, 2.2.7
>
>
> Current logic:
> {code:java}
> private List getRegionInfo() {
> return TEST_UTIL.getHBaseCluster().getLiveRegionServerThreads().stream()
> .map(JVMClusterUtil.RegionServerThread::getRegionServer)
> .flatMap(r -> r.getRegions().stream())
> .filter(r -> r.getTableDescriptor().getTableName().equals(tableName))
> .collect(Collectors.toList());
> }
> {code}
> The MiniHBaseCluster has similar method to do same thing.
> So it could just directly call:
> {code:java}
> private List getRegionInfo() {
> return TEST_UTIL.getHBaseCluster().getRegions(tableName);
> }
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (HBASE-24822) Add a command to support archive the earliest log file manually
Zheng Wang created HBASE-24822:
--
Summary: Add a command to support archive the earliest log file
manually
Key: HBASE-24822
URL: https://issues.apache.org/jira/browse/HBASE-24822
Project: HBase
Issue Type: New Feature
Components: wal
Reporter: Zheng Wang
Assignee: Zheng Wang
This new command named wal_archive, belong to the group of "tools".
It is a sync call, if the data related to earliest log file not all persisted,
they will be flushed immediately.
Usage:
{code:java}
wal_archive 'desktop-ud83s84,16020,1596595277350'{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24821) Simplify the logic of getRegionInfo in TestFlushFromClient to reduce redundancy code
[
https://issues.apache.org/jira/browse/HBASE-24821?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Zheng Wang updated HBASE-24821:
---
Summary: Simplify the logic of getRegionInfo in TestFlushFromClient to
reduce redundancy code (was: simplify the logic of getRegionInfo in
TestFlushFromClient to reduce redundancy code)
> Simplify the logic of getRegionInfo in TestFlushFromClient to reduce
> redundancy code
>
>
> Key: HBASE-24821
> URL: https://issues.apache.org/jira/browse/HBASE-24821
> Project: HBase
> Issue Type: Improvement
> Components: test
>Reporter: Zheng Wang
>Assignee: Zheng Wang
>Priority: Minor
>
> Current logic:
> {code:java}
> private List getRegionInfo() {
> return TEST_UTIL.getHBaseCluster().getLiveRegionServerThreads().stream()
> .map(JVMClusterUtil.RegionServerThread::getRegionServer)
> .flatMap(r -> r.getRegions().stream())
> .filter(r -> r.getTableDescriptor().getTableName().equals(tableName))
> .collect(Collectors.toList());
> }
> {code}
> The MiniHBaseCluster has similar method to do same thing.
> So it could just directly call:
> {code:java}
> private List getRegionInfo() {
> return TEST_UTIL.getHBaseCluster().getRegions(tableName);
> }
> {code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Created] (HBASE-24821) simplify the logic of getRegionInfo in TestFlushFromClient to reduce redundancy code
Zheng Wang created HBASE-24821:
--
Summary: simplify the logic of getRegionInfo in
TestFlushFromClient to reduce redundancy code
Key: HBASE-24821
URL: https://issues.apache.org/jira/browse/HBASE-24821
Project: HBase
Issue Type: Improvement
Components: test
Reporter: Zheng Wang
Assignee: Zheng Wang
Current logic:
{code:java}
private List getRegionInfo() {
return TEST_UTIL.getHBaseCluster().getLiveRegionServerThreads().stream()
.map(JVMClusterUtil.RegionServerThread::getRegionServer)
.flatMap(r -> r.getRegions().stream())
.filter(r -> r.getTableDescriptor().getTableName().equals(tableName))
.collect(Collectors.toList());
}
{code}
The MiniHBaseCluster has similar method to do same thing.
So it could just directly call:
{code:java}
private List getRegionInfo() {
return TEST_UTIL.getHBaseCluster().getRegions(tableName);
}
{code}
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Updated] (HBASE-24704) Make the Table Schema easier to view even there are multiple families
[ https://issues.apache.org/jira/browse/HBASE-24704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24704: --- Release Note: Improve the layout of column family from vertical to horizontal in table UI. > Make the Table Schema easier to view even there are multiple families > - > > Key: HBASE-24704 > URL: https://issues.apache.org/jira/browse/HBASE-24704 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > Attachments: HBASE-24704-beforepatch.png, > HBASE-24704-branch1-afterpatch.png, HBASE-24704-branch2-afterpatch.png, > HBASE-24704-master-afterpatch.png > > > Currently we display the column family as vertical, it is not easy to view, > my proposal is to display them as horizontal. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24704) Make the Table Schema easier to view even there are multiple families
[ https://issues.apache.org/jira/browse/HBASE-24704?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17169669#comment-17169669 ] Zheng Wang commented on HBASE-24704: Thanks a lot. [~vjasani] > Make the Table Schema easier to view even there are multiple families > - > > Key: HBASE-24704 > URL: https://issues.apache.org/jira/browse/HBASE-24704 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > Attachments: HBASE-24704-beforepatch.png, > HBASE-24704-branch1-afterpatch.png, HBASE-24704-branch2-afterpatch.png, > HBASE-24704-master-afterpatch.png > > > Currently we display the column family as vertical, it is not easy to view, > my proposal is to display them as horizontal. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24704) Make the Table Schema easier to view even there are multiple families
[ https://issues.apache.org/jira/browse/HBASE-24704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24704: --- Attachment: HBASE-24704-branch1-afterpatch.png > Make the Table Schema easier to view even there are multiple families > - > > Key: HBASE-24704 > URL: https://issues.apache.org/jira/browse/HBASE-24704 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24704-beforepatch.png, > HBASE-24704-branch1-afterpatch.png, HBASE-24704-branch2-afterpatch.png, > HBASE-24704-master-afterpatch.png > > > Currently we display the column family as vertical, it is not easy to view, > my proposal is to display them as horizontal. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24704) Make the Table Schema easier to view even there are multiple families
[ https://issues.apache.org/jira/browse/HBASE-24704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24704: --- Attachment: (was: HBASE-24704-afterpatch.png) > Make the Table Schema easier to view even there are multiple families > - > > Key: HBASE-24704 > URL: https://issues.apache.org/jira/browse/HBASE-24704 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24704-beforepatch.png, > HBASE-24704-branch2-afterpatch.png, HBASE-24704-master-afterpatch.png > > > Currently we display the column family as vertical, it is not easy to view, > my proposal is to display them as horizontal. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24704) Make the Table Schema easier to view even there are multiple families
[ https://issues.apache.org/jira/browse/HBASE-24704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24704: --- Attachment: HBASE-24704-branch2-afterpatch.png HBASE-24704-master-afterpatch.png > Make the Table Schema easier to view even there are multiple families > - > > Key: HBASE-24704 > URL: https://issues.apache.org/jira/browse/HBASE-24704 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24704-beforepatch.png, > HBASE-24704-branch2-afterpatch.png, HBASE-24704-master-afterpatch.png > > > Currently we display the column family as vertical, it is not easy to view, > my proposal is to display them as horizontal. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24404: --- Release Note: Enhance the flush region command that support to specify a column family. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' was: Enhance the flush region command that support specify a column family. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24404: --- Release Note: Enhance the flush region command that support specify a column family. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' was: Enhance the flush region command that support to specify a column family. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24404: --- Release Note: Enhance the flush region command that support specify a column family. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' was: Enhance the flush region command that support specify a column. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24404: --- Release Note: Enhance the flush region command that support specify a column. Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' was: Enhance the flush region command that support specify a column Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24404: --- Release Note: Enhance the flush region command that support specify a column Usage: flush 'REGIONNAME','FAMILYNAME' flush 'ENCODED_REGIONNAME','FAMILYNAME' > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17168624#comment-17168624 ] Zheng Wang commented on HBASE-24404: bq. Mind put a release note text on the jira, Zheng Wang? No prob. bq. Since it's a new API, I guess it's better backport to branch-2 only, as 2.3.0 was already released. Makes sense. Thanks. > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17168545#comment-17168545 ] Zheng Wang commented on HBASE-24404: [~wchevreuil]Thanks. > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17168545#comment-17168545 ] Zheng Wang edited comment on HBASE-24404 at 7/31/20, 9:09 AM: -- [~wchevreuil] Thanks. was (Author: filtertip): [~wchevreuil]Thanks. > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17164311#comment-17164311 ] Zheng Wang commented on HBASE-24404: [~wchevreuil] I had pushed a PR for branch-2, could you help to merge it if you have time? Thanks. > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24664) Some changing of split region by overall region size rather than only one store size
[ https://issues.apache.org/jira/browse/HBASE-24664?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17164302#comment-17164302 ] Zheng Wang edited comment on HBASE-24664 at 7/24/20, 9:22 AM: -- [~wchevreuil] I had pushed PR for branch-2.2 and branch-2.3, could you help to merge them if you have time? Thanks. was (Author: filtertip): [~wchevreuil] I had pushed PR for branch-2 and branch-3, could you help to merge them if you have time? Thanks. > Some changing of split region by overall region size rather than only one > store size > > > Key: HBASE-24664 > URL: https://issues.apache.org/jira/browse/HBASE-24664 > Project: HBase > Issue Type: Improvement > Components: regionserver >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > As a distributed cluster, HBase distribute loads in unit of region, so if > region grows too big, > it will bring some negative effects, such as: > 1. Harder to homogenize disk usage(consider locality) > 2. Might cost more time on region opening > 3. After split, the daughter region might lead to more io cost on compaction > in a short time(if write evenly) > I tried to introduce a new SteppingAllStoresSizeSplitPolicy in HBASE-24530, > but after discussed in comments and related > [thread|https://lists.apache.org/thread.html/r08a8103e2532eb667a0fcb4efa8a4117b3f82e6251bc4bd0bc157c26%40%3Cdev.hbase.apache.org%3E], > finally we decide to change the existing split policy with a new option that > if it should count all store files, and for master it would be true, else > false. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24664) Some changing of split region by overall region size rather than only one store size
[ https://issues.apache.org/jira/browse/HBASE-24664?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17164302#comment-17164302 ] Zheng Wang commented on HBASE-24664: [~wchevreuil] I had pushed PR for branch-2 and branch-3, could you help to merge them if you have time? Thanks. > Some changing of split region by overall region size rather than only one > store size > > > Key: HBASE-24664 > URL: https://issues.apache.org/jira/browse/HBASE-24664 > Project: HBase > Issue Type: Improvement > Components: regionserver >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > As a distributed cluster, HBase distribute loads in unit of region, so if > region grows too big, > it will bring some negative effects, such as: > 1. Harder to homogenize disk usage(consider locality) > 2. Might cost more time on region opening > 3. After split, the daughter region might lead to more io cost on compaction > in a short time(if write evenly) > I tried to introduce a new SteppingAllStoresSizeSplitPolicy in HBASE-24530, > but after discussed in comments and related > [thread|https://lists.apache.org/thread.html/r08a8103e2532eb667a0fcb4efa8a4117b3f82e6251bc4bd0bc157c26%40%3Cdev.hbase.apache.org%3E], > finally we decide to change the existing split policy with a new option that > if it should count all store files, and for master it would be true, else > false. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24709) Support MoveCostFunction use a lower multiplier in offpeak hours
[ https://issues.apache.org/jira/browse/HBASE-24709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17162517#comment-17162517 ] Zheng Wang commented on HBASE-24709: I have add a loop for the internal code of testMoveCostMultiplier() to run 100 times, no fail occurs, and also review the related code carefully, seems no reason to fail since the code changing is so small and clear. [~bharathv] > Support MoveCostFunction use a lower multiplier in offpeak hours > > > Key: HBASE-24709 > URL: https://issues.apache.org/jira/browse/HBASE-24709 > Project: HBase > Issue Type: Improvement > Components: Balancer >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > Attachments: HBASE-24709-jenkins-log > > > Currently the default multiplier of MoveCostFunction is 7, it should be lower > in offpeak hours, thus balancer could do more things. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24709) Support MoveCostFunction use a lower multiplier in offpeak hours
[ https://issues.apache.org/jira/browse/HBASE-24709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17162421#comment-17162421 ] Zheng Wang commented on HBASE-24709: [~bharathv] Thanks for the log, will dig more. > Support MoveCostFunction use a lower multiplier in offpeak hours > > > Key: HBASE-24709 > URL: https://issues.apache.org/jira/browse/HBASE-24709 > Project: HBase > Issue Type: Improvement > Components: Balancer >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > Attachments: HBASE-24709-jenkins-log > > > Currently the default multiplier of MoveCostFunction is 7, it should be lower > in offpeak hours, thus balancer could do more things. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24709) Support MoveCostFunction use a lower multiplier in offpeak hours
[ https://issues.apache.org/jira/browse/HBASE-24709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17161679#comment-17161679 ] Zheng Wang commented on HBASE-24709: Sorry, but I can not reproduce the problem in my local env, tested with jdk1.7+hadoop2. Did I miss something? [~bharathv] > Support MoveCostFunction use a lower multiplier in offpeak hours > > > Key: HBASE-24709 > URL: https://issues.apache.org/jira/browse/HBASE-24709 > Project: HBase > Issue Type: Improvement > Components: Balancer >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > > Currently the default multiplier of MoveCostFunction is 7, it should be lower > in offpeak hours, thus balancer could do more things. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24709) Support MoveCostFunction use a lower multiplier in offpeak hours
[ https://issues.apache.org/jira/browse/HBASE-24709?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17161652#comment-17161652 ] Zheng Wang commented on HBASE-24709: No prob, will do it later. [~bharathv] > Support MoveCostFunction use a lower multiplier in offpeak hours > > > Key: HBASE-24709 > URL: https://issues.apache.org/jira/browse/HBASE-24709 > Project: HBase > Issue Type: Improvement > Components: Balancer >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 1.7.0, 2.4.0 > > > Currently the default multiplier of MoveCostFunction is 7, it should be lower > in offpeak hours, thus balancer could do more things. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24748) Add hbase.master.balancer.stochastic.moveCost.offpeak to doc as support dynamically change
Zheng Wang created HBASE-24748: -- Summary: Add hbase.master.balancer.stochastic.moveCost.offpeak to doc as support dynamically change Key: HBASE-24748 URL: https://issues.apache.org/jira/browse/HBASE-24748 Project: HBase Issue Type: Bug Components: documentation Reporter: Zheng Wang Assignee: Zheng Wang A follow-on task of HBASE-24709, since it is always fail on jdk11 with this doc change. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24746) The sort icons overlap the col name in master UI
[ https://issues.apache.org/jira/browse/HBASE-24746?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17160313#comment-17160313 ] Zheng Wang commented on HBASE-24746: Thanks. [~stack] > The sort icons overlap the col name in master UI > - > > Key: HBASE-24746 > URL: https://issues.apache.org/jira/browse/HBASE-24746 > Project: HBase > Issue Type: Bug > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.3.1, 2.4.0 > > Attachments: HBASE-24746-afterpatch.png, HBASE-24746-beforepatch.png > > > In tables section, when description not few, the sort icons and col name are > overlap. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24664) Some changing of split region by overall region size rather than only one store size
[ https://issues.apache.org/jira/browse/HBASE-24664?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17159847#comment-17159847 ] Zheng Wang commented on HBASE-24664: No porb, will do it later. Thanks.[~wchevreuil] > Some changing of split region by overall region size rather than only one > store size > > > Key: HBASE-24664 > URL: https://issues.apache.org/jira/browse/HBASE-24664 > Project: HBase > Issue Type: Improvement > Components: regionserver >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > As a distributed cluster, HBase distribute loads in unit of region, so if > region grows too big, > it will bring some negative effects, such as: > 1. Harder to homogenize disk usage(consider locality) > 2. Might cost more time on region opening > 3. After split, the daughter region might lead to more io cost on compaction > in a short time(if write evenly) > I tried to introduce a new SteppingAllStoresSizeSplitPolicy in HBASE-24530, > but after discussed in comments and related > [thread|https://lists.apache.org/thread.html/r08a8103e2532eb667a0fcb4efa8a4117b3f82e6251bc4bd0bc157c26%40%3Cdev.hbase.apache.org%3E], > finally we decide to change the existing split policy with a new option that > if it should count all store files, and for master it would be true, else > false. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24664) Some changing of split region by overall region size rather than only one store size
[ https://issues.apache.org/jira/browse/HBASE-24664?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17159847#comment-17159847 ] Zheng Wang edited comment on HBASE-24664 at 7/17/20, 10:16 AM: --- No prob, will do it later. Thanks.[~wchevreuil] was (Author: filtertip): No porb, will do it later. Thanks.[~wchevreuil] > Some changing of split region by overall region size rather than only one > store size > > > Key: HBASE-24664 > URL: https://issues.apache.org/jira/browse/HBASE-24664 > Project: HBase > Issue Type: Improvement > Components: regionserver >Affects Versions: 3.0.0-alpha-1, 2.4.0 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1, 2.4.0 > > > As a distributed cluster, HBase distribute loads in unit of region, so if > region grows too big, > it will bring some negative effects, such as: > 1. Harder to homogenize disk usage(consider locality) > 2. Might cost more time on region opening > 3. After split, the daughter region might lead to more io cost on compaction > in a short time(if write evenly) > I tried to introduce a new SteppingAllStoresSizeSplitPolicy in HBASE-24530, > but after discussed in comments and related > [thread|https://lists.apache.org/thread.html/r08a8103e2532eb667a0fcb4efa8a4117b3f82e6251bc4bd0bc157c26%40%3Cdev.hbase.apache.org%3E], > finally we decide to change the existing split policy with a new option that > if it should count all store files, and for master it would be true, else > false. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24555) Clear the description of hbase.hregion.max.filesize
[ https://issues.apache.org/jira/browse/HBASE-24555?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24555: --- Description: After we improve the splitpolicy in HBASE-24664, seems it is better to clear this option's meaning. (was: After we improve the splitpolicy ) > Clear the description of hbase.hregion.max.filesize > --- > > Key: HBASE-24555 > URL: https://issues.apache.org/jira/browse/HBASE-24555 > Project: HBase > Issue Type: Improvement > Components: documentation >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > > After we improve the splitpolicy in HBASE-24664, seems it is better to clear > this option's meaning. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24555) Clear the description of hbase.hregion.max.filesize
[ https://issues.apache.org/jira/browse/HBASE-24555?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24555: --- Description: After we improve the splitpolicy (was: For now, we actually split region by store size.) > Clear the description of hbase.hregion.max.filesize > --- > > Key: HBASE-24555 > URL: https://issues.apache.org/jira/browse/HBASE-24555 > Project: HBase > Issue Type: Improvement > Components: documentation >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > > After we improve the splitpolicy -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24555) Clear the description of hbase.hregion.max.filesize
[ https://issues.apache.org/jira/browse/HBASE-24555?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24555: --- Summary: Clear the description of hbase.hregion.max.filesize (was: Correct the description of hbase.hregion.max.filesize) > Clear the description of hbase.hregion.max.filesize > --- > > Key: HBASE-24555 > URL: https://issues.apache.org/jira/browse/HBASE-24555 > Project: HBase > Issue Type: Improvement > Components: documentation >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > > For now, we actually split region by store size. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24455) Correct the doc of "On the number of column families"
[ https://issues.apache.org/jira/browse/HBASE-24455?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24455: --- Summary: Correct the doc of "On the number of column families" (was: clear the doc of "On the number of column families") > Correct the doc of "On the number of column families" > - > > Key: HBASE-24455 > URL: https://issues.apache.org/jira/browse/HBASE-24455 > Project: HBase > Issue Type: Improvement > Components: documentation >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > Fix For: 3.0.0-alpha-1, 2.3.0, 1.7.0, 2.2.6 > > > Currently all the compaction is store basis yet, so correct the content. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24455) clear the doc of "On the number of column families"
[ https://issues.apache.org/jira/browse/HBASE-24455?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24455: --- Summary: clear the doc of "On the number of column families" (was: Correct the doc of "On the number of column families") > clear the doc of "On the number of column families" > --- > > Key: HBASE-24455 > URL: https://issues.apache.org/jira/browse/HBASE-24455 > Project: HBase > Issue Type: Improvement > Components: documentation >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Minor > Fix For: 3.0.0-alpha-1, 2.3.0, 1.7.0, 2.2.6 > > > Currently all the compaction is store basis yet, so correct the content. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24746) The sort icons overlap the col name in master UI
[ https://issues.apache.org/jira/browse/HBASE-24746?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24746: --- Attachment: HBASE-24746-afterpatch.png HBASE-24746-beforepatch.png > The sort icons overlap the col name in master UI > - > > Key: HBASE-24746 > URL: https://issues.apache.org/jira/browse/HBASE-24746 > Project: HBase > Issue Type: Bug > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24746-afterpatch.png, HBASE-24746-beforepatch.png > > > In tables section, when description not few, the sort icons and col name are > overlap. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24746) The sort icons overlap the col name in master UI
[ https://issues.apache.org/jira/browse/HBASE-24746?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24746: --- Attachment: (was: HBASE-24746.png) > The sort icons overlap the col name in master UI > - > > Key: HBASE-24746 > URL: https://issues.apache.org/jira/browse/HBASE-24746 > Project: HBase > Issue Type: Bug > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > In tables section, when description not few, the sort icons and col name are > overlap. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Assigned] (HBASE-24746) The sort icons overlap the col name in master UI
[ https://issues.apache.org/jira/browse/HBASE-24746?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang reassigned HBASE-24746: -- Assignee: Zheng Wang > The sort icons overlap the col name in master UI > - > > Key: HBASE-24746 > URL: https://issues.apache.org/jira/browse/HBASE-24746 > Project: HBase > Issue Type: Bug > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24746.png > > > In tables section, when description not few, the sort icons and col name are > overlap. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24746) The sort icons overlap the col name in master UI
[ https://issues.apache.org/jira/browse/HBASE-24746?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24746: --- Attachment: HBASE-24746.png > The sort icons overlap the col name in master UI > - > > Key: HBASE-24746 > URL: https://issues.apache.org/jira/browse/HBASE-24746 > Project: HBase > Issue Type: Bug > Components: UI >Reporter: Zheng Wang >Priority: Major > Attachments: HBASE-24746.png > > > In tables section, when description not few, the sort icons and col name are > overlap. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Created] (HBASE-24746) The sort icons overlap the col name in master UI
Zheng Wang created HBASE-24746: -- Summary: The sort icons overlap the col name in master UI Key: HBASE-24746 URL: https://issues.apache.org/jira/browse/HBASE-24746 Project: HBase Issue Type: Bug Components: UI Reporter: Zheng Wang In tables section, when description not few, the sort icons and col name are overlap. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24694) Support flush a single column family of table
[ https://issues.apache.org/jira/browse/HBASE-24694?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17157830#comment-17157830 ] Zheng Wang commented on HBASE-24694: As discussed in HBASE-24404, flush table was implemented through execProcedure(use pv1), and compact table was implemented in client side by iterate all its regions, should flush table go same way? BTW, there is a preFlushTable cp-hook, so maybe we still need call execProcedure(just trigger the cp). > Support flush a single column family of table > - > > Key: HBASE-24694 > URL: https://issues.apache.org/jira/browse/HBASE-24694 > Project: HBase > Issue Type: New Feature >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > This is follow-on work of HBASE-24404, do it as a seprate issue could make it > easier to reveiw. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24530) Introduce a split policy similar with SteppingSplitPolicy but count all store size
[ https://issues.apache.org/jira/browse/HBASE-24530?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17156506#comment-17156506 ] Zheng Wang commented on HBASE-24530: Done it with HBASE-24664 instead. [~wchevreuil] Could you help to close this issue and related PR, seems i cant do it, thanks. > Introduce a split policy similar with SteppingSplitPolicy but count all store > size > -- > > Key: HBASE-24530 > URL: https://issues.apache.org/jira/browse/HBASE-24530 > Project: HBase > Issue Type: Improvement > Components: regionserver >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > SteppingSplitPolicy decide whether split by max store size, if there are > several column families and write evenly, then the region will grows very big. > > In this case, count by all store size would be a better choise. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Updated] (HBASE-24704) Make the Table Schema easier to view even there are multiple families
[ https://issues.apache.org/jira/browse/HBASE-24704?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Zheng Wang updated HBASE-24704: --- Description: Currently we display the column family as vertical, it is not easy to view, my proposal is to display them as horizontal. (was: Currently we display the column family as vertical, it is no easy to view, my proposal is to display them as horizontal.) > Make the Table Schema easier to view even there are multiple families > - > > Key: HBASE-24704 > URL: https://issues.apache.org/jira/browse/HBASE-24704 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24704-afterpatch.png, HBASE-24704-beforepatch.png > > > Currently we display the column family as vertical, it is not easy to view, > my proposal is to display them as horizontal. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24586) Add table level locality in table.jsp
[ https://issues.apache.org/jira/browse/HBASE-24586?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17155962#comment-17155962 ] Zheng Wang commented on HBASE-24586: Thanks~ > Add table level locality in table.jsp > - > > Key: HBASE-24586 > URL: https://issues.apache.org/jira/browse/HBASE-24586 > Project: HBase > Issue Type: Improvement > Components: UI >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Attachments: HBASE-24586-afterpatch.png, HBASE-24586-beforepatch.png > > > For now, we only show locality for each region, it should be apply to table > as what Base Stats and Compactions does. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Commented] (HBASE-24382) Flush partial stores of region filtered by seqId when archive wal due to too many wals
[ https://issues.apache.org/jira/browse/HBASE-24382?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17155940#comment-17155940 ] Zheng Wang commented on HBASE-24382: I have push a new PR for branch-2, could you help to merge it? Thanks.[~stack] > Flush partial stores of region filtered by seqId when archive wal due to too > many wals > -- > > Key: HBASE-24382 > URL: https://issues.apache.org/jira/browse/HBASE-24382 > Project: HBase > Issue Type: Improvement > Components: wal >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1 > > > When the logRoller archive the oldest wal due to too many wals, if a region > should be flushed, we flush all stores of it, but it is not necessary, maybe > we can use unflushedSeqId of store to filter them. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24382) Flush partial stores of region filtered by seqId when archive wal due to too many wals
[ https://issues.apache.org/jira/browse/HBASE-24382?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17155940#comment-17155940 ] Zheng Wang edited comment on HBASE-24382 at 7/11/20, 10:37 AM: --- I have pushed a new PR for branch-2, could you help to merge it? Thanks.[~stack] was (Author: filtertip): I have push a new PR for branch-2, could you help to merge it? Thanks.[~stack] > Flush partial stores of region filtered by seqId when archive wal due to too > many wals > -- > > Key: HBASE-24382 > URL: https://issues.apache.org/jira/browse/HBASE-24382 > Project: HBase > Issue Type: Improvement > Components: wal >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > Fix For: 3.0.0-alpha-1 > > > When the logRoller archive the oldest wal due to too many wals, if a region > should be flushed, we flush all stores of it, but it is not necessary, maybe > we can use unflushedSeqId of store to filter them. -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Comment Edited] (HBASE-24404) Support flush a single column family of region
[ https://issues.apache.org/jira/browse/HBASE-24404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=17155909#comment-17155909 ] Zheng Wang edited comment on HBASE-24404 at 7/11/20, 7:56 AM: -- The HBASE-24382(support flush on CF level) has not backported to branch-2 yet, so let me do that first. was (Author: filtertip): The HBASE-24382(support flush multi CFs) has not backported to branch-2 yet, so let me do that first. > Support flush a single column family of region > -- > > Key: HBASE-24404 > URL: https://issues.apache.org/jira/browse/HBASE-24404 > Project: HBase > Issue Type: Improvement > Components: shell >Affects Versions: 3.0.0-alpha-1 >Reporter: Zheng Wang >Assignee: Zheng Wang >Priority: Major > > It is useful when we set many column families for a table, could avoid > generate too many small files. > -- This message was sent by Atlassian Jira (v8.3.4#803005)
