You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@mesos.apache.org by Rafael Capucho <ra...@gmail.com> on 2015/10/12 02:42:35 UTC
(Unknown)
Hello!,
I'm using the follow marathon script [1] to launch cassandra non-seeds
nodes, it is working properly.
[1] - http://hastebin.com/visujikela.lua
As you can see in the script, I'm limiting CPU and Memory.
But some nodes of my cluster is not that big, even because the cluster
isn't big yet. As we know, cassandra generally uses a lot of memory as
cache etc..
1) I would like to know if Mesos will kill (and keep killing) Cassandra
process if it reach the memory limit? if Yes, how can I block Mesos from
kill it?
2) If I have one server with 4gb memory where I deployed Mesos Slave, and I
create a container (by using marathon) with mem=1024 (for example) the
processing within that container, when they ask about the memory available,
they will receive 4gb or 1gb?
Thank you!
--
Rafael Capucho
Bachelor of Computer Science
Federal University of São Paulo
Institute of Science and Technology - ICT
PGP-Public Key: 2048R/7389A96F pgp.mit.edu
FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F
Re:
Posted by Brenden Matthews <br...@diddyinc.com>.
I can assure you it's much closer to being "production ready" than running
C* on Marathon.
On Mon, Oct 12, 2015 at 8:48 AM, Rafael Capucho <ra...@gmail.com>
wrote:
> Hello @Brender and @Rad,
>
> @Brender, we know about mesos-cassandra, but it says that it isnt
> production ready:
>
> "*DISCLAIMER* *This is a very early version of Cassandra-Mesos framework.
> This document, code behavior, and anything else may change without notice
> and/or break older installations."*
>
> And we will use it in production, then we think that its better to hand
> craft the configuration (because we can config like we need) instead of use
> something that could change soon and we don't have much control.
>
> @Rad, Thank you, It will help a lot.
>
>
>
>
>
> On Mon, Oct 12, 2015 at 11:45 AM, Brenden Matthews <br...@diddyinc.com>
> wrote:
>
>> It's worth noting that there's a purpose-built framework for Cassandra:
>> https://github.com/mesosphere/cassandra-mesos
>>
>> You probably want to use this instead of trying to run C* on Marathon.
>>
>> On Sun, Oct 11, 2015 at 6:05 PM, Rad Gruchalski <ra...@gruchalski.com>
>> wrote:
>>
>>> Rafael,
>>>
>>> According to the cassandra documentation, you should not be affected at
>>> all:
>>>
>>> http://docs.datastax.com/en/cassandra/2.0/cassandra/operations/ops_tune_jvm_c.html
>>>
>>> However, your performance with these settings will be rather poor.
>>>
>>> Kind regards,
>>> Radek Gruchalski
>>> radek@gruchalski.com <ra...@gruchalski.com>
>>> de.linkedin.com/in/radgruchalski/
>>>
>>>
>>> *Confidentiality:*This communication is intended for the above-named
>>> person and may be confidential and/or legally privileged.
>>> If it has come to you in error you must take no action based on it, nor
>>> must you copy or show it to anyone; please delete/destroy and inform the
>>> sender immediately.
>>>
>>> On Monday, 12 October 2015 at 02:42, Rafael Capucho wrote:
>>>
>>> Hello!,
>>>
>>> I'm using the follow marathon script [1] to launch cassandra non-seeds
>>> nodes, it is working properly.
>>>
>>> [1] - http://hastebin.com/visujikela.lua
>>>
>>> As you can see in the script, I'm limiting CPU and Memory.
>>>
>>> But some nodes of my cluster is not that big, even because the cluster
>>> isn't big yet. As we know, cassandra generally uses a lot of memory as
>>> cache etc..
>>>
>>> 1) I would like to know if Mesos will kill (and keep killing) Cassandra
>>> process if it reach the memory limit? if Yes, how can I block Mesos from
>>> kill it?
>>>
>>> 2) If I have one server with 4gb memory where I deployed Mesos Slave,
>>> and I create a container (by using marathon) with mem=1024 (for example)
>>> the processing within that container, when they ask about the memory
>>> available, they will receive 4gb or 1gb?
>>>
>>> Thank you!
>>>
>>>
>>> --
>>> Rafael Capucho
>>>
>>> Bachelor of Computer Science
>>> Federal University of São Paulo
>>> Institute of Science and Technology - ICT
>>>
>>> PGP-Public Key: 2048R/7389A96F pgp.mit.edu
>>> FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F
>>>
>>>
>>>
>>
>
>
> --
> --
> Rafael Capucho
>
> Bachelor of Computer Science
> Federal University of São Paulo
> Institute of Science and Technology - ICT
>
> PGP-Public Key: 2048R/7389A96F pgp.mit.edu
> FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F
>
Re:
Posted by Rafael Capucho <ra...@gmail.com>.
Hello @Brender and @Rad,
@Brender, we know about mesos-cassandra, but it says that it isnt
production ready:
"*DISCLAIMER* *This is a very early version of Cassandra-Mesos framework.
This document, code behavior, and anything else may change without notice
and/or break older installations."*
And we will use it in production, then we think that its better to hand
craft the configuration (because we can config like we need) instead of use
something that could change soon and we don't have much control.
@Rad, Thank you, It will help a lot.
On Mon, Oct 12, 2015 at 11:45 AM, Brenden Matthews <br...@diddyinc.com>
wrote:
> It's worth noting that there's a purpose-built framework for Cassandra:
> https://github.com/mesosphere/cassandra-mesos
>
> You probably want to use this instead of trying to run C* on Marathon.
>
> On Sun, Oct 11, 2015 at 6:05 PM, Rad Gruchalski <ra...@gruchalski.com>
> wrote:
>
>> Rafael,
>>
>> According to the cassandra documentation, you should not be affected at
>> all:
>>
>> http://docs.datastax.com/en/cassandra/2.0/cassandra/operations/ops_tune_jvm_c.html
>>
>> However, your performance with these settings will be rather poor.
>>
>> Kind regards,
>> Radek Gruchalski
>> radek@gruchalski.com <ra...@gruchalski.com>
>> de.linkedin.com/in/radgruchalski/
>>
>>
>> *Confidentiality:*This communication is intended for the above-named
>> person and may be confidential and/or legally privileged.
>> If it has come to you in error you must take no action based on it, nor
>> must you copy or show it to anyone; please delete/destroy and inform the
>> sender immediately.
>>
>> On Monday, 12 October 2015 at 02:42, Rafael Capucho wrote:
>>
>> Hello!,
>>
>> I'm using the follow marathon script [1] to launch cassandra non-seeds
>> nodes, it is working properly.
>>
>> [1] - http://hastebin.com/visujikela.lua
>>
>> As you can see in the script, I'm limiting CPU and Memory.
>>
>> But some nodes of my cluster is not that big, even because the cluster
>> isn't big yet. As we know, cassandra generally uses a lot of memory as
>> cache etc..
>>
>> 1) I would like to know if Mesos will kill (and keep killing) Cassandra
>> process if it reach the memory limit? if Yes, how can I block Mesos from
>> kill it?
>>
>> 2) If I have one server with 4gb memory where I deployed Mesos Slave, and
>> I create a container (by using marathon) with mem=1024 (for example) the
>> processing within that container, when they ask about the memory available,
>> they will receive 4gb or 1gb?
>>
>> Thank you!
>>
>>
>> --
>> Rafael Capucho
>>
>> Bachelor of Computer Science
>> Federal University of São Paulo
>> Institute of Science and Technology - ICT
>>
>> PGP-Public Key: 2048R/7389A96F pgp.mit.edu
>> FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F
>>
>>
>>
>
--
--
Rafael Capucho
Bachelor of Computer Science
Federal University of São Paulo
Institute of Science and Technology - ICT
PGP-Public Key: 2048R/7389A96F pgp.mit.edu
FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F
Re:
Posted by Brenden Matthews <br...@diddyinc.com>.
It's worth noting that there's a purpose-built framework for Cassandra:
https://github.com/mesosphere/cassandra-mesos
You probably want to use this instead of trying to run C* on Marathon.
On Sun, Oct 11, 2015 at 6:05 PM, Rad Gruchalski <ra...@gruchalski.com>
wrote:
> Rafael,
>
> According to the cassandra documentation, you should not be affected at
> all:
>
> http://docs.datastax.com/en/cassandra/2.0/cassandra/operations/ops_tune_jvm_c.html
>
> However, your performance with these settings will be rather poor.
>
> Kind regards,
> Radek Gruchalski
> radek@gruchalski.com <ra...@gruchalski.com>
> de.linkedin.com/in/radgruchalski/
>
>
> *Confidentiality:*This communication is intended for the above-named
> person and may be confidential and/or legally privileged.
> If it has come to you in error you must take no action based on it, nor
> must you copy or show it to anyone; please delete/destroy and inform the
> sender immediately.
>
> On Monday, 12 October 2015 at 02:42, Rafael Capucho wrote:
>
> Hello!,
>
> I'm using the follow marathon script [1] to launch cassandra non-seeds
> nodes, it is working properly.
>
> [1] - http://hastebin.com/visujikela.lua
>
> As you can see in the script, I'm limiting CPU and Memory.
>
> But some nodes of my cluster is not that big, even because the cluster
> isn't big yet. As we know, cassandra generally uses a lot of memory as
> cache etc..
>
> 1) I would like to know if Mesos will kill (and keep killing) Cassandra
> process if it reach the memory limit? if Yes, how can I block Mesos from
> kill it?
>
> 2) If I have one server with 4gb memory where I deployed Mesos Slave, and
> I create a container (by using marathon) with mem=1024 (for example) the
> processing within that container, when they ask about the memory available,
> they will receive 4gb or 1gb?
>
> Thank you!
>
>
> --
> Rafael Capucho
>
> Bachelor of Computer Science
> Federal University of São Paulo
> Institute of Science and Technology - ICT
>
> PGP-Public Key: 2048R/7389A96F pgp.mit.edu
> FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F
>
>
>
Re:
Posted by Rad Gruchalski <ra...@gruchalski.com>.
Rafael,
According to the cassandra documentation, you should not be affected at all:
http://docs.datastax.com/en/cassandra/2.0/cassandra/operations/ops_tune_jvm_c.html
However, your performance with these settings will be rather poor.
Kind regards,
Radek Gruchalski
radek@gruchalski.com (mailto:radek@gruchalski.com)
(mailto:radek@gruchalski.com)
de.linkedin.com/in/radgruchalski/ (http://de.linkedin.com/in/radgruchalski/)
Confidentiality:
This communication is intended for the above-named person and may be confidential and/or legally privileged.
If it has come to you in error you must take no action based on it, nor must you copy or show it to anyone; please delete/destroy and inform the sender immediately.
On Monday, 12 October 2015 at 02:42, Rafael Capucho wrote:
> Hello!,
>
> I'm using the follow marathon script [1] to launch cassandra non-seeds nodes, it is working properly.
>
> [1] - http://hastebin.com/visujikela.lua
>
> As you can see in the script, I'm limiting CPU and Memory.
>
> But some nodes of my cluster is not that big, even because the cluster isn't big yet. As we know, cassandra generally uses a lot of memory as cache etc..
>
> 1) I would like to know if Mesos will kill (and keep killing) Cassandra process if it reach the memory limit? if Yes, how can I block Mesos from kill it?
>
> 2) If I have one server with 4gb memory where I deployed Mesos Slave, and I create a container (by using marathon) with mem=1024 (for example) the processing within that container, when they ask about the memory available, they will receive 4gb or 1gb?
>
> Thank you!
>
>
> --
> Rafael Capucho
>
> Bachelor of Computer Science
> Federal University of São Paulo
> Institute of Science and Technology - ICT
>
> PGP-Public Key: 2048R/7389A96F pgp.mit.edu (http://pgp.mit.edu/)
> FP: EDB5 CDEE 8442 99CC C92D 9173 6B32 A5C9 7389 A96F