Re: Total resources available versus actually available

[Benjamin Mahler]

Mesos doesn't know anything about non-Mesos processes running on the
machine consuming resources. So yes, Mesos would still offer the resources
and allow a task to be placed there. The assumption is that all of the
resources specified via --resources are available for Mesos to allocate.

On Mon, Feb 9, 2015 at 2:34 AM, craig w  wrote:

> Mesos slaves report the total resources they have available, such as 2
> CPU, 8GB Memory and 100GB disk.
>
> Does Mesos keep track of the amount of resources that are actually
> available to best schedule tasks?
>
> For example, imagine a slave has some other processes running on it (that
> are not mesos tasks) that are taking up 6GB of RAM (out of the total 8GB).
> A new task is created in Mesos that needs 4GB of RAM, would Mesos still try
> to put the new task on the slave?
>

Final week for #MesosCon 2015 CFP and early-bird registration

[Dave Lester]

Just a friendly-reminder, two deadlines are fast-approaching for
#MesosCon on this upcoming Saturday, February 14th:

The call for speaking proposals:
http://events.linuxfoundation.org/events/mesoscon/program/cfp

Early-bird registration:
http://events.linuxfoundation.org/events/mesoscon/attend/register

MesosCon is an annual conference organized by the Apache Mesos
community, bringing together users and developers to share and learn
about the project and its growing ecosystem. This year's conference will
take place in Seattle, WA, on August 20-21, 2015.

Are you using Mesos in production and do you have an interesting,
challenging, or unique use case? Have you been contributing to the
project core? Have you developed a custom framework, or used one in a
novel way? How have you integrated Mesos into other parts of your stack?
What have you learned as a result of deploying Mesos in production? How
are you using Docker containers with Apache Mesos? I hope these
questions help illustrate some of the topics we hope speakers will
explore in their talks. If you have any questions as you prepare your
talks, feel free to reach out to me. The CFP closes February 14, 2015 at
11:55PM, PST.

Huge thanks, I'll see you at #MesosCon!

Dave

Re: Total resources available versus actually available

[craig w]

Is this the issue? https://issues.apache.org/jira/browse/MESOS-354

On Mon, Feb 9, 2015 at 12:30 PM, Niklas Nielsen 
wrote:

> The slave picks up total available resources (not free) on startup, but
> this can be overridden by the --resources="" flag.
> That way, you can leave resources for your out of bound processes.
>
> If you want to compute the slack (difference between allocated and
> actually used), you can compute that from the slave's
> /monitor/statistics.json endpoint.
> You should be able to pick up some documentation on that endpoint from
> /help/monitor/statistics.json
>
> If you are looking for something more dynamic, you should follow the
> discussion on oversubscription :)
>
> Cheers,
> Niklas
>
> On 9 February 2015 at 02:34, craig w  wrote:
>
>> Mesos slaves report the total resources they have available, such as 2
>> CPU, 8GB Memory and 100GB disk.
>>
>> Does Mesos keep track of the amount of resources that are actually
>> available to best schedule tasks?
>>
>> For example, imagine a slave has some other processes running on it (that
>> are not mesos tasks) that are taking up 6GB of RAM (out of the total 8GB).
>> A new task is created in Mesos that needs 4GB of RAM, would Mesos still try
>> to put the new task on the slave?
>>
>
>


-- 

https://github.com/mindscratch
https://www.google.com/+CraigWickesser
https://twitter.com/mind_scratch
https://twitter.com/craig_links

Re: Total resources available versus actually available

[Niklas Nielsen]

The slave picks up total available resources (not free) on startup, but
this can be overridden by the --resources="" flag.
That way, you can leave resources for your out of bound processes.

If you want to compute the slack (difference between allocated and actually
used), you can compute that from the slave's /monitor/statistics.json
endpoint.
You should be able to pick up some documentation on that endpoint from
/help/monitor/statistics.json

If you are looking for something more dynamic, you should follow the
discussion on oversubscription :)

Cheers,
Niklas

On 9 February 2015 at 02:34, craig w  wrote:

> Mesos slaves report the total resources they have available, such as 2
> CPU, 8GB Memory and 100GB disk.
>
> Does Mesos keep track of the amount of resources that are actually
> available to best schedule tasks?
>
> For example, imagine a slave has some other processes running on it (that
> are not mesos tasks) that are taking up 6GB of RAM (out of the total 8GB).
> A new task is created in Mesos that needs 4GB of RAM, would Mesos still try
> to put the new task on the slave?
>

Re: Mesos Marathon weird behavior

[Dario Rexin]

It should be in the Mesosphere repos since Thursday last week.

> On 09 Feb 2015, at 17:02, Emre Tekelioğlu  wrote:
> 
> I installed mesos from repo, did you know when ubuntu repo will update?
> 
> 2015-02-09 17:46 GMT+02:00 Dario Rexin  >:
> This looks normal. But could you try updating to Marathon 0.8.0? We fixed a 
> couple of deployment related bugs in there.
> 
> 
>> On 09 Feb 2015, at 16:38, Emre Tekelioğlu > > wrote:
>> 
>> Hi Dario,
>> 
>> I found these logs in syslog.
>> Is it normal?
>> 
>> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,537] INFO 
>> Computed new deployment plan: DeploymentPlan(2015-02-09T15:26:38.437Z, 
>> (Step(Vector(Start(App(/testtest1, None)), 0))), 
>> Step(List(Scale(App(/testtest1, None)), 5) 
>> (mesosphere.marathon.upgrade.DeploymentPlan$:246)
>> Feb  9 15:26:38 maas01 marathon[8808]: [INFO] [02/09/2015 15:26:38.539] 
>> [marathon-akka.actor.default-dispatcher-14] 
>> [akka://marathon/user/MarathonScheduler/$a/UpgradeManager/5cf28be2-af8e-427d-90a2-5b0388b7009a/$a
>>  <>] Successfully started 0 instances of /testtest1
>> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO 
>> Starting app /testtest1 (mesosphere.marathon.SchedulerActions:363)
>> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO 
>> Already running 0 instances of /testtest1. Not scaling. 
>> (mesosphere.marathon.SchedulerActions:512)
>> 
>> 
>> 2015-02-09 16:10 GMT+02:00 Dario Rexin > >:
>> Could you please post the relevant log output of Mesos and Marathon?
>> 
>>> On 09 Feb 2015, at 15:06, Emre Tekelioğlu >> > wrote:
>>> 
>>> I added the app with REST API, but nothing changed.
>>> Status stuck in "Deploying"
>>> I also run these commands;
>>> 
>>> echo 'docker,mesos' > /etc/mesos-slave/containerizers
>>> echo '5mins' > /etc/mesos-slave/executor_registration_timeout
>>> 
>>> App config is like that
>>> 
>>> {  
>>>"id":"/test2",
>>>"cmd":null,
>>>"args":[  
>>> 
>>>],
>>>"user":null,
>>>"env":{  
>>> 
>>>},
>>>"instances":5,
>>>"cpus":0.1,
>>>"mem":128.0,
>>>"disk":512.0,
>>>"executor":"",
>>>"constraints":[  
>>> 
>>>],
>>>"uris":[  
>>> 
>>>],
>>>"storeUrls":[  
>>> 
>>>],
>>>"ports":[  
>>>   10001
>>>],
>>>"requirePorts":false,
>>>"backoffSeconds":1,
>>>"backoffFactor":1.15,
>>>"maxLaunchDelaySeconds":3600,
>>>"container":{  
>>>   "type":"DOCKER",
>>>   "volumes":[  
>>> 
>>>   ],
>>>   "docker":{  
>>>  "image":"fehmicansaglam/progressed.io ",
>>>  "network":null,
>>>  "portMappings":null,
>>>  "privileged":false,
>>>  "parameters":[  
>>> 
>>>  ]
>>>   }
>>>},
>>>"healthChecks":[  
>>> 
>>>],
>>>"dependencies":[  
>>> 
>>>],
>>>"upgradeStrategy":{  
>>>   "minimumHealthCapacity":1.0,
>>>   "maximumOverCapacity":1.0
>>>},
>>>"labels":{  
>>> 
>>>},
>>>"version":"2015-02-09T13:55:52.577Z"
>>> }
>>> 
>>> 
>>> Thanks
>>> 
>>> 
>>> 2015-02-09 15:43 GMT+02:00 Dario Rexin >> >:
>>> Yes, it’s currently not possible to deploy containers with the web UI.
>>> 
 On 09 Feb 2015, at 14:42, Emre Tekelioğlu >>> > wrote:
 
 Hi Dario,
 
 I couldn't find the container section in my Marathon UI when i click New 
 App, should i use only REST API to do that?
 
 Regards,
 Emre
 
 2015-02-09 15:09 GMT+02:00 Dario Rexin >>> >:
 Hi Emre,
 
 Mesos and Marathon have full support for running Docker containers, you 
 can just specify them in the app definition, no need to run `docker run …` 
 as a command. Try this app definition instead:
 
 {  
   "id":"/progressed-io",
   "instances":5,
   "cpus":0.1,
   "mem":128.0,
   "disk":512.0,
   “container”: {
  “docker”: {
 “image”: “fehmicansaglam/progressed.io "
   }
   }
 }
 
 Please make sure to start your mesos-slave process with 
 “—containerizers=docker,mesos”
 
 Hope this helps.
 
 Cheers,
 Dario
 
> On 09 Feb 2015, at 14:02, Emre Tekelioğlu  > wrote:
> 
> Hi Dario
> 
> Mesos version 0.21.1, Marathon version 0.7.6.
> App config;
> {  
>"id":"/progressed-io",
>"cmd":"sudo docker run fehmicansaglam/progressed.io 
> ",
>"args":null,
>"user":null,
>"env":{  
> 
>},
>"instances":5,
>"cpus":0.1,
>"mem":128.0,
>"disk":512.0,
>"executor":"",
>"constraints":[  
> 
>],
>"uris":[  

Re: Mesos Marathon weird behavior

[Emre Tekelioğlu]

I installed mesos from repo, did you know when ubuntu repo will update?

2015-02-09 17:46 GMT+02:00 Dario Rexin :

> This looks normal. But could you try updating to Marathon 0.8.0? We fixed
> a couple of deployment related bugs in there.
>
>
> On 09 Feb 2015, at 16:38, Emre Tekelioğlu 
> wrote:
>
> Hi Dario,
>
> I found these logs in syslog.
> Is it normal?
>
> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,537] INFO
> Computed new deployment plan: DeploymentPlan(2015-02-09T15:26:38.437Z,
> (Step(Vector(Start(App(/testtest1, None)), 0))),
> Step(List(Scale(App(/testtest1, None)), 5)
> (mesosphere.marathon.upgrade.DeploymentPlan$:246)
> Feb  9 15:26:38 maas01 marathon[8808]: [INFO] [02/09/2015 15:26:38.539]
> [marathon-akka.actor.default-dispatcher-14] [
> akka://marathon/user/MarathonScheduler/$a/UpgradeManager/5cf28be2-af8e-427d-90a2-5b0388b7009a/$a]
> Successfully started 0 instances of /testtest1
> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO
> Starting app /testtest1 (mesosphere.marathon.SchedulerActions:363)
> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO
> Already running 0 instances of /testtest1. Not scaling.
> (mesosphere.marathon.SchedulerActions:512)
>
>
> 2015-02-09 16:10 GMT+02:00 Dario Rexin :
>
>> Could you please post the relevant log output of Mesos and Marathon?
>>
>> On 09 Feb 2015, at 15:06, Emre Tekelioğlu 
>> wrote:
>>
>> I added the app with REST API, but nothing changed.
>> Status stuck in "Deploying"
>> I also run these commands;
>>
>> echo 'docker,mesos' > /etc/mesos-slave/containerizers
>> echo '5mins' > /etc/mesos-slave/executor_registration_timeout
>>
>> App config is like that
>>
>> {
>>"id":"/test2",
>>"cmd":null,
>>"args":[
>>
>>],
>>"user":null,
>>"env":{
>>
>>},
>>"instances":5,
>>"cpus":0.1,
>>"mem":128.0,
>>"disk":512.0,
>>"executor":"",
>>"constraints":[
>>
>>],
>>"uris":[
>>
>>],
>>"storeUrls":[
>>
>>],
>>"ports":[
>>   10001
>>],
>>"requirePorts":false,
>>"backoffSeconds":1,
>>"backoffFactor":1.15,
>>"maxLaunchDelaySeconds":3600,
>>"container":{
>>   "type":"DOCKER",
>>   "volumes":[
>>
>>   ],
>>   "docker":{
>>  "image":"fehmicansaglam/progressed.io",
>>  "network":null,
>>  "portMappings":null,
>>  "privileged":false,
>>  "parameters":[
>>
>>  ]
>>   }
>>},
>>"healthChecks":[
>>
>>],
>>"dependencies":[
>>
>>],
>>"upgradeStrategy":{
>>   "minimumHealthCapacity":1.0,
>>   "maximumOverCapacity":1.0
>>},
>>"labels":{
>>
>>},
>>"version":"2015-02-09T13:55:52.577Z"
>> }
>>
>>
>> Thanks
>>
>>
>> 2015-02-09 15:43 GMT+02:00 Dario Rexin :
>>
>>> Yes, it’s currently not possible to deploy containers with the web UI.
>>>
>>> On 09 Feb 2015, at 14:42, Emre Tekelioğlu 
>>> wrote:
>>>
>>> Hi Dario,
>>>
>>> I couldn't find the container section in my Marathon UI when i click New
>>> App, should i use only REST API to do that?
>>>
>>> Regards,
>>> Emre
>>>
>>> 2015-02-09 15:09 GMT+02:00 Dario Rexin :
>>>
 Hi Emre,

 Mesos and Marathon have full support for running Docker containers, you
 can just specify them in the app definition, no need to run `docker run …`
 as a command. Try this app definition instead:

 {
   "id":"/progressed-io",
   "instances":5,
   "cpus":0.1,
   "mem":128.0,
   "disk":512.0,
   “container”: {
  “docker”: {
 “image”: “fehmicansaglam/progressed.io"
   }
   }
 }

 Please make sure to start your mesos-slave process with
 “—containerizers=docker,mesos”

 Hope this helps.

 Cheers,
 Dario

 On 09 Feb 2015, at 14:02, Emre Tekelioğlu 
 wrote:

 Hi Dario

 Mesos version 0.21.1, Marathon version 0.7.6.
 App config;
 {
"id":"/progressed-io",
"cmd":"sudo docker run fehmicansaglam/progressed.io",
"args":null,
"user":null,
"env":{

},
"instances":5,
"cpus":0.1,
"mem":128.0,
"disk":512.0,
"executor":"",
"constraints":[

],
"uris":[

],
"storeUrls":[

],
"ports":[

],
"requirePorts":false,
"backoffSeconds":1,
"backoffFactor":1.15,
"container":null,
"healthChecks":[

],
"dependencies":[

],
"upgradeStrategy":{
   "minimumHealthCapacity":1.0
}


 Thanks,
 Emre

 2015-02-09 14:52 GMT+02:00 Dario Rexin :

> Hi Emre,
>
> what are the versions of Mesos and Marathon and what does your app
> definition look like?
>
> Cheers,
> Dario
>
> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu 
> wrote:
> >
> > Hi,

Re: Mesos Marathon weird behavior

[Emre Tekelioğlu]

Yes, they are on the samebox.

2015-02-09 17:47 GMT+02:00 David Kesler :

>  Are you running marathon and mesos on the same box?  If not, check the
> frameworks  page in mesos.  If marathon has a ‘re-registered’ value that is
> always ‘just now’, you’re suffering from the issue described in
> https://github.com/mesosphere/marathon/issues/1140 (either due to ports
> being blocked or due to marathon registering with mesos as 127.0.0.1).
>
>
>
> *From:* Emre Tekelioğlu [mailto:emre.tekelio...@gmail.com]
> *Sent:* Monday, February 09, 2015 7:50 AM
> *To:* user@mesos.apache.org
> *Subject:* Mesos Marathon weird behavior
>
>
>
> Hi,
>
> I have a single node mesos cluster on Ubuntu 14.04 host machine, with one
> node mesos-master and one node mesos-slave.
>
> I try to deploy and run my dockerized services with marathon but i have a
> weird problem.
>
>
>
> For example If i start a new app with 5 instance, only one task start and
> the status stuck in "deploying". If i restart the mesos-master, then the
> second task start to run.
>
>
>
> Each restart for mesos-master cause an increment of a task.
>
>
>
> Any idea?
>
>
>
> Regards,
>
> Emre
>

RE: Mesos Marathon weird behavior

[David Kesler]

Are you running marathon and mesos on the same box?  If not, check the 
frameworks  page in mesos.  If marathon has a ‘re-registered’ value that is 
always ‘just now’, you’re suffering from the issue described in 
https://github.com/mesosphere/marathon/issues/1140 (either due to ports being 
blocked or due to marathon registering with mesos as 127.0.0.1).

From: Emre Tekelioğlu [mailto:emre.tekelio...@gmail.com]
Sent: Monday, February 09, 2015 7:50 AM
To: user@mesos.apache.org
Subject: Mesos Marathon weird behavior

Hi,
I have a single node mesos cluster on Ubuntu 14.04 host machine, with one node 
mesos-master and one node mesos-slave.
I try to deploy and run my dockerized services with marathon but i have a weird 
problem.

For example If i start a new app with 5 instance, only one task start and the 
status stuck in "deploying". If i restart the mesos-master, then the second 
task start to run.

Each restart for mesos-master cause an increment of a task.

Any idea?

Regards,
Emre

Re: Mesos Marathon weird behavior

[Dario Rexin]

This looks normal. But could you try updating to Marathon 0.8.0? We fixed a 
couple of deployment related bugs in there.

> On 09 Feb 2015, at 16:38, Emre Tekelioğlu  wrote:
> 
> Hi Dario,
> 
> I found these logs in syslog.
> Is it normal?
> 
> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,537] INFO 
> Computed new deployment plan: DeploymentPlan(2015-02-09T15:26:38.437Z, 
> (Step(Vector(Start(App(/testtest1, None)), 0))), 
> Step(List(Scale(App(/testtest1, None)), 5) 
> (mesosphere.marathon.upgrade.DeploymentPlan$:246)
> Feb  9 15:26:38 maas01 marathon[8808]: [INFO] [02/09/2015 15:26:38.539] 
> [marathon-akka.actor.default-dispatcher-14] 
> [akka://marathon/user/MarathonScheduler/$a/UpgradeManager/5cf28be2-af8e-427d-90a2-5b0388b7009a/$a]
>  Successfully started 0 instances of /testtest1
> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO 
> Starting app /testtest1 (mesosphere.marathon.SchedulerActions:363)
> Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO Already 
> running 0 instances of /testtest1. Not scaling. 
> (mesosphere.marathon.SchedulerActions:512)
> 
> 
> 2015-02-09 16:10 GMT+02:00 Dario Rexin  >:
> Could you please post the relevant log output of Mesos and Marathon?
> 
>> On 09 Feb 2015, at 15:06, Emre Tekelioğlu > > wrote:
>> 
>> I added the app with REST API, but nothing changed.
>> Status stuck in "Deploying"
>> I also run these commands;
>> 
>> echo 'docker,mesos' > /etc/mesos-slave/containerizers
>> echo '5mins' > /etc/mesos-slave/executor_registration_timeout
>> 
>> App config is like that
>> 
>> {  
>>"id":"/test2",
>>"cmd":null,
>>"args":[  
>> 
>>],
>>"user":null,
>>"env":{  
>> 
>>},
>>"instances":5,
>>"cpus":0.1,
>>"mem":128.0,
>>"disk":512.0,
>>"executor":"",
>>"constraints":[  
>> 
>>],
>>"uris":[  
>> 
>>],
>>"storeUrls":[  
>> 
>>],
>>"ports":[  
>>   10001
>>],
>>"requirePorts":false,
>>"backoffSeconds":1,
>>"backoffFactor":1.15,
>>"maxLaunchDelaySeconds":3600,
>>"container":{  
>>   "type":"DOCKER",
>>   "volumes":[  
>> 
>>   ],
>>   "docker":{  
>>  "image":"fehmicansaglam/progressed.io ",
>>  "network":null,
>>  "portMappings":null,
>>  "privileged":false,
>>  "parameters":[  
>> 
>>  ]
>>   }
>>},
>>"healthChecks":[  
>> 
>>],
>>"dependencies":[  
>> 
>>],
>>"upgradeStrategy":{  
>>   "minimumHealthCapacity":1.0,
>>   "maximumOverCapacity":1.0
>>},
>>"labels":{  
>> 
>>},
>>"version":"2015-02-09T13:55:52.577Z"
>> }
>> 
>> 
>> Thanks
>> 
>> 
>> 2015-02-09 15:43 GMT+02:00 Dario Rexin > >:
>> Yes, it’s currently not possible to deploy containers with the web UI.
>> 
>>> On 09 Feb 2015, at 14:42, Emre Tekelioğlu >> > wrote:
>>> 
>>> Hi Dario,
>>> 
>>> I couldn't find the container section in my Marathon UI when i click New 
>>> App, should i use only REST API to do that?
>>> 
>>> Regards,
>>> Emre
>>> 
>>> 2015-02-09 15:09 GMT+02:00 Dario Rexin >> >:
>>> Hi Emre,
>>> 
>>> Mesos and Marathon have full support for running Docker containers, you can 
>>> just specify them in the app definition, no need to run `docker run …` as a 
>>> command. Try this app definition instead:
>>> 
>>> {  
>>>   "id":"/progressed-io",
>>>   "instances":5,
>>>   "cpus":0.1,
>>>   "mem":128.0,
>>>   "disk":512.0,
>>>   “container”: {
>>>  “docker”: {
>>> “image”: “fehmicansaglam/progressed.io "
>>>   }
>>>   }
>>> }
>>> 
>>> Please make sure to start your mesos-slave process with 
>>> “—containerizers=docker,mesos”
>>> 
>>> Hope this helps.
>>> 
>>> Cheers,
>>> Dario
>>> 
 On 09 Feb 2015, at 14:02, Emre Tekelioğlu >>> > wrote:
 
 Hi Dario
 
 Mesos version 0.21.1, Marathon version 0.7.6.
 App config;
 {  
"id":"/progressed-io",
"cmd":"sudo docker run fehmicansaglam/progressed.io 
 ",
"args":null,
"user":null,
"env":{  
 
},
"instances":5,
"cpus":0.1,
"mem":128.0,
"disk":512.0,
"executor":"",
"constraints":[  
 
],
"uris":[  
 
],
"storeUrls":[  
 
],
"ports":[  
 
],
"requirePorts":false,
"backoffSeconds":1,
"backoffFactor":1.15,
"container":null,
"healthChecks":[  
 
],
"dependencies":[  
 
],
"upgradeStrategy":{  
   "minimumHealthCapacity":1.0
}
 
 
 Thanks,
 Emre
 
 2015-02-09 14:52 GMT+02:00 Dario Rexin >>> >:

Re: Mesos Marathon weird behavior

[Emre Tekelioğlu]

Hi Dario,

I found these logs in syslog.
Is it normal?

Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,537] INFO
Computed new deployment plan: DeploymentPlan(2015-02-09T15:26:38.437Z,
(Step(Vector(Start(App(/testtest1, None)), 0))),
Step(List(Scale(App(/testtest1, None)), 5)
(mesosphere.marathon.upgrade.DeploymentPlan$:246)
Feb  9 15:26:38 maas01 marathon[8808]: [INFO] [02/09/2015 15:26:38.539]
[marathon-akka.actor.default-dispatcher-14]
[akka://marathon/user/MarathonScheduler/$a/UpgradeManager/5cf28be2-af8e-427d-90a2-5b0388b7009a/$a]
Successfully started 0 instances of /testtest1
Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO
Starting app /testtest1 (mesosphere.marathon.SchedulerActions:363)
Feb  9 15:26:38 maas01 marathon[8808]: [2015-02-09 15:26:38,557] INFO
Already running 0 instances of /testtest1. Not scaling.
(mesosphere.marathon.SchedulerActions:512)


2015-02-09 16:10 GMT+02:00 Dario Rexin :

> Could you please post the relevant log output of Mesos and Marathon?
>
> On 09 Feb 2015, at 15:06, Emre Tekelioğlu 
> wrote:
>
> I added the app with REST API, but nothing changed.
> Status stuck in "Deploying"
> I also run these commands;
>
> echo 'docker,mesos' > /etc/mesos-slave/containerizers
> echo '5mins' > /etc/mesos-slave/executor_registration_timeout
>
> App config is like that
>
> {
>"id":"/test2",
>"cmd":null,
>"args":[
>
>],
>"user":null,
>"env":{
>
>},
>"instances":5,
>"cpus":0.1,
>"mem":128.0,
>"disk":512.0,
>"executor":"",
>"constraints":[
>
>],
>"uris":[
>
>],
>"storeUrls":[
>
>],
>"ports":[
>   10001
>],
>"requirePorts":false,
>"backoffSeconds":1,
>"backoffFactor":1.15,
>"maxLaunchDelaySeconds":3600,
>"container":{
>   "type":"DOCKER",
>   "volumes":[
>
>   ],
>   "docker":{
>  "image":"fehmicansaglam/progressed.io",
>  "network":null,
>  "portMappings":null,
>  "privileged":false,
>  "parameters":[
>
>  ]
>   }
>},
>"healthChecks":[
>
>],
>"dependencies":[
>
>],
>"upgradeStrategy":{
>   "minimumHealthCapacity":1.0,
>   "maximumOverCapacity":1.0
>},
>"labels":{
>
>},
>"version":"2015-02-09T13:55:52.577Z"
> }
>
>
> Thanks
>
>
> 2015-02-09 15:43 GMT+02:00 Dario Rexin :
>
>> Yes, it’s currently not possible to deploy containers with the web UI.
>>
>> On 09 Feb 2015, at 14:42, Emre Tekelioğlu 
>> wrote:
>>
>> Hi Dario,
>>
>> I couldn't find the container section in my Marathon UI when i click New
>> App, should i use only REST API to do that?
>>
>> Regards,
>> Emre
>>
>> 2015-02-09 15:09 GMT+02:00 Dario Rexin :
>>
>>> Hi Emre,
>>>
>>> Mesos and Marathon have full support for running Docker containers, you
>>> can just specify them in the app definition, no need to run `docker run …`
>>> as a command. Try this app definition instead:
>>>
>>> {
>>>   "id":"/progressed-io",
>>>   "instances":5,
>>>   "cpus":0.1,
>>>   "mem":128.0,
>>>   "disk":512.0,
>>>   “container”: {
>>>  “docker”: {
>>> “image”: “fehmicansaglam/progressed.io"
>>>   }
>>>   }
>>> }
>>>
>>> Please make sure to start your mesos-slave process with
>>> “—containerizers=docker,mesos”
>>>
>>> Hope this helps.
>>>
>>> Cheers,
>>> Dario
>>>
>>> On 09 Feb 2015, at 14:02, Emre Tekelioğlu 
>>> wrote:
>>>
>>> Hi Dario
>>>
>>> Mesos version 0.21.1, Marathon version 0.7.6.
>>> App config;
>>> {
>>>"id":"/progressed-io",
>>>"cmd":"sudo docker run fehmicansaglam/progressed.io",
>>>"args":null,
>>>"user":null,
>>>"env":{
>>>
>>>},
>>>"instances":5,
>>>"cpus":0.1,
>>>"mem":128.0,
>>>"disk":512.0,
>>>"executor":"",
>>>"constraints":[
>>>
>>>],
>>>"uris":[
>>>
>>>],
>>>"storeUrls":[
>>>
>>>],
>>>"ports":[
>>>
>>>],
>>>"requirePorts":false,
>>>"backoffSeconds":1,
>>>"backoffFactor":1.15,
>>>"container":null,
>>>"healthChecks":[
>>>
>>>],
>>>"dependencies":[
>>>
>>>],
>>>"upgradeStrategy":{
>>>   "minimumHealthCapacity":1.0
>>>}
>>>
>>>
>>> Thanks,
>>> Emre
>>>
>>> 2015-02-09 14:52 GMT+02:00 Dario Rexin :
>>>
 Hi Emre,

 what are the versions of Mesos and Marathon and what does your app
 definition look like?

 Cheers,
 Dario

 > On 09 Feb 2015, at 13:50, Emre Tekelioğlu 
 wrote:
 >
 > Hi,
 > I have a single node mesos cluster on Ubuntu 14.04 host machine, with
 one node mesos-master and one node mesos-slave.
 > I try to deploy and run my dockerized services with marathon but i
 have a weird problem.
 >
 > For example If i start a new app with 5 instance, only one task start
 and the status stuck in "deploying". If i restart the mesos-master, then
 the second task start to run.
 >
 > Each restart for mesos-master cause an increment of a task.
 >
 > Any idea?
 >

Re: Mesos Marathon weird behavior

[Dario Rexin]

Could you please post the relevant log output of Mesos and Marathon?

> On 09 Feb 2015, at 15:06, Emre Tekelioğlu  wrote:
> 
> I added the app with REST API, but nothing changed.
> Status stuck in "Deploying"
> I also run these commands;
> 
> echo 'docker,mesos' > /etc/mesos-slave/containerizers
> echo '5mins' > /etc/mesos-slave/executor_registration_timeout
> 
> App config is like that
> 
> {  
>"id":"/test2",
>"cmd":null,
>"args":[  
> 
>],
>"user":null,
>"env":{  
> 
>},
>"instances":5,
>"cpus":0.1,
>"mem":128.0,
>"disk":512.0,
>"executor":"",
>"constraints":[  
> 
>],
>"uris":[  
> 
>],
>"storeUrls":[  
> 
>],
>"ports":[  
>   10001
>],
>"requirePorts":false,
>"backoffSeconds":1,
>"backoffFactor":1.15,
>"maxLaunchDelaySeconds":3600,
>"container":{  
>   "type":"DOCKER",
>   "volumes":[  
> 
>   ],
>   "docker":{  
>  "image":"fehmicansaglam/progressed.io ",
>  "network":null,
>  "portMappings":null,
>  "privileged":false,
>  "parameters":[  
> 
>  ]
>   }
>},
>"healthChecks":[  
> 
>],
>"dependencies":[  
> 
>],
>"upgradeStrategy":{  
>   "minimumHealthCapacity":1.0,
>   "maximumOverCapacity":1.0
>},
>"labels":{  
> 
>},
>"version":"2015-02-09T13:55:52.577Z"
> }
> 
> 
> Thanks
> 
> 
> 2015-02-09 15:43 GMT+02:00 Dario Rexin  >:
> Yes, it’s currently not possible to deploy containers with the web UI.
> 
>> On 09 Feb 2015, at 14:42, Emre Tekelioğlu > > wrote:
>> 
>> Hi Dario,
>> 
>> I couldn't find the container section in my Marathon UI when i click New 
>> App, should i use only REST API to do that?
>> 
>> Regards,
>> Emre
>> 
>> 2015-02-09 15:09 GMT+02:00 Dario Rexin > >:
>> Hi Emre,
>> 
>> Mesos and Marathon have full support for running Docker containers, you can 
>> just specify them in the app definition, no need to run `docker run …` as a 
>> command. Try this app definition instead:
>> 
>> {  
>>   "id":"/progressed-io",
>>   "instances":5,
>>   "cpus":0.1,
>>   "mem":128.0,
>>   "disk":512.0,
>>   “container”: {
>>  “docker”: {
>> “image”: “fehmicansaglam/progressed.io "
>>   }
>>   }
>> }
>> 
>> Please make sure to start your mesos-slave process with 
>> “—containerizers=docker,mesos”
>> 
>> Hope this helps.
>> 
>> Cheers,
>> Dario
>> 
>>> On 09 Feb 2015, at 14:02, Emre Tekelioğlu >> > wrote:
>>> 
>>> Hi Dario
>>> 
>>> Mesos version 0.21.1, Marathon version 0.7.6.
>>> App config;
>>> {  
>>>"id":"/progressed-io",
>>>"cmd":"sudo docker run fehmicansaglam/progressed.io 
>>> ",
>>>"args":null,
>>>"user":null,
>>>"env":{  
>>> 
>>>},
>>>"instances":5,
>>>"cpus":0.1,
>>>"mem":128.0,
>>>"disk":512.0,
>>>"executor":"",
>>>"constraints":[  
>>> 
>>>],
>>>"uris":[  
>>> 
>>>],
>>>"storeUrls":[  
>>> 
>>>],
>>>"ports":[  
>>> 
>>>],
>>>"requirePorts":false,
>>>"backoffSeconds":1,
>>>"backoffFactor":1.15,
>>>"container":null,
>>>"healthChecks":[  
>>> 
>>>],
>>>"dependencies":[  
>>> 
>>>],
>>>"upgradeStrategy":{  
>>>   "minimumHealthCapacity":1.0
>>>}
>>> 
>>> 
>>> Thanks,
>>> Emre
>>> 
>>> 2015-02-09 14:52 GMT+02:00 Dario Rexin >> >:
>>> Hi Emre,
>>> 
>>> what are the versions of Mesos and Marathon and what does your app 
>>> definition look like?
>>> 
>>> Cheers,
>>> Dario
>>> 
>>> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu >> > > wrote:
>>> >
>>> > Hi,
>>> > I have a single node mesos cluster on Ubuntu 14.04 host machine, with one 
>>> > node mesos-master and one node mesos-slave.
>>> > I try to deploy and run my dockerized services with marathon but i have a 
>>> > weird problem.
>>> >
>>> > For example If i start a new app with 5 instance, only one task start and 
>>> > the status stuck in "deploying". If i restart the mesos-master, then the 
>>> > second task start to run.
>>> >
>>> > Each restart for mesos-master cause an increment of a task.
>>> >
>>> > Any idea?
>>> >
>>> > Regards,
>>> > Emre
>>> 
>>> 
>> 
>> 
> 
> 

Re: Mesos Marathon weird behavior

[Emre Tekelioğlu]

I added the app with REST API, but nothing changed.
Status stuck in "Deploying"
I also run these commands;

echo 'docker,mesos' > /etc/mesos-slave/containerizers
echo '5mins' > /etc/mesos-slave/executor_registration_timeout

App config is like that

{
   "id":"/test2",
   "cmd":null,
   "args":[

   ],
   "user":null,
   "env":{

   },
   "instances":5,
   "cpus":0.1,
   "mem":128.0,
   "disk":512.0,
   "executor":"",
   "constraints":[

   ],
   "uris":[

   ],
   "storeUrls":[

   ],
   "ports":[
  10001
   ],
   "requirePorts":false,
   "backoffSeconds":1,
   "backoffFactor":1.15,
   "maxLaunchDelaySeconds":3600,
   "container":{
  "type":"DOCKER",
  "volumes":[

  ],
  "docker":{
 "image":"fehmicansaglam/progressed.io",
 "network":null,
 "portMappings":null,
 "privileged":false,
 "parameters":[

 ]
  }
   },
   "healthChecks":[

   ],
   "dependencies":[

   ],
   "upgradeStrategy":{
  "minimumHealthCapacity":1.0,
  "maximumOverCapacity":1.0
   },
   "labels":{

   },
   "version":"2015-02-09T13:55:52.577Z"
}


Thanks


2015-02-09 15:43 GMT+02:00 Dario Rexin :

> Yes, it’s currently not possible to deploy containers with the web UI.
>
> On 09 Feb 2015, at 14:42, Emre Tekelioğlu 
> wrote:
>
> Hi Dario,
>
> I couldn't find the container section in my Marathon UI when i click New
> App, should i use only REST API to do that?
>
> Regards,
> Emre
>
> 2015-02-09 15:09 GMT+02:00 Dario Rexin :
>
>> Hi Emre,
>>
>> Mesos and Marathon have full support for running Docker containers, you
>> can just specify them in the app definition, no need to run `docker run …`
>> as a command. Try this app definition instead:
>>
>> {
>>   "id":"/progressed-io",
>>   "instances":5,
>>   "cpus":0.1,
>>   "mem":128.0,
>>   "disk":512.0,
>>   “container”: {
>>  “docker”: {
>> “image”: “fehmicansaglam/progressed.io"
>>   }
>>   }
>> }
>>
>> Please make sure to start your mesos-slave process with
>> “—containerizers=docker,mesos”
>>
>> Hope this helps.
>>
>> Cheers,
>> Dario
>>
>> On 09 Feb 2015, at 14:02, Emre Tekelioğlu 
>> wrote:
>>
>> Hi Dario
>>
>> Mesos version 0.21.1, Marathon version 0.7.6.
>> App config;
>> {
>>"id":"/progressed-io",
>>"cmd":"sudo docker run fehmicansaglam/progressed.io",
>>"args":null,
>>"user":null,
>>"env":{
>>
>>},
>>"instances":5,
>>"cpus":0.1,
>>"mem":128.0,
>>"disk":512.0,
>>"executor":"",
>>"constraints":[
>>
>>],
>>"uris":[
>>
>>],
>>"storeUrls":[
>>
>>],
>>"ports":[
>>
>>],
>>"requirePorts":false,
>>"backoffSeconds":1,
>>"backoffFactor":1.15,
>>"container":null,
>>"healthChecks":[
>>
>>],
>>"dependencies":[
>>
>>],
>>"upgradeStrategy":{
>>   "minimumHealthCapacity":1.0
>>}
>>
>>
>> Thanks,
>> Emre
>>
>> 2015-02-09 14:52 GMT+02:00 Dario Rexin :
>>
>>> Hi Emre,
>>>
>>> what are the versions of Mesos and Marathon and what does your app
>>> definition look like?
>>>
>>> Cheers,
>>> Dario
>>>
>>> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu 
>>> wrote:
>>> >
>>> > Hi,
>>> > I have a single node mesos cluster on Ubuntu 14.04 host machine, with
>>> one node mesos-master and one node mesos-slave.
>>> > I try to deploy and run my dockerized services with marathon but i
>>> have a weird problem.
>>> >
>>> > For example If i start a new app with 5 instance, only one task start
>>> and the status stuck in "deploying". If i restart the mesos-master, then
>>> the second task start to run.
>>> >
>>> > Each restart for mesos-master cause an increment of a task.
>>> >
>>> > Any idea?
>>> >
>>> > Regards,
>>> > Emre
>>>
>>>
>>
>>
>
>

Re: Mesos Marathon weird behavior

[Dario Rexin]

Yes, it’s currently not possible to deploy containers with the web UI.

> On 09 Feb 2015, at 14:42, Emre Tekelioğlu  wrote:
> 
> Hi Dario,
> 
> I couldn't find the container section in my Marathon UI when i click New App, 
> should i use only REST API to do that?
> 
> Regards,
> Emre
> 
> 2015-02-09 15:09 GMT+02:00 Dario Rexin  >:
> Hi Emre,
> 
> Mesos and Marathon have full support for running Docker containers, you can 
> just specify them in the app definition, no need to run `docker run …` as a 
> command. Try this app definition instead:
> 
> {  
>   "id":"/progressed-io",
>   "instances":5,
>   "cpus":0.1,
>   "mem":128.0,
>   "disk":512.0,
>   “container”: {
>  “docker”: {
> “image”: “fehmicansaglam/progressed.io "
>   }
>   }
> }
> 
> Please make sure to start your mesos-slave process with 
> “—containerizers=docker,mesos”
> 
> Hope this helps.
> 
> Cheers,
> Dario
> 
>> On 09 Feb 2015, at 14:02, Emre Tekelioğlu > > wrote:
>> 
>> Hi Dario
>> 
>> Mesos version 0.21.1, Marathon version 0.7.6.
>> App config;
>> {  
>>"id":"/progressed-io",
>>"cmd":"sudo docker run fehmicansaglam/progressed.io 
>> ",
>>"args":null,
>>"user":null,
>>"env":{  
>> 
>>},
>>"instances":5,
>>"cpus":0.1,
>>"mem":128.0,
>>"disk":512.0,
>>"executor":"",
>>"constraints":[  
>> 
>>],
>>"uris":[  
>> 
>>],
>>"storeUrls":[  
>> 
>>],
>>"ports":[  
>> 
>>],
>>"requirePorts":false,
>>"backoffSeconds":1,
>>"backoffFactor":1.15,
>>"container":null,
>>"healthChecks":[  
>> 
>>],
>>"dependencies":[  
>> 
>>],
>>"upgradeStrategy":{  
>>   "minimumHealthCapacity":1.0
>>}
>> 
>> 
>> Thanks,
>> Emre
>> 
>> 2015-02-09 14:52 GMT+02:00 Dario Rexin > >:
>> Hi Emre,
>> 
>> what are the versions of Mesos and Marathon and what does your app 
>> definition look like?
>> 
>> Cheers,
>> Dario
>> 
>> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu > > > wrote:
>> >
>> > Hi,
>> > I have a single node mesos cluster on Ubuntu 14.04 host machine, with one 
>> > node mesos-master and one node mesos-slave.
>> > I try to deploy and run my dockerized services with marathon but i have a 
>> > weird problem.
>> >
>> > For example If i start a new app with 5 instance, only one task start and 
>> > the status stuck in "deploying". If i restart the mesos-master, then the 
>> > second task start to run.
>> >
>> > Each restart for mesos-master cause an increment of a task.
>> >
>> > Any idea?
>> >
>> > Regards,
>> > Emre
>> 
>> 
> 
> 

Re: Mesos Marathon weird behavior

[Emre Tekelioğlu]

Hi Dario,

I couldn't find the container section in my Marathon UI when i click New
App, should i use only REST API to do that?

Regards,
Emre

2015-02-09 15:09 GMT+02:00 Dario Rexin :

> Hi Emre,
>
> Mesos and Marathon have full support for running Docker containers, you
> can just specify them in the app definition, no need to run `docker run …`
> as a command. Try this app definition instead:
>
> {
>   "id":"/progressed-io",
>   "instances":5,
>   "cpus":0.1,
>   "mem":128.0,
>   "disk":512.0,
>   “container”: {
>  “docker”: {
> “image”: “fehmicansaglam/progressed.io"
>   }
>   }
> }
>
> Please make sure to start your mesos-slave process with
> “—containerizers=docker,mesos”
>
> Hope this helps.
>
> Cheers,
> Dario
>
> On 09 Feb 2015, at 14:02, Emre Tekelioğlu 
> wrote:
>
> Hi Dario
>
> Mesos version 0.21.1, Marathon version 0.7.6.
> App config;
> {
>"id":"/progressed-io",
>"cmd":"sudo docker run fehmicansaglam/progressed.io",
>"args":null,
>"user":null,
>"env":{
>
>},
>"instances":5,
>"cpus":0.1,
>"mem":128.0,
>"disk":512.0,
>"executor":"",
>"constraints":[
>
>],
>"uris":[
>
>],
>"storeUrls":[
>
>],
>"ports":[
>
>],
>"requirePorts":false,
>"backoffSeconds":1,
>"backoffFactor":1.15,
>"container":null,
>"healthChecks":[
>
>],
>"dependencies":[
>
>],
>"upgradeStrategy":{
>   "minimumHealthCapacity":1.0
>}
>
>
> Thanks,
> Emre
>
> 2015-02-09 14:52 GMT+02:00 Dario Rexin :
>
>> Hi Emre,
>>
>> what are the versions of Mesos and Marathon and what does your app
>> definition look like?
>>
>> Cheers,
>> Dario
>>
>> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu 
>> wrote:
>> >
>> > Hi,
>> > I have a single node mesos cluster on Ubuntu 14.04 host machine, with
>> one node mesos-master and one node mesos-slave.
>> > I try to deploy and run my dockerized services with marathon but i have
>> a weird problem.
>> >
>> > For example If i start a new app with 5 instance, only one task start
>> and the status stuck in "deploying". If i restart the mesos-master, then
>> the second task start to run.
>> >
>> > Each restart for mesos-master cause an increment of a task.
>> >
>> > Any idea?
>> >
>> > Regards,
>> > Emre
>>
>>
>
>

Re: Mesos Marathon weird behavior

[Dario Rexin]

Hi Emre,

Mesos and Marathon have full support for running Docker containers, you can 
just specify them in the app definition, no need to run `docker run …` as a 
command. Try this app definition instead:

{  
  "id":"/progressed-io",
  "instances":5,
  "cpus":0.1,
  "mem":128.0,
  "disk":512.0,
  “container”: {
 “docker”: {
“image”: “fehmicansaglam/progressed.io "
  }
  }
}

Please make sure to start your mesos-slave process with 
“—containerizers=docker,mesos”

Hope this helps.

Cheers,
Dario

> On 09 Feb 2015, at 14:02, Emre Tekelioğlu  wrote:
> 
> Hi Dario
> 
> Mesos version 0.21.1, Marathon version 0.7.6.
> App config;
> {  
>"id":"/progressed-io",
>"cmd":"sudo docker run fehmicansaglam/progressed.io 
> ",
>"args":null,
>"user":null,
>"env":{  
> 
>},
>"instances":5,
>"cpus":0.1,
>"mem":128.0,
>"disk":512.0,
>"executor":"",
>"constraints":[  
> 
>],
>"uris":[  
> 
>],
>"storeUrls":[  
> 
>],
>"ports":[  
> 
>],
>"requirePorts":false,
>"backoffSeconds":1,
>"backoffFactor":1.15,
>"container":null,
>"healthChecks":[  
> 
>],
>"dependencies":[  
> 
>],
>"upgradeStrategy":{  
>   "minimumHealthCapacity":1.0
>}
> 
> 
> Thanks,
> Emre
> 
> 2015-02-09 14:52 GMT+02:00 Dario Rexin  >:
> Hi Emre,
> 
> what are the versions of Mesos and Marathon and what does your app definition 
> look like?
> 
> Cheers,
> Dario
> 
> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu  > > wrote:
> >
> > Hi,
> > I have a single node mesos cluster on Ubuntu 14.04 host machine, with one 
> > node mesos-master and one node mesos-slave.
> > I try to deploy and run my dockerized services with marathon but i have a 
> > weird problem.
> >
> > For example If i start a new app with 5 instance, only one task start and 
> > the status stuck in "deploying". If i restart the mesos-master, then the 
> > second task start to run.
> >
> > Each restart for mesos-master cause an increment of a task.
> >
> > Any idea?
> >
> > Regards,
> > Emre
> 
> 

Re: Mesos Marathon weird behavior

[Emre Tekelioğlu]

Hi Dario

Mesos version 0.21.1, Marathon version 0.7.6.
App config;
{
   "id":"/progressed-io",
   "cmd":"sudo docker run fehmicansaglam/progressed.io",
   "args":null,
   "user":null,
   "env":{

   },
   "instances":5,
   "cpus":0.1,
   "mem":128.0,
   "disk":512.0,
   "executor":"",
   "constraints":[

   ],
   "uris":[

   ],
   "storeUrls":[

   ],
   "ports":[

   ],
   "requirePorts":false,
   "backoffSeconds":1,
   "backoffFactor":1.15,
   "container":null,
   "healthChecks":[

   ],
   "dependencies":[

   ],
   "upgradeStrategy":{
  "minimumHealthCapacity":1.0
   }


Thanks,
Emre

2015-02-09 14:52 GMT+02:00 Dario Rexin :

> Hi Emre,
>
> what are the versions of Mesos and Marathon and what does your app
> definition look like?
>
> Cheers,
> Dario
>
> > On 09 Feb 2015, at 13:50, Emre Tekelioğlu 
> wrote:
> >
> > Hi,
> > I have a single node mesos cluster on Ubuntu 14.04 host machine, with
> one node mesos-master and one node mesos-slave.
> > I try to deploy and run my dockerized services with marathon but i have
> a weird problem.
> >
> > For example If i start a new app with 5 instance, only one task start
> and the status stuck in "deploying". If i restart the mesos-master, then
> the second task start to run.
> >
> > Each restart for mesos-master cause an increment of a task.
> >
> > Any idea?
> >
> > Regards,
> > Emre
>
>

Re: Mesos Marathon weird behavior

[Dario Rexin]

Hi Emre,

what are the versions of Mesos and Marathon and what does your app definition 
look like?

Cheers,
Dario

> On 09 Feb 2015, at 13:50, Emre Tekelioğlu  wrote:
> 
> Hi,
> I have a single node mesos cluster on Ubuntu 14.04 host machine, with one 
> node mesos-master and one node mesos-slave.
> I try to deploy and run my dockerized services with marathon but i have a 
> weird problem.
> 
> For example If i start a new app with 5 instance, only one task start and the 
> status stuck in "deploying". If i restart the mesos-master, then the second 
> task start to run. 
> 
> Each restart for mesos-master cause an increment of a task.
> 
> Any idea?
> 
> Regards,
> Emre

Mesos Marathon weird behavior

[Emre Tekelioğlu]

Hi,
I have a single node mesos cluster on Ubuntu 14.04 host machine, with one
node mesos-master and one node mesos-slave.
I try to deploy and run my dockerized services with marathon but i have a
weird problem.

For example If i start a new app with 5 instance, only one task start and
the status stuck in "deploying". If i restart the mesos-master, then the
second task start to run.

Each restart for mesos-master cause an increment of a task.

Any idea?

Regards,
Emre

Total resources available versus actually available

[craig w]

Mesos slaves report the total resources they have available, such as 2 CPU,
8GB Memory and 100GB disk.

Does Mesos keep track of the amount of resources that are actually
available to best schedule tasks?

For example, imagine a slave has some other processes running on it (that
are not mesos tasks) that are taking up 6GB of RAM (out of the total 8GB).
A new task is created in Mesos that needs 4GB of RAM, would Mesos still try
to put the new task on the slave?