,
Joshua Cohen
-mail. To reply, visit:
https://reviews.apache.org/r/25108/#review51709
---
On Aug. 27, 2014, 6:40 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply
bound for size (and an upper bound for overall vertical hiehgt) and
simply let each box represent multiple instances?
- Joshua Cohen
On Aug. 28, 2014, 6:13 p.m., David McLaughlin wrote:
---
This is an automatically generated e-mail
/test/sh/org/apache/aurora/e2e/test_end_to_end.sh
Unfortunately the v2 client end to end tests are currently broken (filed
https://issues.apache.org/jira/browse/AURORA-677), but I verified that the test
for aurora task run works in that script as well.
Thanks,
Joshua Cohen
boilerplate.
- Joshua Cohen
On Sept. 2, 2014, 4:36 p.m., Mark Chu-Carroll wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25255
On Sept. 2, 2014, 6:31 p.m., Joshua Cohen wrote:
src/test/java/org/apache/aurora/scheduler/app/local/FakeMaster.java, lines
134-135
https://reviews.apache.org/r/25257/diff/1/?file=674033#file674033line134
Any way to avoid magic numbers here (and would it be worthwhile
)?
- Joshua
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25206/#review52121
---
On Aug. 29, 2014, 11:17 p.m., Joshua Cohen wrote
out to a separate function
that's reused by both and documenting the reasoning there?
src/test/sh/org/apache/aurora/e2e/test_end_to_end_v2.sh
https://reviews.apache.org/r/25297/#comment90940
Kill trailing whitespace?
- Joshua Cohen
On Sept. 3, 2014, 2:07 p.m., Mark Chu-Carroll wrote
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25285/#review52186
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 3, 2014, 5:23 a.m
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25300/#review52193
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 3, 2014, 4:02 p.m
verified that the test
for aurora task run works in that script as well.
Thanks,
Joshua Cohen
/browse/AURORA-677), but I verified that the test
for aurora task run works in that script as well.
Thanks,
Joshua Cohen
to have
the e2e tests fixed before I ship https://reviews.apache.org/r/25208/
Joshua Cohen wrote:
It defaults to 'aurora run', for the aurora1 client, but we pass in
'aurora2 job run' from the test_end_to_end_v2.sh script. Are you saying to
switch the default (or maybe I should just remove
---
On Sept. 3, 2014, 6:33 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25206
for aurora task run works in that script as well.
Thanks,
Joshua Cohen
broken (filed
https://issues.apache.org/jira/browse/AURORA-677), but I verified that the test
for aurora task run works in that script as well.
Thanks,
Joshua Cohen
: https://reviews.apache.org/r/25319/diff/
Testing
---
Thanks,
Joshua Cohen
/apache/aurora/scheduler/TaskVars.java
https://reviews.apache.org/r/25357/#comment91130
use host (declared above) here in place of
task.getAssignedTask().getSlaveHost()?
- Joshua Cohen
On Sept. 4, 2014, 9:42 p.m., Maxim Khutornenko wrote
Thanks,
Joshua Cohen
/aurora/scheduler/http/slaves.st
defe1c5ececc8d742c78e52b9c43bebebba401ea
Diff: https://reviews.apache.org/r/25365/diff/
Testing
---
Ran scheduler, verified updated url.
Thanks,
Joshua Cohen
a get call, not sure if there's any downside there to
the implicit string conversion done by getProperty).
- Joshua Cohen
On Sept. 4, 2014, 11:16 p.m., Zameer Manji wrote:
---
This is an automatically generated e-mail. To reply, visit
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25366/#review52377
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 4, 2014, 11:36 p.m
/
Testing
---
gradle -Pq build
Thanks,
Joshua Cohen
(updated)
-
docs/contributing.md 44512c273e5f24c2b9352b3a817c2a1bb5996ce4
Diff: https://reviews.apache.org/r/25319/diff/
Testing
---
Thanks,
Joshua Cohen
https://reviews.apache.org/r/25398/#comment91463
Should these be @Positive instead (i.e. do we want to allow for a zero hold
time/jitter window)?
- Joshua Cohen
On Sept. 8, 2014, 5:54 p.m., Joshua Cohen wrote
---
gradle -Pq build
Thanks,
Joshua Cohen
---
gradle -Pq build
Thanks,
Joshua Cohen
.
- Joshua
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25398/#review52630
---
On Sept. 8, 2014, 9:09 p.m., Joshua Cohen wrote
/
Testing
---
gradle -Pq build
Thanks,
Joshua Cohen
/25398/diff/
Testing
---
gradle -Pq build
Thanks,
Joshua Cohen
, 2014, 4:54 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25398/
---
(Updated Sept. 9, 2014, 4:54 p.m
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25552/#review53123
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 11, 2014, 9:55 p.m
, is there danger in the addition of additional statuses in the future
rendering that assumption invalid?)
- Joshua Cohen
On Sept. 11, 2014, 4:53 a.m., Bill Farner wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
On Sept. 12, 2014, 5:09 p.m., Joe Smith wrote:
src/test/python/apache/aurora/client/cli/test_task_run.py, line 228
https://reviews.apache.org/r/25582/diff/1/?file=687672#file687672line228
Test the ssh command for proper behavior when no tasks are found
within a job or something,
about passing subset in as a param to isSubset
instead of picking it up via a closure?
- Joshua Cohen
On Sept. 9, 2014, 11:50 p.m., David McLaughlin wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
/JettyServerModule.java
https://reviews.apache.org/r/25593/#comment92730
Obviously not related, but should these all use Objects.requireNonNull
instead of Preconditions.checkNotNull?
- Joshua Cohen
On Sept. 12, 2014, 9:28 p.m., Kevin Sweeney wrote
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25593/#review53226
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 12, 2014, 9:43 p.m
now!
src/main/resources/org/apache/aurora/scheduler/http/ui/js/services.js
https://reviews.apache.org/r/25259/#comment93077
Nitpick: I'd name this subset param so it doesn't collide w/ the subset var
above, just for extra future-safety.
- Joshua Cohen
On Sept. 15, 2014, 6:53 p.m., David
the config validation logic from the
UpdateFactory creation?
- Joshua Cohen
On Sept. 15, 2014, 7:40 p.m., Bill Farner wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25529
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25721/#review53733
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 17, 2014, 5:49 p.m
/JobUpdateControllerImpl.java
https://reviews.apache.org/r/25791/#comment93729
Do we have any sort of guidelines on using Preconditions.checkNotNull vs
Objects.requireNonNull, or should it always be the latter?
- Joshua Cohen
On Sept. 18, 2014, 8:14 p.m., Bill Farner wrote
to swap out
responses by replacing the mock data on disk...).
- Joshua Cohen
On Sept. 18, 2014, 9:38 p.m., David McLaughlin wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25794
., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25835/
---
(Updated Sept. 19, 2014, 6:41 p.m.)
Review request
On Sept. 19, 2014, 6:58 p.m., Bill Farner wrote:
Ran locally, looks like assets are no longer gzip-encoded. Does that repro
for you?
Joshua Cohen wrote:
Hmm, I didn't verify that explicitly, but the gzip encoding tests passed,
so I assumed we were ok on that front. I'll take
UI in vagrant image.
One thing I've seen locally is lost tasks in vagrant, but I *think* that's just
due to my image being somehow misconfigured, and not these changes. I'm going
to destroy and rebuild the image to see if that cleans things up.
Thanks,
Joshua Cohen
line
(request for /js/angular.js returns a 404) even though it's passing in
IntelliJ. So that's neat.
- Joshua Cohen
On Sept. 19, 2014, 8:45 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit
works in scheduler UI in vagrant image.
One thing I've seen locally is lost tasks in vagrant, but I *think* that's just
due to my image being somehow misconfigured, and not these changes. I'm going
to destroy and rebuild the image to see if that cleans things up.
Thanks,
Joshua Cohen
files to the same
directory. You need to do a git clean and try again.
On Fri, Sep 19, 2014 at 2:24 PM, Joshua Cohen jco...@twopensource.com
wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org
to include
screenshots comparable to those in the ticket showing the proper alignment?
- Joshua Cohen
On Sept. 22, 2014, 11:23 p.m., David McLaughlin wrote:
---
This is an automatically generated e-mail. To reply, visit:
https
the
data source). Does angular have partials or something similar so we can avoid
replicating this?
- Joshua Cohen
On Sept. 23, 2014, 11:46 p.m., David McLaughlin wrote:
---
This is an automatically generated e-mail. To reply, visit
/org/apache/aurora/scheduler/http/ui/latestUpdates.html
https://reviews.apache.org/r/25963/#comment94608
Nit: should we parameterize this to maintain the previous distinction
between in progress vs completed?
- Joshua Cohen
On Sept. 24, 2014, 9:46 p.m., David McLaughlin wrote
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25963/#review54474
---
Ship it!
Ship It!
- Joshua Cohen
On Sept. 24, 2014, 10:52 p.m
of the desired behavior for
someone adding new code to a legacy class but not adding code coverage because
it means writing tests for the entire class as part of their change.
- Joshua Cohen
On Sept. 27, 2014, 11:25 p.m., Bill Farner wrote
for assets that were added mid-development).
The third commit is us forking the graphview resources from twitter commons
The fourth/final commit is the bulk of the actual code changes to get
everything working.
- Joshua Cohen
On Sept. 19, 2014, 9:39 p.m., Joshua Cohen wrote
On Sept. 26, 2014, 8:19 p.m., Joshua Cohen wrote:
src/main/resources/org/apache/aurora/scheduler/http/ui/js/services.js, line
440
https://reviews.apache.org/r/26095/diff/1/?file=706256#file706256line440
As far as I can see, allInstances is only used here to get the length,
can
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26095/#review54833
---
Ship it!
Looks good to me.
- Joshua Cohen
On Sept. 29, 2014, 4
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26123/#review54908
---
Ship it!
Thanks for clarifying the assert message.
- Joshua Cohen
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26269/#review55212
---
Ship it!
Ship It!
- Joshua Cohen
On Oct. 2, 2014, 3:48 p.m
://reviews.apache.org/r/26320/diff/
Testing
---
Committed file w/ precommit hook in place.
Thanks,
Joshua Cohen
, 9:39 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/25835/
---
(Updated Sept. 19, 2014, 9:39 p.m
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26320/#review55511
---
*ping* Kevin.
- Joshua Cohen
On Oct. 3, 2014, 4:51 p.m., Joshua
in... many
(all?) of these tests. Can you refactor to remove the repetition?
- Joshua Cohen
On Oct. 6, 2014, 2:57 p.m., Mark Chu-Carroll wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26363
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26378/#review55595
---
Ship it!
Ship It!
- Joshua Cohen
On Oct. 6, 2014, 7:53 p.m
c722eae2d04dec90e9c772f49c578184a2bdf76c
Diff: https://reviews.apache.org/r/26424/diff/
Testing
---
Added hokey test, ran hokey test.
Also verified the lack of http connection logs when running client commands
directly.
Thanks,
Joshua Cohen
commands
directly.
Thanks,
Joshua Cohen
I'm +1 on removing the time control as well. If you need to extend the
snooze you could always touch -m the snooze file?
On Wed, Oct 8, 2014 at 9:51 AM, Bill Farner wfar...@apache.org wrote:
On Oct. 6, 2014, 10:40 p.m., Brian Wickman wrote:
docs/configuration-reference.md, lines 359-360
On Oct. 6, 2014, 4:33 p.m., Joshua Cohen wrote:
*ping* Kevin.
*ping* again. Once this ships the static assets changes from
https://github.com/jcohen/incubator-aurora/commits/jcohen/static-assets
(reviewed at https://reviews.apache.org/r/25835/) should be mergeable.
- Joshua
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26320/#review55847
---
On Oct. 3, 2014, 4:51 p.m., Joshua Cohen wrote
On Oct. 6, 2014, 4:44 p.m., Joshua Cohen wrote:
src/test/python/apache/aurora/client/cli/test_update.py, lines 342-353
https://reviews.apache.org/r/26363/diff/1/?file=714123#file714123line342
This sequence of mocks and writing config to a file is repeated in...
many (all
that we're just
fixing a symptom of the problem, and not the problem itself (i.e. we're going
to be adding random classes to the legacy list despite their actually having
coverage).
- Joshua Cohen
On Oct. 8, 2014, 11:09 p.m., Zameer Manji wrote
On Oct. 8, 2014, 11:50 p.m., Joshua Cohen wrote:
This doesn't actually fail when I run it locally. I'm worried that we're
just fixing a symptom of the problem, and not the problem itself (i.e.
we're going to be adding random classes to the legacy list despite their
actually having
-Type: text/html
Vary: Accept-Encoding
Content-Length: 0
Server: Jetty(7.6.15.v20140411)
Thanks,
Joshua Cohen
On Oct. 8, 2014, 11:50 p.m., Joshua Cohen wrote:
This doesn't actually fail when I run it locally. I'm worried that we're
just fixing a symptom of the problem, and not the problem itself (i.e.
we're going to be adding random classes to the legacy list despite their
actually having
you see that one
as well?
- Bill
On October 9th, 2014, 12:15 a.m. UTC, Joshua Cohen wrote:
Review request for Aurora, Bill Farner and Zameer Manji.
By Joshua Cohen.
*Updated Oct. 9, 2014, 12:15 a.m.*
*Repository: * aurora
Description
Kill code to serve ApiBeta help pages that's
I can also verify that running the entire test suite in a debugger, a
breakpoint set in OfferQueue$OfferQueueImpl$2 is not hit.
On Thu, Oct 9, 2014 at 10:05 AM, Joshua Cohen jco...@twopensource.com
wrote:
Nope, it's passing for me, but it's similar to the behavior I was seeing
with ApiBeta$2
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26504/#review56011
---
Ship it!
Ship It!
- Joshua Cohen
On Oct. 9, 2014, 4:50 p.m
report
if needed.
On Thu, Oct 9, 2014 at 10:42 AM, Bill Farner wfar...@apache.org wrote:
Can you post your xml report to a ticket? It might help get to the bottom
of this flakiness.
On Thursday, October 9, 2014, Joshua Cohen jco...@twopensource.com
wrote:
I can also verify that running
I feel like we should err on the side of correctness here, rather than
simplicity? The dangers of someone accidentally leaving health checks
disabled indefinitely (on a service that has opted in to health checks) are
not insignificant.
On Thu, Oct 9, 2014 at 1:34 PM, Kevin Sweeney
On Oct. 9, 2014, 2:53 p.m., Bill Farner wrote:
src/main/python/apache/aurora/executor/common/health_checker.py, line 66
https://reviews.apache.org/r/26383/diff/3/?file=716355#file716355line66
FWIW i actually meant to suggest that the snooze has no concept of time
at all. If the
://reviews.apache.org/r/26531/#comment96529
This doesn't match the doc you sent out? Shouldn't this just be one of [OK,
STOP, ERRROR]? (JobUpdateStatus has a much wider set of values of which only
ERROR is included)
- Joshua Cohen
On Oct. 9, 2014, 11:23 p.m., Maxim Khutornenko wrote
to
externalize it[1], rather than having build.gradle take on more and more logic?
[1] http://www.gradle.org/docs/current/userguide/custom_tasks.html
- Joshua Cohen
On Oct. 10, 2014, 9:41 p.m., Bill Farner wrote
that just says int rather than
str?
- Joshua Cohen
On Oct. 14, 2014, 3:07 p.m., Mark Chu-Carroll wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26688
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26711/#review56594
---
Ship it!
Ship It!
- Joshua Cohen
On Oct. 14, 2014, 10:53 p.m
On Oct. 15, 2014, 12:02 a.m., Bill Farner wrote:
Sorry for the delay on this, i didn't realize i was on the hook to commit.
Now on master:
$ git log -1
commit 72fed752fcdbacf0f29e7eaef56c177f5ec9fcf0
Author: Joshua Cohen jco...@twopensource.com
Date: Tue Oct 14
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26664/#review56813
---
Ship it!
Ship It!
- Joshua Cohen
On Oct. 13, 2014, 11:22 p.m
Diff: https://reviews.apache.org/r/26790/diff/
Testing
---
./gradlew build -Pq
Also started up multiple schedulers in vagrant image and confirmed that I'm
properly redirected when accessing the non-leading scheduler.
Thanks,
Joshua Cohen
https://reviews.apache.org/r/26821/#comment97432
nit, move to previous line.
src/main/java/org/apache/aurora/scheduler/MesosSchedulerImpl.java
https://reviews.apache.org/r/26821/#comment97433
s/Augment/augment
- Joshua Cohen
On Oct. 16, 2014, 8:26 p.m., Zameer Manji wrote
://reviews.apache.org/r/26834/diff/
Testing
---
Rendered here:
https://github.com/jcohen/incubator-aurora/blob/jcohen/docs/clusters.json/docs/client-cluster-configuration.md
Thanks,
Joshua Cohen
On Oct. 16, 2014, 11:03 p.m., Joshua Cohen wrote:
docs/client-cluster-configuration.md, line 34
https://reviews.apache.org/r/26834/diff/1/?file=723472#file723472line34
I think an argument could be made for removing this configuration
entirely and just hardcoding
/incubator-aurora/blob/jcohen/docs/howtocontribute-git/docs/contributing.md
Thanks,
Joshua Cohen
6d63eb01c83b976f7183054afac48010d88558e9
Diff: https://reviews.apache.org/r/26841/diff/
Testing
---
https://github.com/jcohen/incubator-aurora/blob/jcohen/docs/howtocontribute-git/docs/contributing.md
Thanks,
Joshua Cohen
---
On Oct. 16, 2014, 11:58 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26834
go in the outcome of AURORA-834, so not your task.
Joshua Cohen wrote:
Well, I just added it, want me to kill it? ;)
(didn't see your follow up for some reason).
- Joshua
---
This is an automatically generated e-mail. To reply
://reviews.apache.org/r/26834/#review57064
---
On Oct. 17, 2014, 12:37 a.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r
-aurora/blob/jcohen/docs/clusters.json/docs/client-cluster-configuration.md
Thanks,
Joshua Cohen
seen users
having when parsing these error messages, it might be more clear that the
indented lines that follow are the message from the server if the previous line
ends in a colon?
Server reported error restarting job west/bozo/test/hello:
Job 'west/bozo/test/hello' not found.
- Joshua Cohen
/docs/client-cluster-configuration.md
Thanks,
Joshua Cohen
---
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/26834/#review57142
---
On Oct. 17, 2014, 8:23 p.m., Joshua Cohen wrote
generated e-mail. To reply, visit:
https://reviews.apache.org/r/26320/#review56991
---
On Oct. 3, 2014, 4:51 p.m., Joshua Cohen wrote:
---
This is an automatically generated e-mail
---
./gradlew build -Pq
Thanks,
Joshua Cohen
1 - 100 of 315 matches
Mail list logo
