Re: Optimal resource allocation

Dario, Thanks for the reply and clarification. How hard is to port to ARM? is there lot of architecture related code? Any idea? Regards, Pradeep On 5 February 2015 at 12:01, Dario Rexin da...@mesosphere.io mailto:da...@mesosphere.io wrote: There is currently no support for ARM

Re: Marathon stability and use-case

Thanks Niklas. Hi Antonin, Marathon should be able to handle tjousands of tasks and that is exactly what it's made for. Unfortunately the latest release (0.7.6) has been very unstable. We fixed a lot of bugs that caused this unstability and just tagged an RC for 0.8.0 yesterday:

Re: Mesos Marathon weird behavior

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 da

Re: Mesos Marathon weird behavior

It should be in the Mesosphere repos since Thursday last week. On 09 Feb 2015, at 17:02, Emre Tekelioğlu emre.tekelio...@gmail.com wrote: I installed mesos from repo, did you know when ubuntu repo will update? 2015-02-09 17:46 GMT+02:00 Dario Rexin da...@mesosphere.io mailto:da

Re: Mesos Marathon weird behavior

a service restart. Now marathon shows version 0.8.0 but nothing changed :| 2015-02-09 18:05 GMT+02:00 Dario Rexin da...@mesosphere.io: It should be in the Mesosphere repos since Thursday last week. On 09 Feb 2015, at 17:02, Emre Tekelioğlu emre.tekelio...@gmail.com wrote: I installed mesos

Re: Mesos Marathon weird behavior

:38,557] INFO Already running 0 instances of /testtest1. Not scaling. Marathon says 'Not scaling' here? But has 0 instances running, is this correct? -- Thanks, Chengwei (mesosphere.marathon.SchedulerActions:512) 2015-02-09 16:10 GMT+02:00 Dario Rexin da...@mesosphere.io: Could

Re: Mesos Marathon weird behavior

: Hi Dario, I gave up for a while. I'm going to try https://github.com/mesosphere/playa-mesos https://github.com/mesosphere/playa-mesos Regards, Emre 2015-02-10 11:41 GMT+02:00 Dario Rexin da...@mesosphere.io mailto:da...@mesosphere.io: Any failed tasks in the list? Does

Re: Mesos Marathon weird behavior

, maximumOverCapacity:1.0 }, labels:{ }, version:2015-02-09T13:55:52.577Z } Thanks 2015-02-09 15:43 GMT+02:00 Dario Rexin da...@mesosphere.io mailto:da...@mesosphere.io: Yes, it’s currently not possible to deploy containers with the web UI. On 09 Feb 2015

Re: Mesos Marathon weird behavior

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 emre.tekelio...@gmail.com wrote: Hi, I have a single node mesos cluster on Ubuntu 14.04 host machine, with one node mesos-master and

Re: Mesos Marathon weird behavior

, Emre 2015-02-09 15:09 GMT+02:00 Dario Rexin da...@mesosphere.io mailto:da...@mesosphere.io: 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

Re: Mesos Marathon weird behavior

:{ minimumHealthCapacity:1.0 } Thanks, Emre 2015-02-09 14:52 GMT+02:00 Dario Rexin da...@mesosphere.io mailto:da...@mesosphere.io: 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

Re: Weird behavior when stopping the mesos master leader of a HA mesos cluster

Hi Geoffrey, we identified the issue and will fix it in Marathon 0.8.2. To prevent this behaviour for now, you just have to make sure that in a fresh setup (Marathon was never connected to Mesos) you first start up a single Marathon and let it register with Mesos and then start the other

Re: Deploying containers to every mesos slave node

Hi all, the question for a way to start a task on all nodes has come up a couple of times on the Marathon ML and GH repo as well. The reason for not having this in Marathon is simple. We can’t guarantee that we’ll ever get offers for all the slaves and also we son’t know how many slaves are in

Re: Does MesosScheduler.resourceOffers need to be reentrant?

No, resourceOffers will only be called once at a time. Cheers, Dario On 26 Feb 2015, at 10:09, Itamar Ostricher ita...@yowza3d.com wrote: Making sure the question is clear: I'm implementing a framework scheduler, and I want to know if the resourceOffers method can be invoked while a

Re: [VOTE] Release Apache Mesos 0.22.0 (rc4)

+1 (non-binding) tested with Marathon On 24 Mar 2015, at 00:03, Adam Bordelon a...@mesosphere.io wrote: +1 (binding) make check passes on Ubuntu 14.04 and Mac OS X (buildbot) On Mon, Mar 23, 2015 at 3:44 PM, Elizabeth Lingg elizab...@mesosphere.io mailto:elizab...@mesosphere.io wrote:

Re: Does mesos scheduler will listen on random port to receive message from mesos master

This a the Mesos library. You can set a fixed port through the env variable $LIBPROCESS_PORT. Cheers, Dario On 20.04.2015, at 09:20, Mingqi Shao mingqi.s...@gmail.com wrote: I deploy marathon and mesos cluster. I found marathon will listen on a random port to receive message from mesos

Re: Mesosphere security

Thanks. On 24 Apr 2015, at 18:10, Nikolay Borodachev nbo...@adobe.com wrote: Here you go: https://github.com/mesosphere/marathon/issues/1428 https://github.com/mesosphere/marathon/issues/1428 From: Dario Rexin [mailto:da...@mesosphere.io] Sent: Friday, April 24, 2015 10:56 AM

Re: Mesosphere security

user id password from a file? From: Dario Rexin [mailto:da...@mesosphere.io] Sent: Friday, April 24, 2015 10:37 AM To: user@mesos.apache.org Subject: Re: Mesosphere security Hi Nikolay, we are close to releasing Marathon 0.8.2 which will support disabling HTTP with the flag

Re: Marathon chage of leader and stalled deployments

? To test it out, should I set LIBPROCESS_IP to an IP address of mesos1 server? Thank you Nikolay From: Dario Rexin [mailto:da...@mesosphere.io] Sent: Tuesday, April 28, 2015 4:31 AM To: user@mesos.apache.org Subject: Re: Marathon chage of leader and stalled deployments Hi Nikolay

Re: Marathon chage of leader and stalled deployments

, and marathon? Thanks Nikolay From: Dario Rexin [mailto:da...@mesosphere.io] Sent: Tuesday, April 28, 2015 9:47 AM To: user@mesos.apache.org Subject: Re: Marathon chage of leader and stalled deployments On each host you have to set it to the interface that is connected to the network your

Re: Marathon chage of leader and stalled deployments

Hi Nikolay, this is an unexpected behavior. Could you please post the log output from the leading node around the time you try to scale? Also, what version of Marathon are you running? Thanks, Dario On 27.04.2015, at 20:41, Nikolay Borodachev nbo...@adobe.com wrote: Hello All, I

Re: Problem using cgroups/mem isolator

Hi Arunabha, you may need to enable memory isolation in the kernel. You can do that by adding cgroup_enable=memory to your kernel boot options. Cheers, Dario On 02.05.2015, at 04:49, Arunabha Ghosh arunabha...@gmail.com wrote: Hi, I'm trying to use the cgroups memory isolator, but

[VOTE] Release Apache Mesos 0.24.0 (rc1)

-1 (non-binding) I found a breaking bug in the new HTTP API. The messages do not conform to the HTTP standard for chunked transfer encoding. in RFC 2616 Sec. 3 (http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html) a chunk is defined as:

Re: [VOTE] Release Apache Mesos 0.24.0 (rc1)

event-data Looking forward to more bug reports as you try out the new API ! -anand On Aug 28, 2015, at 12:56 AM, Dario Rexin dario.re...@me.com wrote: -1 (non-binding) I found a breaking bug in the new HTTP API. The messages do not conform to the HTTP standard for chunked transfer

Re: [VOTE] Release Apache Mesos 0.24.0 (rc1)

ks, is it safe to assume that each chunk contains exactly one >> event? Because that would make parsing the events a lot easier for me. > > No guarantee that a chunk is a full single event. > >> On Tue, Sep 1, 2015 at 1:49 AM, Dario Rexin <dario.re...@me.com> wrote: >&g

Re: [VOTE] Release Apache Mesos 0.24.0 (rc1)

e formats and Record-IO > format was the one we settled on. Note that this format is also used by the > Twitter streaming API (see delimited messages section). > > HTH, > >> On Mon, Aug 31, 2015 at 2:09 PM, Dario Rexin <dario.re...@me.com> wrote: >> Hi Vino, >>

Re: [VOTE] Release Apache Mesos 0.24.0 (rc1)

y the Twitter streaming API >> <https://dev.twitter.com/streaming/overview/processing> (see delimited >> messages section). >> >> HTH, >> >> On Mon, Aug 31, 2015 at 2:09 PM, Dario Rexin <dario.re...@me.com >> <mailto:dario.re...@me.com>> wrote

Re: [VOTE] Release Apache Mesos 0.24.0 (rc1)

that you observe on stdout with CURL is of the Record-IO encoding used for the events stream ( and is not related to the RFC ): event = event-size LF event-data Looking forward to more bug reports as you try out the new API ! -anand On Aug 28, 2015, at 12:56 AM, Dario Rexin

Re: [VOTE] Release Apache Mesos 0.24.0 (rc1)

to more bug reports as you try out the new API ! -anand On Aug 28, 2015, at 12:56 AM, Dario Rexin dario.re...@me.com wrote: -1 (non-binding) I found a breaking bug in the new HTTP API. The messages do not conform to the HTTP standard for chunked transfer encoding. in RFC 2616 Sec. 3

HTTP API

Hi all, since the introduction of the HTTP API in 0.24 around 7.5 months have passed. What are the plans to make this API stable? There are already features (inverse offers) that are exclusively available through this API, so it would be great to have a timeline, as I think for most people

Re: Running Mesos agent on ARM (Raspberry Pi)?

Hi Sharma, I played around with Mesos on RPi a while back and have been able to compile and run it with 2 little patches. 1) Depending on the ZK version, it may be necessary to patch a function that uses inline ASM to use the resp. compiler intrinsics (I don’t remember where exactly in zk it

Re: Debugging Scheduler HTTP API Failures

HinZameer, when you send the SUBSCRIBE to Mesos, the response will contain a header 'Mesos-Stream-Id'. You have to send that header with every subsequent call you send to Mesos for that framework. -- Dario > On Aug 14, 2016, at 5:58 PM, Zameer Manji wrote: > > Hey, > >

Re: Debugging Scheduler HTTP API Failures

Oh, sorry, I didn't see you actually set the header (wall of text ;) ). That's an interesting issue, do you set the header case sensitive? I know headers shouldn't be case sensitive, but maybe there's a bug in the Mesos code. I have not seen this issue before. > On Aug 14, 2016, at 5:58 PM,

Re: What's the official pronounce of mesos?

Mesos comes from the Greek word ‘mesos’, which means ‘middle’, so I guess that would be a valid pronunciation ;). > On Jul 13, 2016, at 5:18 PM, Jie Yu wrote: > > Looks like we don't have an official pronunciation. The previous guideline > was "say it how you like it". >

Re: kill mesos task

Hey, Mesos does not relaunch tasks on its own. What Scheduler do you use? Did you delete the service from that scheduler. -- Dario > On Aug 15, 2016, at 10:35 PM, Luke Adolph wrote: > > > Hi all: > I find a zombie task on mesos. I ssh to the agent and use "sudo kill

Re: what is the status on this?

gt; wrote: >>> If we use the replicated log for leader election, how will frameworks >>> detect the leading master? Right now the scheduler driver uses the >>> MasterInfo in ZK to discover the leader and detect leadership changes. >>> >>>> On Mon,

Re: Performance regression in v1 api vs v0

erReconcileTasks_BENCHMARK_Test.SchedulerDriver/3 > (2711 ms) > ``` > > *SchedulerLibrary* is the HTTP API, *SchedulerDriver* is the old way based on > libmesos.so. > > On Sun, Oct 16, 2016 at 2:41 PM, Dario Rexin <dre...@apple.com > <mailto:dre...@apple.com>> wrote: >

Performance regression in v1 api vs v0

Hi all, I recently did some performance testing on the v1 scheduler API and found that throughput is around 10x lower than for the v0 API. Using 1 connection, I don’t get a lot more than 1,500 calls per second, where the v0 API can do ~15,000. If I use multiple connections, throughput maxes

Re: Performance regression in v1 api vs v0

that measures just the time > taken by the master to write the events. I would file an issue shortly to > address this. > > Do you mind filing an issue with more details on your test setup? > > -anand > > On Sun, Oct 16, 2016 at 12:05 AM, Dario Rexin <dre...@apple.co

Re: Performance regression in v1 api vs v0

rent call? That might explain the behavior > you are seeing. > > -anand > > On Sun, Oct 16, 2016 at 11:58 AM, tommy xiao <xia...@gmail.com > <mailto:xia...@gmail.com>> wrote: > interesting this topic. > > 2016-10-17 2:51 GMT+08:00 Dario Rexin <dre...

Re: Welcome Guangya Liu as Mesos Committer and PMC member!

Congrats! > On Dec 16, 2016, at 4:27 PM, Vinod Kone wrote: > > Congrats Guangya! Welcome to the PMC! > >> On Fri, Dec 16, 2016 at 7:03 PM, Sam wrote: >> congratulations Guangya >> >> Sent from my iPhone >> >>> On 17 Dec 2016, at 3:23 AM, Avinash

Re: task has finished ,but keep resending TASK_RUNNING message

Hey, Mesos keeps resending status updates until they are acknowledged. Just call `driver.acknowledge(status)` and it should stop sending them. -- Dario > On Apr 29, 2017, at 9:07 AM, 梦开始的地方 <382607...@qq.com> wrote: > > Hi,I write a simple scheduler,task finished but keep resending

Re: [VOTE] Release Apache Mesos 1.3.0 (rc2)

-1 (non-binding) We found major performance regressions in the allocator / sorter that make this release pretty much unusable for us. JIRA: https://issues.apache.org/jira/browse/MESOS-7521 --  Dario > On May 17, 2017, at 4:44 AM, Michael