[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16276470#comment-16276470
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16271207#comment-16271207
]
Hudson commented on HBASE-19290:
FAILURE: Integrated in Jenkins build HBase-Trunk_matrix #4138 (See
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16271002#comment-16271002
]
Hudson commented on HBASE-19290:
FAILURE: Integrated in Jenkins build HBase-2.0 #936 (See
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16270571#comment-16270571
]
binlijin commented on HBASE-19290:
--
Push to master and branch-2
> Reduce zk request when doing split
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16270476#comment-16270476
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16270247#comment-16270247
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16269984#comment-16269984
]
binlijin commented on HBASE-19290:
--
HBASE-19290.master.006.patch fix test failed.
> Reduce zk request
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16268618#comment-16268618
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16268414#comment-16268414
]
binlijin commented on HBASE-19290:
--
bq.Many apologies for the wasted time! (will try to make up for it
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16268365#comment-16268365
]
Appy commented on HBASE-19290:
--
(facepalm) Okay, I accept my thick headedness here. Many apologies for the
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16268045#comment-16268045
]
binlijin commented on HBASE-19290:
--
bq. Can you please walk me through a full case explaining need of
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16267542#comment-16267542
]
Appy commented on HBASE-19290:
--
+1
I still don't get it. Am starting to feel thick head on this one :). But
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263840#comment-16263840
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263808#comment-16263808
]
Ted Yu commented on HBASE-19290:
bq. The WAL split speed was stable at 0.2TB/minute
The above is
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263773#comment-16263773
]
Yu Li commented on HBASE-19290:
---
Let me try to add more information:
Once when upgrading the HDFS version,
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263760#comment-16263760
]
binlijin commented on HBASE-19290:
--
{quote}
The patch is adding throttling, so it's certainly changing
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263748#comment-16263748
]
binlijin commented on HBASE-19290:
--
bq. The above example lasted for almost an hour.
With the
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263742#comment-16263742
]
Ted Yu commented on HBASE-19290:
Lijin:
The above example lasted for almost an hour.
With the patch,
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263729#comment-16263729
]
binlijin commented on HBASE-19290:
--
We observe that when the cluster is big, regionserver issue too much
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263720#comment-16263720
]
binlijin commented on HBASE-19290:
--
[~tedyu]
bq. Assuming patch v3 is very close to the version you run
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263699#comment-16263699
]
binlijin commented on HBASE-19290:
--
[~appy]
bq. Yeah, they are definitely clear, but Jingyun Tian's
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263187#comment-16263187
]
Ted Yu commented on HBASE-19290:
Reading the code with patch, it seems calculateAvailableSplitters()
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263094#comment-16263094
]
Appy commented on HBASE-19290:
--
[~tedyu] what kind of performance numbers are you looking for?
> Reduce zk
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263089#comment-16263089
]
Appy commented on HBASE-19290:
--
{quote}
bq. I can see it's doing that. The question really meant - why such
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263086#comment-16263086
]
Ted Yu commented on HBASE-19290:
Lijin:
Assuming patch v3 is very close to the version you run in the
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16263051#comment-16263051
]
Appy commented on HBASE-19290:
--
{quote}
bq. Since expectedTasksPerRS only relay on availableRSs and
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16261933#comment-16261933
]
binlijin commented on HBASE-19290:
--
bq. Since expectedTasksPerRS only relay on availableRSs and
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16261925#comment-16261925
]
binlijin commented on HBASE-19290:
--
bq. I can see it's doing that. The question really meant - why such
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16261597#comment-16261597
]
Appy commented on HBASE-19290:
--
{quote}
bq. So there are 2 available splitters, and one grabbed task, we
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16260541#comment-16260541
]
Jingyun Tian commented on HBASE-19290:
--
{code}
343 int expectedTasksPerRS = (numTasks /
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16260258#comment-16260258
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16260196#comment-16260196
]
binlijin commented on HBASE-19290:
--
bq. Why randomize? Can be constant?
No particular reason for
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16260121#comment-16260121
]
Appy commented on HBASE-19290:
--
bq. int sleepTime = RandomUtils.nextInt(0, 100) + 500;
Why randomize? Can be
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16259078#comment-16259078
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16258913#comment-16258913
]
binlijin commented on HBASE-19290:
--
bq. Add javadoc for availableRSs.
OK.
bq. Grabed -> Grabbed
OK.
bq.
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16258912#comment-16258912
]
binlijin commented on HBASE-19290:
--
bq. Please add a short commit message explaining 'idea' of the
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16257863#comment-16257863
]
Ted Yu commented on HBASE-19290:
{code}
+ * @param numTasks current total number of available tasks
+
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16257575#comment-16257575
]
Hadoop QA commented on HBASE-19290:
---
| (x) *{color:red}-1 overall{color}* |
\\
\\
|| Vote || Subsystem
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16256632#comment-16256632
]
Appy commented on HBASE-19290:
--
- Please add a short commit message explaining 'idea' of the change (makes
[
https://issues.apache.org/jira/browse/HBASE-19290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16256585#comment-16256585
]
binlijin commented on HBASE-19290:
--
The patch is effective and we have test it on our 2000+ nodes
40 matches
Mail list logo