You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mesos.apache.org by 王国栋 <wa...@gmail.com> on 2013/04/23 04:52:01 UTC
help about support/post-review tools
Hi all,
I am using support/post-review.py to submit my review. But the git log of
my commit seems wrong. I don't know the reason. Now, I can not update my
review request with this tool, since it can not parse the correct
review-request-id from git log.
The latest 2 commits of my branch is like the following.
Author: guodong wang <wa...@gmail.com>
Date: Tue Apr 23 10:16:39 2013 +0800
add some comments
commit a688741ae523abb2d83496daa6eae84809d2f2ad
Author: guodong wang <wa...@gmail.com>
Date: Mon Apr 22 10:41:15 2013 +0800
fix NullPointerException in MesosScheduler when using fairscheduler as
the underlying scheduler
Review: Created new window in existing browser session.
I think the green line is not the right log, Is it possible that
support/post-review.py generated the wrong Review information ?
I think I can modify the git log manually to bypass this problem. But still
want to know the reason of it.
Thanks. :)
Guodong
Re: help about support/post-review tools
Posted by 王国栋 <wa...@gmail.com>.
My post-review version is RBTools 0.5.
I delete the line "Review: Created new window in existing browser session.".
And try to update my review again. It works this time. And now, the git log
is OK. :)
Thanks to Vinod.
commit 04e61998f88d80aee247feb51ac64500fa733564
Author: guodong wang <wa...@gmail.com>
Date: Tue Apr 23 11:09:19 2013 +0800
add comment for fair scheduler
Review: https://reviews.apache.org/r/10695
commit f8e3e472c339ebca7d5e872d3c6c4d6ac1e3f979
Author: guodong wang <wa...@gmail.com>
Date: Mon Apr 22 10:41:15 2013 +0800
fix NullPointerException in MesosScheduler when using fairscheduler as
the underlying scheduler
Review: https://reviews.apache.org/r/10695
Guodong
On Tue, Apr 23, 2013 at 11:00 AM, Benjamin Mahler <benjamin.mahler@gmail.com
> wrote:
> That definitely looks like a bug! It looks like your version of post-review
> might have different output than the version we typically use. What version
> of post-review are you running?
>
> Here is an expected git log message:
>
> commit 741d67a65713a349450165d5f6cc9f916e04814d
> Author: Benjamin Mahler <bm...@twitter.com>
> Date: Mon Apr 15 16:10:44 2013 -0700
>
> Removed the hostname:port slave map from the Master.
>
> Review: https://reviews.apache.org/r/10534
>
>
> On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
>
> > Hi all,
> >
> > I am using support/post-review.py to submit my review. But the git log of
> > my commit seems wrong. I don't know the reason. Now, I can not update my
> > review request with this tool, since it can not parse the correct
> > review-request-id from git log.
> >
> > The latest 2 commits of my branch is like the following.
> >
> > Author: guodong wang <wa...@gmail.com>
> > Date: Tue Apr 23 10:16:39 2013 +0800
> >
> > add some comments
> >
> > commit a688741ae523abb2d83496daa6eae84809d2f2ad
> > Author: guodong wang <wa...@gmail.com>
> > Date: Mon Apr 22 10:41:15 2013 +0800
> >
> > fix NullPointerException in MesosScheduler when using fairscheduler
> as
> > the underlying scheduler
> >
> > Review: Created new window in existing browser session.
> >
> > I think the green line is not the right log, Is it possible that
> > support/post-review.py generated the wrong Review information ?
> >
> > I think I can modify the git log manually to bypass this problem. But
> still
> > want to know the reason of it.
> >
> > Thanks. :)
> >
> > Guodong
> >
>
Re: help about support/post-review tools
Posted by Benjamin Mahler <be...@gmail.com>.
Would love to hear a litte more about how you're using mesos!
Are you writing your own framework or have you been using the Hadoop
framework exclusively? Are you using other third-party frameworks?
Ben
On Tue, Apr 23, 2013 at 7:42 PM, 王国栋 <wa...@gmail.com> wrote:
> Hi Ben,
>
> Thanks a lot. I didn't notice the wrong patch in my second post-review. I
> did 2 commits. Sorry about that.
>
> My team is going to use mesos to deploy our service. And I hope to learn
> more details about Mesos. :)
>
> Guodong
>
>
> On Wed, Apr 24, 2013 at 1:47 AM, Benjamin Mahler
> <be...@gmail.com>wrote:
>
> > Hey Guodong, I'm going to get this submitted for you, thanks for the
> patch!
> > To make a commit you would need committer access to the project,
> something
> > we will definitely be considering as you help out and contribute further!
> >
> > I looked at your diff and it looked like you made two commits? With
> > post-reviews, there is a 1:1 commit:review ratio, so ReviewBoard shows
> your
> > patch only having the second commit:
> >
> > diff --git
> > a/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> > b/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> > index
> >
> >
> 3074bebee3ea4b907b3e24541490cd0d80dcd3e1..afe401f5265e3d9494af7eace42eec45943184a3
> > 100644
> > --- a/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> > +++ b/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> > @@ -224,6 +224,7 @@ public class MesosScheduler extends TaskScheduler
> > implements Scheduler {
> > // Let the underlying task scheduler do the actual task scheduling.
> > List<Task> tasks = taskScheduler.assignTasks(taskTracker);
> >
> > + // The Hadoop Fair Scheduler is known to return null.
> > if (tasks != null) {
> > // Keep track of which TaskTracker contains which tasks.
> > for (Task task : tasks) {
> >
> > In the future, we use 'git commit --amend' when looking to update an
> > existing review. Interactive rebasing also works when your commit is
> buried
> > in history.
> >
> > But don't worry about it this time! I'll fix this up manually :)
> >
> >
> > On Mon, Apr 22, 2013 at 8:21 PM, 王国栋 <wa...@gmail.com> wrote:
> >
> > > BTW: how can I submit my code to the remote repo?
> > >
> > > I see you guys use git-svn, but I use git to pull the repo out. Is it
> Ok
> > to
> > > merge my branch to trunk and push my commit to remote repo?
> > >
> > > Thanks.
> > >
> > > Guodong
> > >
> > >
> > > On Tue, Apr 23, 2013 at 11:03 AM, Vinod Kone <vi...@gmail.com>
> > wrote:
> > >
> > > > also, it should be safe to manually delete the bad "Review: ...."
> line
> > > from
> > > > your log message.
> > > >
> > > >
> > > > On Mon, Apr 22, 2013 at 8:00 PM, Benjamin Mahler
> > > > <be...@gmail.com>wrote:
> > > >
> > > > > That definitely looks like a bug! It looks like your version of
> > > > post-review
> > > > > might have different output than the version we typically use. What
> > > > version
> > > > > of post-review are you running?
> > > > >
> > > > > Here is an expected git log message:
> > > > >
> > > > > commit 741d67a65713a349450165d5f6cc9f916e04814d
> > > > > Author: Benjamin Mahler <bm...@twitter.com>
> > > > > Date: Mon Apr 15 16:10:44 2013 -0700
> > > > >
> > > > > Removed the hostname:port slave map from the Master.
> > > > >
> > > > > Review: https://reviews.apache.org/r/10534
> > > > >
> > > > >
> > > > > On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
> > > > >
> > > > > > Hi all,
> > > > > >
> > > > > > I am using support/post-review.py to submit my review. But the
> git
> > > log
> > > > of
> > > > > > my commit seems wrong. I don't know the reason. Now, I can not
> > update
> > > > my
> > > > > > review request with this tool, since it can not parse the correct
> > > > > > review-request-id from git log.
> > > > > >
> > > > > > The latest 2 commits of my branch is like the following.
> > > > > >
> > > > > > Author: guodong wang <wa...@gmail.com>
> > > > > > Date: Tue Apr 23 10:16:39 2013 +0800
> > > > > >
> > > > > > add some comments
> > > > > >
> > > > > > commit a688741ae523abb2d83496daa6eae84809d2f2ad
> > > > > > Author: guodong wang <wa...@gmail.com>
> > > > > > Date: Mon Apr 22 10:41:15 2013 +0800
> > > > > >
> > > > > > fix NullPointerException in MesosScheduler when using
> > > fairscheduler
> > > > > as
> > > > > > the underlying scheduler
> > > > > >
> > > > > > Review: Created new window in existing browser session.
> > > > > >
> > > > > > I think the green line is not the right log, Is it possible that
> > > > > > support/post-review.py generated the wrong Review information ?
> > > > > >
> > > > > > I think I can modify the git log manually to bypass this problem.
> > But
> > > > > still
> > > > > > want to know the reason of it.
> > > > > >
> > > > > > Thanks. :)
> > > > > >
> > > > > > Guodong
> > > > > >
> > > > >
> > > >
> > >
> >
>
Re: help about support/post-review tools
Posted by 王国栋 <wa...@gmail.com>.
Hi Ben,
Thanks a lot. I didn't notice the wrong patch in my second post-review. I
did 2 commits. Sorry about that.
My team is going to use mesos to deploy our service. And I hope to learn
more details about Mesos. :)
Guodong
On Wed, Apr 24, 2013 at 1:47 AM, Benjamin Mahler
<be...@gmail.com>wrote:
> Hey Guodong, I'm going to get this submitted for you, thanks for the patch!
> To make a commit you would need committer access to the project, something
> we will definitely be considering as you help out and contribute further!
>
> I looked at your diff and it looked like you made two commits? With
> post-reviews, there is a 1:1 commit:review ratio, so ReviewBoard shows your
> patch only having the second commit:
>
> diff --git
> a/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> b/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> index
>
> 3074bebee3ea4b907b3e24541490cd0d80dcd3e1..afe401f5265e3d9494af7eace42eec45943184a3
> 100644
> --- a/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> +++ b/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
> @@ -224,6 +224,7 @@ public class MesosScheduler extends TaskScheduler
> implements Scheduler {
> // Let the underlying task scheduler do the actual task scheduling.
> List<Task> tasks = taskScheduler.assignTasks(taskTracker);
>
> + // The Hadoop Fair Scheduler is known to return null.
> if (tasks != null) {
> // Keep track of which TaskTracker contains which tasks.
> for (Task task : tasks) {
>
> In the future, we use 'git commit --amend' when looking to update an
> existing review. Interactive rebasing also works when your commit is buried
> in history.
>
> But don't worry about it this time! I'll fix this up manually :)
>
>
> On Mon, Apr 22, 2013 at 8:21 PM, 王国栋 <wa...@gmail.com> wrote:
>
> > BTW: how can I submit my code to the remote repo?
> >
> > I see you guys use git-svn, but I use git to pull the repo out. Is it Ok
> to
> > merge my branch to trunk and push my commit to remote repo?
> >
> > Thanks.
> >
> > Guodong
> >
> >
> > On Tue, Apr 23, 2013 at 11:03 AM, Vinod Kone <vi...@gmail.com>
> wrote:
> >
> > > also, it should be safe to manually delete the bad "Review: ...." line
> > from
> > > your log message.
> > >
> > >
> > > On Mon, Apr 22, 2013 at 8:00 PM, Benjamin Mahler
> > > <be...@gmail.com>wrote:
> > >
> > > > That definitely looks like a bug! It looks like your version of
> > > post-review
> > > > might have different output than the version we typically use. What
> > > version
> > > > of post-review are you running?
> > > >
> > > > Here is an expected git log message:
> > > >
> > > > commit 741d67a65713a349450165d5f6cc9f916e04814d
> > > > Author: Benjamin Mahler <bm...@twitter.com>
> > > > Date: Mon Apr 15 16:10:44 2013 -0700
> > > >
> > > > Removed the hostname:port slave map from the Master.
> > > >
> > > > Review: https://reviews.apache.org/r/10534
> > > >
> > > >
> > > > On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
> > > >
> > > > > Hi all,
> > > > >
> > > > > I am using support/post-review.py to submit my review. But the git
> > log
> > > of
> > > > > my commit seems wrong. I don't know the reason. Now, I can not
> update
> > > my
> > > > > review request with this tool, since it can not parse the correct
> > > > > review-request-id from git log.
> > > > >
> > > > > The latest 2 commits of my branch is like the following.
> > > > >
> > > > > Author: guodong wang <wa...@gmail.com>
> > > > > Date: Tue Apr 23 10:16:39 2013 +0800
> > > > >
> > > > > add some comments
> > > > >
> > > > > commit a688741ae523abb2d83496daa6eae84809d2f2ad
> > > > > Author: guodong wang <wa...@gmail.com>
> > > > > Date: Mon Apr 22 10:41:15 2013 +0800
> > > > >
> > > > > fix NullPointerException in MesosScheduler when using
> > fairscheduler
> > > > as
> > > > > the underlying scheduler
> > > > >
> > > > > Review: Created new window in existing browser session.
> > > > >
> > > > > I think the green line is not the right log, Is it possible that
> > > > > support/post-review.py generated the wrong Review information ?
> > > > >
> > > > > I think I can modify the git log manually to bypass this problem.
> But
> > > > still
> > > > > want to know the reason of it.
> > > > >
> > > > > Thanks. :)
> > > > >
> > > > > Guodong
> > > > >
> > > >
> > >
> >
>
Re: help about support/post-review tools
Posted by Benjamin Mahler <be...@gmail.com>.
Hey Guodong, I'm going to get this submitted for you, thanks for the patch!
To make a commit you would need committer access to the project, something
we will definitely be considering as you help out and contribute further!
I looked at your diff and it looked like you made two commits? With
post-reviews, there is a 1:1 commit:review ratio, so ReviewBoard shows your
patch only having the second commit:
diff --git
a/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
b/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
index
3074bebee3ea4b907b3e24541490cd0d80dcd3e1..afe401f5265e3d9494af7eace42eec45943184a3
100644
--- a/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
+++ b/hadoop/mesos/src/java/org/apache/hadoop/mapred/MesosScheduler.java
@@ -224,6 +224,7 @@ public class MesosScheduler extends TaskScheduler
implements Scheduler {
// Let the underlying task scheduler do the actual task scheduling.
List<Task> tasks = taskScheduler.assignTasks(taskTracker);
+ // The Hadoop Fair Scheduler is known to return null.
if (tasks != null) {
// Keep track of which TaskTracker contains which tasks.
for (Task task : tasks) {
In the future, we use 'git commit --amend' when looking to update an
existing review. Interactive rebasing also works when your commit is buried
in history.
But don't worry about it this time! I'll fix this up manually :)
On Mon, Apr 22, 2013 at 8:21 PM, 王国栋 <wa...@gmail.com> wrote:
> BTW: how can I submit my code to the remote repo?
>
> I see you guys use git-svn, but I use git to pull the repo out. Is it Ok to
> merge my branch to trunk and push my commit to remote repo?
>
> Thanks.
>
> Guodong
>
>
> On Tue, Apr 23, 2013 at 11:03 AM, Vinod Kone <vi...@gmail.com> wrote:
>
> > also, it should be safe to manually delete the bad "Review: ...." line
> from
> > your log message.
> >
> >
> > On Mon, Apr 22, 2013 at 8:00 PM, Benjamin Mahler
> > <be...@gmail.com>wrote:
> >
> > > That definitely looks like a bug! It looks like your version of
> > post-review
> > > might have different output than the version we typically use. What
> > version
> > > of post-review are you running?
> > >
> > > Here is an expected git log message:
> > >
> > > commit 741d67a65713a349450165d5f6cc9f916e04814d
> > > Author: Benjamin Mahler <bm...@twitter.com>
> > > Date: Mon Apr 15 16:10:44 2013 -0700
> > >
> > > Removed the hostname:port slave map from the Master.
> > >
> > > Review: https://reviews.apache.org/r/10534
> > >
> > >
> > > On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
> > >
> > > > Hi all,
> > > >
> > > > I am using support/post-review.py to submit my review. But the git
> log
> > of
> > > > my commit seems wrong. I don't know the reason. Now, I can not update
> > my
> > > > review request with this tool, since it can not parse the correct
> > > > review-request-id from git log.
> > > >
> > > > The latest 2 commits of my branch is like the following.
> > > >
> > > > Author: guodong wang <wa...@gmail.com>
> > > > Date: Tue Apr 23 10:16:39 2013 +0800
> > > >
> > > > add some comments
> > > >
> > > > commit a688741ae523abb2d83496daa6eae84809d2f2ad
> > > > Author: guodong wang <wa...@gmail.com>
> > > > Date: Mon Apr 22 10:41:15 2013 +0800
> > > >
> > > > fix NullPointerException in MesosScheduler when using
> fairscheduler
> > > as
> > > > the underlying scheduler
> > > >
> > > > Review: Created new window in existing browser session.
> > > >
> > > > I think the green line is not the right log, Is it possible that
> > > > support/post-review.py generated the wrong Review information ?
> > > >
> > > > I think I can modify the git log manually to bypass this problem. But
> > > still
> > > > want to know the reason of it.
> > > >
> > > > Thanks. :)
> > > >
> > > > Guodong
> > > >
> > >
> >
>
Re: help about support/post-review tools
Posted by 王国栋 <wa...@gmail.com>.
BTW: how can I submit my code to the remote repo?
I see you guys use git-svn, but I use git to pull the repo out. Is it Ok to
merge my branch to trunk and push my commit to remote repo?
Thanks.
Guodong
On Tue, Apr 23, 2013 at 11:03 AM, Vinod Kone <vi...@gmail.com> wrote:
> also, it should be safe to manually delete the bad "Review: ...." line from
> your log message.
>
>
> On Mon, Apr 22, 2013 at 8:00 PM, Benjamin Mahler
> <be...@gmail.com>wrote:
>
> > That definitely looks like a bug! It looks like your version of
> post-review
> > might have different output than the version we typically use. What
> version
> > of post-review are you running?
> >
> > Here is an expected git log message:
> >
> > commit 741d67a65713a349450165d5f6cc9f916e04814d
> > Author: Benjamin Mahler <bm...@twitter.com>
> > Date: Mon Apr 15 16:10:44 2013 -0700
> >
> > Removed the hostname:port slave map from the Master.
> >
> > Review: https://reviews.apache.org/r/10534
> >
> >
> > On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
> >
> > > Hi all,
> > >
> > > I am using support/post-review.py to submit my review. But the git log
> of
> > > my commit seems wrong. I don't know the reason. Now, I can not update
> my
> > > review request with this tool, since it can not parse the correct
> > > review-request-id from git log.
> > >
> > > The latest 2 commits of my branch is like the following.
> > >
> > > Author: guodong wang <wa...@gmail.com>
> > > Date: Tue Apr 23 10:16:39 2013 +0800
> > >
> > > add some comments
> > >
> > > commit a688741ae523abb2d83496daa6eae84809d2f2ad
> > > Author: guodong wang <wa...@gmail.com>
> > > Date: Mon Apr 22 10:41:15 2013 +0800
> > >
> > > fix NullPointerException in MesosScheduler when using fairscheduler
> > as
> > > the underlying scheduler
> > >
> > > Review: Created new window in existing browser session.
> > >
> > > I think the green line is not the right log, Is it possible that
> > > support/post-review.py generated the wrong Review information ?
> > >
> > > I think I can modify the git log manually to bypass this problem. But
> > still
> > > want to know the reason of it.
> > >
> > > Thanks. :)
> > >
> > > Guodong
> > >
> >
>
Re: help about support/post-review tools
Posted by Vinod Kone <vi...@gmail.com>.
also, it should be safe to manually delete the bad "Review: ...." line from
your log message.
On Mon, Apr 22, 2013 at 8:00 PM, Benjamin Mahler
<be...@gmail.com>wrote:
> That definitely looks like a bug! It looks like your version of post-review
> might have different output than the version we typically use. What version
> of post-review are you running?
>
> Here is an expected git log message:
>
> commit 741d67a65713a349450165d5f6cc9f916e04814d
> Author: Benjamin Mahler <bm...@twitter.com>
> Date: Mon Apr 15 16:10:44 2013 -0700
>
> Removed the hostname:port slave map from the Master.
>
> Review: https://reviews.apache.org/r/10534
>
>
> On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
>
> > Hi all,
> >
> > I am using support/post-review.py to submit my review. But the git log of
> > my commit seems wrong. I don't know the reason. Now, I can not update my
> > review request with this tool, since it can not parse the correct
> > review-request-id from git log.
> >
> > The latest 2 commits of my branch is like the following.
> >
> > Author: guodong wang <wa...@gmail.com>
> > Date: Tue Apr 23 10:16:39 2013 +0800
> >
> > add some comments
> >
> > commit a688741ae523abb2d83496daa6eae84809d2f2ad
> > Author: guodong wang <wa...@gmail.com>
> > Date: Mon Apr 22 10:41:15 2013 +0800
> >
> > fix NullPointerException in MesosScheduler when using fairscheduler
> as
> > the underlying scheduler
> >
> > Review: Created new window in existing browser session.
> >
> > I think the green line is not the right log, Is it possible that
> > support/post-review.py generated the wrong Review information ?
> >
> > I think I can modify the git log manually to bypass this problem. But
> still
> > want to know the reason of it.
> >
> > Thanks. :)
> >
> > Guodong
> >
>
Re: help about support/post-review tools
Posted by Benjamin Mahler <be...@gmail.com>.
That definitely looks like a bug! It looks like your version of post-review
might have different output than the version we typically use. What version
of post-review are you running?
Here is an expected git log message:
commit 741d67a65713a349450165d5f6cc9f916e04814d
Author: Benjamin Mahler <bm...@twitter.com>
Date: Mon Apr 15 16:10:44 2013 -0700
Removed the hostname:port slave map from the Master.
Review: https://reviews.apache.org/r/10534
On Mon, Apr 22, 2013 at 7:52 PM, 王国栋 <wa...@gmail.com> wrote:
> Hi all,
>
> I am using support/post-review.py to submit my review. But the git log of
> my commit seems wrong. I don't know the reason. Now, I can not update my
> review request with this tool, since it can not parse the correct
> review-request-id from git log.
>
> The latest 2 commits of my branch is like the following.
>
> Author: guodong wang <wa...@gmail.com>
> Date: Tue Apr 23 10:16:39 2013 +0800
>
> add some comments
>
> commit a688741ae523abb2d83496daa6eae84809d2f2ad
> Author: guodong wang <wa...@gmail.com>
> Date: Mon Apr 22 10:41:15 2013 +0800
>
> fix NullPointerException in MesosScheduler when using fairscheduler as
> the underlying scheduler
>
> Review: Created new window in existing browser session.
>
> I think the green line is not the right log, Is it possible that
> support/post-review.py generated the wrong Review information ?
>
> I think I can modify the git log manually to bypass this problem. But still
> want to know the reason of it.
>
> Thanks. :)
>
> Guodong
>