Hi Li,
Why do you think the slave was OOM killed? Is there something that pointed
you to that conclusion? All I see is the slave launched an executor, and
the executor was killed by framework a few seconds after the task was
launched.
Also, what version are you running?
Ben
On Thu, Aug 22,
Looks like 0.13.0 does not have the fix cherry-picked. 0.14.0 is incoming
and will include the fix, until then you can cherry-pick the following
commit, although it may depend on additional patches!
commit 028d500c6d00023c8a56b37f885207adfd1e9a50
Author: Charles Reiss wog...@apache.org
Date:
:
Hi,
I'm happy to announce the passing of 0.14.0 vote with 3 +1 binding votes,
no -1 and no 0 votes:
+1:
Benjamin Hindman (binding)
Benjamin Mahler (binding)
Dave Lester (binding)
-1:
0:
Please find the release at http://www.apache.org/dist/mesos/0.14.0 (or
preferably please use
Benjamin Mahler benjamin.mah...@gmail.com
gpg: WARNING: This key is not certified with a trusted signature!
gpg: There is no indication that the signature belongs to the
owner.
Primary key fingerprint: E3A6 E5EF 7B67 C142 5B53 F072 D0BE BB95 D141 A5B6
[chipotle:~/tmp/apache-mesos-0.14.1
There could be an issue related to using the ':' character in your executor
id (which ultimately gets mapped to a path). Awhile back I filed this:
https://issues.apache.org/jira/browse/MESOS-361
That's all I have to go on with the given information, the slave log might
be more informative here,
Hi All,
I'd like to mention some changes that have been discussed amongst the
committers but have not yet been shared broadly with the list.
The central component of Mesos is the Master. The Master is responsible
for administering slaves, frameworks, and resource offers. It also handles
task
We should fix that so that it reconnects with Mesos after a restart of
Jenkins!
Can you file an issue for this?
On Thu, Nov 7, 2013 at 12:31 PM, Whitney Sorenson wsoren...@hubspot.comwrote:
I should also point out the scheduler didn't seem to survive a reboot of
Jenkins - I had to delete the
, Nov 7, 2013 at 5:10 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
We should fix that so that it reconnects with Mesos after a restart of
Jenkins!
Can you file an issue for this?
On Thu, Nov 7, 2013 at 12:31 PM, Whitney Sorenson
wsoren...@hubspot.comwrote:
I should also point out
In the longer term, utilization will be improved through oversubscription
of the slave's resources. This means, giving out resources inside existing
resource allocations if they remain unused. These over-subscribed
resources would be revokable at any time.
On Mon, Nov 25, 2013 at 12:29 PM, ricky
On Sun, Jan 12, 2014 at 9:14 PM, Ankur Chauhan an...@malloc64.com wrote:
Thanks everyone for all the help.
Marathon does seem like a good framework but my use case requires the app
to evaluate it's own health and scale up based on internal load stats (SLA
requirements) and I don't know if
Can you provide the commands you're using to run the master and slaves?
Can you provide the master and slave logs?
On Mon, Jan 13, 2014 at 6:07 PM, Sai Sagar jsaisa...@gmail.com wrote:
Hi all,
I am implementing a framework on the top of mesos. The framework is
registered successfully but
Running the master / slave with GLOG_v=2 in your environment may prove more
insightful.
Can you provide the full textual version of the logs instead of screenshots
please? :)
On Mon, Jan 13, 2014 at 8:36 PM, Sai Sagar jsaisa...@gmail.com wrote:
Hi,
Please find the attached images of master
I'm afraid that document is out of date, please ignore the comments related
to MESOS_HOME in the interim of us fixing that document.
https://issues.apache.org/jira/browse/MESOS-934
--log_dir is a parameter of the mesos-master and mesos-slave binaries.
The directory you linked is to an executor
On Wed, Jan 22, 2014 at 10:27 PM, coocood cooc...@gmail.com wrote:
I want to run redis server cluster on mesos, but have some problems.
The first problem is the storage path, since it is storage service, I need
to set the storage path out of the sandbox, so the next run of the service
will
I don't believe that we compile with C++11 on gcc 4.2, and C++11 support
did not land in 0.16.0 IIRC.
You should remove your -std=c++11 flag. Let us know if that does not work.
On Thu, Jan 30, 2014 at 11:40 AM, Tom Arnfeld t...@duedil.com wrote:
I'm trying to get going with Mesos to do a bit
]: *** [stl_logging_unittest-stl_logging_unittest.o] Error 1
#
--
Tom Arnfeld
Developer // DueDil
On 30 Jan 2014, at 19:46, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
I don't believe that we compile with C++11
Looks like we the 'm4' directory in stout was not added to the distribution:
$ ./bootstrap
...
m4/acx_pthread.m4:363: ACX_PTHREAD is expanded from...
configure.ac:87: the top level
autoreconf: configure.ac: adding subdirectory 3rdparty/stout to autoreconf
autoreconf: Entering directory
to track the fix?
Meanwhile, I don't think this should be a blocker for releasing this? If
you agree, I'm still looking for 2 more votes.
On Tue, Feb 4, 2014 at 12:47 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
0.15.0 has this same issue actually.
Probably we should either
Any link to the project? :)
On Fri, Feb 7, 2014 at 9:09 AM, Benjamin Hindman benjamin.hind...@gmail.com
wrote:
Very cool Tomas!
On Fri, Feb 7, 2014 at 8:15 AM, Tomas Barton barton.to...@gmail.comwrote:
Hi,
I wrote a Mesos framework for executing CI tasks from GitLab CI. The
GitLab
+jie who has been doing the 0.17.0 release
On Wed, Mar 5, 2014 at 11:44 PM, Chengwei Yang
chengwei.yang...@gmail.comwrote:
Hi List,
I'm new to mesos and trying to download one from
http://mesos.apache.org/downloads/
And it's saying that the most recent stable release is 0.16.0, however,
I have the status of TASK_COMPLETED being sent via the driver, followed by
a wait of about 5 secs
This is needed because of https://issues.apache.org/jira/browse/MESOS-243.
A 1 second sleep should be ample.
However, I'd still like to hear any thoughts on the approach of using one
task per
You can run N slaves on one machine, or you can run meta-slaves (slaves
within slaves). We've used meta-slaves in the past to run scaling
simulations as it is more accurate and easier than stubbing out the task
launching.
On Fri, Mar 21, 2014 at 12:55 PM, Sharma Podila spod...@netflix.com wrote:
Looks like --without-cxx11 is broken for gcc-4.2.1, not sure if that should
be a blocker for this because I don't believe there is a fix for this yet!
[bmahler@smf1-aye-26-sr4 mesos-0.18.1]$ ./configure --disable-optimize
--without-cxx11 make check -j8
...
libtool: compile: g++
You can set LIBPROCESS_PORT in the environment.
On Wed, May 14, 2014 at 1:58 PM, Scott Clasen sc...@heroku.com wrote:
I raised this question on the Spark ML but it may be more a Mesos question.
I would like to be able to configure the port used to communicate between
the Mesos master and
. What about tasks that Mesos is running for my
framework, but my framework lost track of them (there could be some
operational causes for this, even if we assume my code is bug free)? How
are frameworks handling such a scenario?
On Wed, May 14, 2014 at 4:05 PM, Benjamin Mahler
benjamin.mah
In addition to cpu and memory isolation, you will get process isolation.
With posix isolation, processes can escape from the slave (e.g. something
that double-forks and uses setsid).
On Mon, Jun 9, 2014 at 9:02 AM, Jie Yu yujie@gmail.com wrote:
Hi Dick,
what croup isolation provides
OS X (clang-503.0.40) and Ubuntu 13.10 (gcc 4.8.1)
Thanks for the hard work everyone - this is going to be a great release!
Niklas
On Thu, Jun 5, 2014 at 11:43 PM, Benjamin Mahler
benjamin.mah...@gmail.com
wrote:
Hi all,
Please vote on releasing the following candidate
On Mon, Jun 23, 2014 at 9:24 AM, Chris Burroughs chris.burrou...@gmail.com
wrote:
I have two related questions about the cpu and memory subsystems and at
most vs at least semantics.
## cpu
As I understand it, by default cpu.shares is used to give tasks at least
the specified cpu resources.
This is great! If only supporting 0.19.0+ I would recommend just collecting
from /metrics/snapshot because it obviates /stats.json.
On Thu, Jun 26, 2014 at 12:13 PM, Ray Rodriguez rayrod2...@gmail.com
wrote:
Hey everyone just thought I'd post a simple collectd python plugin that I
just
I created a ticket to track it for 0.19.1:
https://issues.apache.org/jira/browse/MESOS-1551
On Fri, Jun 27, 2014 at 8:26 AM, Benjamin Hindman
benjamin.hind...@gmail.com wrote:
Yes, this was fixed in this commit
https://github.com/apache/mesos/commit/1ce8d31fda545d69aea0637107f507c2b512adc9
Just to clarify:
You can run a single master (--quorum=1) if you are looking to experiment
and don't care about high availability.
You can run 3 masters (--quorum=2) if you want to remain operating with 1
machine being down (planned or unplanned). You can also operate in the face
of the complete
I've added it to the 0.19.1 list since it's trivial and helps those using
S3.
On Fri, Jul 4, 2014 at 12:52 PM, Tom Arnfeld t...@duedil.com wrote:
Happy to. It surprised me that this wasn't supported, especially
considering the fetcher is supposed to be able to download URIs from any
URL using
/ to
--work_dir the process crashes as well.
For ex: --work_dir=/mnt/data/mesos/
On Fri, Jun 27, 2014 at 2:02 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
I created a ticket to track it for 0.19.1:
https://issues.apache.org/jira/browse/MESOS-1551
On Fri, Jun 27, 2014 at 8:26 AM
Naming suggestion, let's call these pure language bindings. Native is
overloaded.
On Fri, Jul 11, 2014 at 8:40 AM, Niklas Nielsen nik...@mesosphere.io
wrote:
Embraced language repos is a great path too. +1 for not having to tie
automake into the respective language build systems.
Ensuring
Will also pull in MESOS-1538
https://issues.apache.org/jira/browse/MESOS-1538 for this.
On Wed, Jul 9, 2014 at 11:12 AM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
I've added it to the 0.19.1 list since it's trivial and helps those using
S3.
On Fri, Jul 4, 2014 at 12:52 PM, Tom
Hi all,
Please vote on releasing the following candidate as Apache Mesos 0.19.1.
0.19.1 includes the following:
Fixes a long standing critical bug in the JNI bindings that can lead to
framework unregistration.
vinodk...@gmail.com
wrote:
+1 (binding)
Tested on OSX Mavericks w/ gcc-4.8
On Mon, Jul 14, 2014 at 2:35 PM, Timothy Chen tnac...@gmail.com
wrote:
+1 (non-binding).
Tim
On Mon, Jul 14, 2014 at 2:32 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
Hi all
At the current time, you need an odd number of masters as there is an
assumption built into the replicated that the number of masters = 2*quorum
- 1. This assumption is present when bootstrapping the log from no data.
To recover from this, you need to run an odd number of masters, and set
your
Hey Whitney, sorry to hear you were bit by this!
I just realized that I misclassified MESOS-1406, the problem ran deeper
than I initially thought.
In particular, all of the boolean JSON values that were being exported
implicitly changed from 1/0 to true/false due to this change:
commit
Thanks for bringing this up! This is part of the ZK C library. We have seen
failing slaves with sporadic DNS lookup failures in our clusters.
After speaking to a ZK expert, I believe one of the things going into 3.5.0
is the ability to only need to resolve one of the zk hosts correctly, as
you
It might work to use
0.19 with a 0.20 mesos (or visa versa), but there be dragons =)
Is there a deprecation cycle? How should folks be upgrading Python
schedulers
and executors to 0.20.0 if they are not statically bundling libmesos?
Is there an upgrade order required?
We will need to
Just to confirm I understand correctly:
You have a framework, it needs to be a framework because it launches tasks
in the cluster. As part of this framework, you need to look into all of the
sandboxes of all currently running mesos tasks. This includes those tasks
that don't belong to your
Mesos also provides the ability to reserve resources, if you need
guarantees about the resources available to a particular framework.
For now, resources can be reserved at the per-slave level and they will
*only* be offered to the role that has them reserved.
On Mon, Aug 18, 2014 at 2:13 AM,
Hi all,
I wanted to take a moment to thank Alexandra Sava, who completed her OPW
internship this past week. We worked together in the second half of her
internship to create a design document for maintenance primitives in Mesos
(the original ticket is MESOS-1474
even when the deadline has passed if tasks
are still running? This is not explicit in the document and we want to make
sure operators have the information about this and could avoid unfortunate
rolling restarts.
On Aug 25, 2014 9:25 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Hi all
Looks like you're using the JVM, can you set all of your JVM flags to limit
the memory consumption? This would favor an OutOfMemoryError instead of
OOMing the cgroup.
On Thu, Aug 28, 2014 at 5:51 AM, Whitney Sorenson wsoren...@hubspot.com
wrote:
Recently, I've seen at least one case where a
a link between a previous offerID and this but then I saw the
Resource field. Wouldn't it be clearer to have InverseOfferID?
Thanks for the work! I really want to have these primitives.
On Aug 26, 2014 10:59 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
You're right, we don't account
to the
total usage of memory for the task. So you can't have the same amount of
memory for the task as you pass to java, -Xmx parameter.
On 2 September 2014 20:43, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Looks like you're using the JVM, can you set all of your JVM flags to
limit
Can you show us the the slave log and more of the master log?
There should be a TASK_LOST somewhere within them.
On Wed, Sep 17, 2014 at 10:43 AM, Luyi Wang wangluyi1...@gmail.com wrote:
Have anyone experience TASK_LOST status for storm tasks on mesos.
I checked the stderr. Everything seems
-185627-326871232-5050-8074-2/frameworks/20140915-230424-326871232-5050-13574-'
for gc 6.9960553185days in the future
-Luyi.
On Wed, Sep 17, 2014 at 12:56 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
Can you show us the the slave log and more of the master log
We want reconciliation to be a process that eventually terminates.
In = 0.19.0, the following two cases are conflated through no update being
sent:
(1) No state difference.
(2) Master temporarily cannot reply / dropped message.
As a result, a scheduler cannot determine when it is finished
that the taskStatus objects sent as a
result of calling reconciliation themselves are the exact same object as
was (possibly) sent before, or would they have updated timestamps?
On Tue, Sep 30, 2014 at 4:14 AM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
We want reconciliation to be a process
also
slaveId?) without knowing the current taskState or sending other metadata?
Thanks,
-Whitney
On Wed, Oct 1, 2014 at 8:33 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Updated timestamps will be generated.
On Tue, Sep 30, 2014 at 6:21 AM, Whitney Sorenson wsoren...@hubspot.com
Can you link to the full logs from make check?
P.S. This ended up in my spam folder.
On Tue, Oct 7, 2014 at 10:47 AM, Sammy Steele sammy_ste...@stanford.edu
wrote:
Hi,
I am trying to install the newest version of Mesos using Ubuntu 14.04 and
libsasl2-dev 2.1.25.dfsg1-17build1. When I run
But as a note, I would recommend against relying on updated timestamps,
because the clocks may not be synchronized in the system.
On Wed, Oct 1, 2014 at 12:33 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Updated timestamps will be generated.
On Tue, Sep 30, 2014 at 6:21 AM, Whitney
Hi all,
I've sent a review out for a document describing reconciliation, you can
see the draft here:
https://gist.github.com/bmahler/18409fc4f052df43f403
Would love to gather high level feedback on it from framework developers.
Feel free to reply here, or on the review:
status update was ACKed, but the scheduler failed over before
this information could be persisted. What task status (if any) does Mesos
respond with?
--
Connor Doyle
http://mesosphere.io
On Oct 15, 2014, at 14:05, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Hi all,
I've sent
Inline.
On Thu, Oct 16, 2014 at 7:43 PM, Sharma Podila spod...@netflix.com wrote:
Response inline, below.
On Thu, Oct 16, 2014 at 5:41 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
Thanks for the thoughtful questions, I will take these into account in
the document.
Addressing
for its task's either.
On Mon, Nov 3, 2014 at 6:15 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
I'm pretty confused by what's occurring in your scheduler, let's start
by looking at a particular task:
https://gist.github.com/bmahler/6f6bdb0385ec245b2346
You received an update from
Which version of the master are you using and do you have the logs? The
fact that no offers were coming back sounds like a bug!
As for using O1 after a disconnection, all offers are invalid once a
disconnection occurs. The scheduler driver does not automatically rescind
offers upon disconnection,
In addition to what Dick said, you need to make sure that you have a quorum
of masters *online* in order for a master to recover correctly. This means
you'll want to run the master under a tool (e.g. Monit) that restarts it
promptly upon failure.
You'll want to do this for the slaves as well.
On
. There is no longer a split between the planned schedule and the
actual draining, also useful for persistent frameworks.
The updated high level design is here:
https://docs.google.com/document/d/16k0lVwpSGVOyxPSyXKmGC-gbNmRlisNEe4p-fAUSojk/edit?usp=sharing
On Mon, Aug 25, 2014 at 12:24 PM, Benjamin Mahler
+till
FYI for Mac OS X 10.10.
Had to install subversion and apr:
$ brew tap homebrew/versions
$ brew tap homebrew/apache
$ brew install subversion apr
There is a flaky test in libprocess:
[ RUN ] IO.Write
make[5]: *** [check-local] Broken pipe: 13
Filed:
Here's the patch for 10.10 that we should cherry-pick:
commit 8adb36e3f72b575dea53013e7e790cb6c7957ae0
Author: Benjamin Mahler bmah...@twitter.com
Date: Tue Nov 11 15:11:32 2014 -0800
Added missing ZooKeeper patch file to the Makefile.
I'll mark https://issues.apache.org/jira/browse/MESOS
://gist.github.com/wsorenson/d2e49b96e84af86c9492
On Fri, Sep 12, 2014 at 9:12 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
+Ian
Sorry for the delay, when your cgroup OOMs a few things will occur:
(1) The kernel will notify mesos-slave about the OOM event.
(2) The kernel's OOM killer will pick
+1
Thanks Ian! Compiles on Mac OS X 10.10 now, all tests pass.
On Thu, Nov 13, 2014 at 1:07 PM, Cosmin Lehene cleh...@adobe.com wrote:
Ian,
Are there rpms in some public repo for this RC?
Thanks,
Cosmin
From: Ian Downes
It shouldn't be that high, especially with the size of the cluster I see in
your stats.
Which scheduler(s) are you running, and do they create large TaskInfo
objects? Just a hunch, as I do not recall any leaks in 0.19.1.
On Tue, Nov 18, 2014 at 1:00 AM, Tom Arnfeld t...@duedil.com wrote:
I've
On Thu, Nov 20, 2014 at 10:33 PM, Benjamin Mahler
benjamin.mah...@gmail.com wrote:
It shouldn't be that high, especially with the size of the cluster I see
in your stats.
Which scheduler(s) are you running, and do they create large TaskInfo
objects? Just a hunch, as I do not recall any
Benjamin Mahler benjamin.mah...@gmail.com:
I would like to be able to shutdown a mesos-slave for maintenance
without altering the current tasks.
What are you trying to do? If your maintenance operation does not affect
the tasks, why do you need to stop the slave in the first place?
On Wed
Sorry this is a bit of a tangent to the thread:
For ex, if the executor crashes Mesos would get a TASK_LOST while the
container might still be running.
Mesos should destroy the container when the executor exits, are you seeing
otherwise?
So we are doing something similar to Aurora's GC
in a
certain way, for example attach elastic network interfaces to containers,
mount ZFS volumes etc, and that's why we had to write a custom executor in
the first place.
On Mon, Dec 1, 2014 at 1:01 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Sorry this is a bit of a tangent
If anyone is interested in driving a 0.21.1 bug fix release, we could get
bug fixes out more quickly than waiting for 0.22.0.
On Tue, Dec 2, 2014 at 2:28 PM, Tim Chen t...@mesosphere.io wrote:
Hi Scott,
The patch for MESOS-1925 is already merged into master, so you should be
able to just
Sorry for the delay, yes it's possible but I don't think anyone has looked
at modularizing it. Hopefully others will speak up otherwise!
We have only one authorizer implementation in the code (a LocalAuthorizer
https://github.com/apache/mesos/blob/0.21.0/src/authorizer/authorizer.hpp#L64
that
Hey Thomas,
Could you share the scripts you're using to publish to pypi? It's not part
of the release process as of yet:
http://mesos.apache.org/documentation/latest/release-guide/
The 0.21.1 eggs were never published:
https://issues.apache.org/jira/browse/MESOS-2310
On Thu, Aug 14, 2014 at
It doesn't look like the exception here is being caught correctly:
https://github.com/apache/mesos/blob/0.21.1/3rdparty/libprocess/3rdparty/stout/include/stout/numify.hpp#L32
Just to be sure, can you show your compilation output, in particular, that
you're not seeing -fno-exceptions as a compiler
Hi all,
As part of https://issues.apache.org/jira/browse/MESOS-2347, there is a
scalability concern with the reconciliation API. Performing an implicit
reconciliation results in a status update being sent for each task in the
cluster. For large clusters in the tens of thousands of slaves, this
Hey Tim,
Can you release 0.21.1 on JIRA with the correct date?
https://issues.apache.org/jira/plugins/servlet/project-config/MESOS/versions
Thanks!
Ben
On Fri, Jan 2, 2015 at 12:30 PM, Dave Lester daveles...@gmail.com wrote:
Any of the recent blog posts are good templates, located in the
' is set.
On Fri, Feb 13, 2015 at 11:46 AM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Ok, don't see any issues with that approach, since executors will have to
create their own UUIDs anyway with pure language bindings. Will go back to
just updating TaskStatus to keep the API compatible
I posted a review of the proper documentation for this:
https://reviews.apache.org/r/29796/
On Tue, Jan 6, 2015 at 5:38 PM, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
Sorry for the delay, operational documentation for the replicated log has
been badly needed, I'll get some basic stuff
Sorry for the delay, operational documentation for the replicated log has
been badly needed, I'll get some basic stuff up on the website by next week.
In the interim, if you're using *--registry_strict=false (the default)*,
you can simply stop the original N masters, rm -rf all the data in
:35 AM, Kevin Sweeney kevi...@apache.org wrote:
Regarding the backwards-compatibility concern, would it make sense to add a
TaskStatusID field to the existing TaskStatus message instead of changing
the Scheduler signature?
On Friday, February 13, 2015, Benjamin Mahler benjamin.mah...@gmail.com
Ok, don't see any issues with that approach, since executors will have to
create their own UUIDs anyway with pure language bindings. Will go back to
just updating TaskStatus to keep the API compatible. Thanks Kevin!
On Fri, Feb 13, 2015 at 10:33 AM, Benjamin Mahler benjamin.mah...@gmail.com
Did you find what you were looking for?
At a quick glance, this kind of configuration is left to the framework
(Spark).
Mesos doesn't make any decisions with respect to how the resources offered
to Spark are being used.
On Tue, Dec 23, 2014 at 5:44 AM, Josh Devins j...@soundcloud.com wrote:
Guaranteeing fairness in such situations requires pre-emption of running
tasks/executors, which is not yet provided in mesos.
For now, you can try reserving a minimum amount of resources for each
framework, note however that this may reduce your efficiency if you
over-estimate the minimum
that, etc...
our plan is to log all tasks remotely to a fluentd or logstash agent
on each slave.
On 25 February 2015 at 18:05, Benjamin Mahler benjamin.mah...@gmail.com
wrote:
+drob
0.22.0 introduces a flag on the master / slave called
--external_log_file.
This allows you to point
Great to see so many proposals!
Is it intentional that we have to review them in small subsets? It's hard
to tell what to consider as an Average proposal when you can only see a
small subset at a time. Just curious on the reasoning behind that.
On Wed, Feb 18, 2015 at 2:44 PM, Dave Lester
$ sudo apt-get install libsvn-dev
From here: http://mesos.apache.org/gettingstarted/
On Thu, Jan 29, 2015 at 1:35 PM, Dan Dong dongda...@gmail.com wrote:
Hi,
When I tried to build mesos-0.21.0 on Ubuntu-14.04, I get this error:
This message can be a bit misleading, do you have perf installed?
On Thu, Apr 30, 2015 at 11:18 AM, Brian Topping brian.topp...@gmail.com
wrote:
Getting closer. After finding
http://garyzhu.net/notes/CentOS7-Systemd-Mesos-Marathon.html, I set up
another new CentOS 7 machine, got a lot further
Fixed.
On Mon, May 4, 2015 at 12:56 AM, Ryan Thomas r.n.tho...@gmail.com wrote:
Whilst this is a bit old, the docs here
http://mesos.apache.org/blog/mesos-0-21-0-released/ for 0.21.0 link to
the wrong ticket for the shared filesystem isolator.
Cheers,
ryan
Glad to hear it, are you able to share a little more with the list about how
you're using mesos? :)
Sent from my iPhone
On May 7, 2015, at 10:26 AM, John Miller john.mil...@brigade.com wrote:
We're utilizing Mesos within our organization for multiple projects.
Anyone with access please
It sounds like that is expected, since it says nonexistent:
http://lists.gnu.org/archive/html/autoconf/2011-03/msg9.html
On Sat, May 16, 2015 at 2:38 PM, Joerg Maurer dev-ma...@gmx.net wrote:
Hello,
I am building from source git clone --branch 0.22.1 --depth 1
If that's really what you're seeing, it is a bug and a very surprising one,
so please provide evidence :)
See the detailed description here:
http://mesos.apache.org/api/latest/c++/classmesos_1_1Scheduler.html
The scheduler driver will serially invoke methods on your Scheduler
implementation.
The existing Mesos API provides unreliable messaging passing for framework
- executor communication:
-- Schedulers can call 'sendFrameworkMessage(executor, slave, data)' on
the driver [1], this sends a message to the executor. This has a
best-effort optimization to bypass the master, and send the
Should be removed from the website now.
On Wed, Jun 17, 2015 at 7:21 AM, Ken Sipe kens...@gmail.com wrote:
thanks! https://github.com/apache/mesos/pull/46
On Jun 17, 2015, at 5:22 AM, Brian Candler b.cand...@pobox.com wrote:
At
The offer will contain the running executors on that slave:
https://github.com/apache/mesos/blob/0.22.1/include/mesos/mesos.proto#L619
However, there are a few bugs to note if you're planning to use this in any
kind of production setup:
(1) If an executor exits, or a new one is started, we won't
Interesting to see HTCondor has a defragmentation feature, this kind of
thing has come up before for Mesos as well.
Specifically, adding Inverse Offers as a generic mechanism for obtaining
resources back from a framework unlocks a lot of functionality. The first
use case was cluster maintenance.
This is because stop() is asynchronously processed. A message is sent to
the scheduler process and it will eventually send the message to the
master. This is why you've noticed that sleeping helps to ensure that this
occurs.
There is no scheduler driver specific issue for this, but the executor
Hey all!
I'm an engineer at Twitter and Mesos committer / PMC member. Looking
forward to seeing some of the new faces this year and re-connect with old
friends :)
I'll be talking about the upcoming maintenance primitives in mesos
http://sched.co/35Cu, which is critical for operating
Currently, recovered resources are not immediately re-offered as you
noticed, and the default allocation interval is 1 second. I'd recommend
lowering that (e.g. --allocation_interval=50ms), that should improve the
second bullet you listed. Although, in your case it would be better to
immediately
there. I will try to find some logs that provide some insight into the
execution times.
I am using a command task. I haven't looked into executors yet; I had a
hard time finding some examples in my language (Scala).
On Fri, Jul 17, 2015 at 2:00 PM, Benjamin Mahler
benjamin.mah...@gmail.com
1 - 100 of 317 matches
Mail list logo