Control shard placement

It's possible to control shard placement on nodes. But how we can control
primary shard and replica shard placement independently ?

Thanks.

--

Hi,

What is the reason that you control the primary and replica shard
placement independently?
By default the primary shards's replicas aren't placed on the same node.

Martijn

On 16 November 2012 15:41, elasticuser merik2004-elastic@yahoo.fr wrote:

It's possible to control shard placement on nodes. But how we can control
primary shard and replica shard placement independently ?

Thanks.

--

--
Met vriendelijke groet,

Martijn van Groningen

--

Hi Martijn,

Because I have 10 servers with each about 500Go of hard drive space. So,
for my data, hard drive space is limited.
I would like to control where the primary shard and the replica shard are
stored in order to put my primary shard on the hdd and the replica shard on
a SAN.

On Friday, November 16, 2012 5:44:05 PM UTC+1, Martijn v Groningen wrote:

Hi,

What is the reason that you control the primary and replica shard
placement independently?
By default the primary shards's replicas aren't placed on the same node.

Martijn

On 16 November 2012 15:41, elasticuser <merik200...@yahoo.fr <javascript:>>
wrote:

It's possible to control shard placement on nodes. But how we can
control
primary shard and replica shard placement independently ?

Thanks.

--

--
Met vriendelijke groet,

Martijn van Groningen

--

So, the problem here is that if you will even manage to place all primary
shards to the right places, the moment one of the nodes with primary shards
goes away all corresponding replicas will be promoted to primaries.

Could you talk a little bit about how you want to take advantage of having
primaries on HDD and replicas on SAN. Is the idea to use nodes with HDD for
requests and keep nodes that are using SAN as hot standby? What are you
trying to achieve by such allocation?

On Friday, November 16, 2012 12:08:48 PM UTC-5, elasticuser wrote:

Hi Martijn,

Because I have 10 servers with each about 500Go of hard drive space. So,
for my data, hard drive space is limited.
I would like to control where the primary shard and the replica shard are
stored in order to put my primary shard on the hdd and the replica shard on
a SAN.

On Friday, November 16, 2012 5:44:05 PM UTC+1, Martijn v Groningen wrote:

Hi,

What is the reason that you control the primary and replica shard
placement independently?
By default the primary shards's replicas aren't placed on the same node.

Martijn

On 16 November 2012 15:41, elasticuser merik200...@yahoo.fr wrote:

It's possible to control shard placement on nodes. But how we can
control
primary shard and replica shard placement independently ?

Thanks.

--

--
Met vriendelijke groet,

Martijn van Groningen

--

My goal is to save space on the HDD. In my case, I have 5 To on my cluster
but with replica shards just 2.5 To. So, I would like to keep 5 To for my
primary shards and store the replica shards on a SAN.
I am aware that if a node with primary shards down all corresponding
replicas will be promoted to primaries but just during the time to repair
the node.
Indeed, I would like to use nodes with HDD for requests and nodes with SAN
just in case the nodes with primary shard has a problem.

On Saturday, November 17, 2012 12:11:15 AM UTC+1, Igor Motov wrote:

So, the problem here is that if you will even manage to place all primary
shards to the right places, the moment one of the nodes with primary shards
goes away all corresponding replicas will be promoted to primaries.

Could you talk a little bit about how you want to take advantage of having
primaries on HDD and replicas on SAN. Is the idea to use nodes with HDD for
requests and keep nodes that are using SAN as hot standby? What are you
trying to achieve by such allocation?

On Friday, November 16, 2012 12:08:48 PM UTC-5, elasticuser wrote:

Hi Martijn,

Because I have 10 servers with each about 500Go of hard drive space. So,
for my data, hard drive space is limited.
I would like to control where the primary shard and the replica shard are
stored in order to put my primary shard on the hdd and the replica shard on
a SAN.

On Friday, November 16, 2012 5:44:05 PM UTC+1, Martijn v Groningen wrote:

Hi,

What is the reason that you control the primary and replica shard
placement independently?
By default the primary shards's replicas aren't placed on the same node.

Martijn

On 16 November 2012 15:41, elasticuser merik200...@yahoo.fr wrote:

It's possible to control shard placement on nodes. But how we can
control
primary shard and replica shard placement independently ?

Thanks.

--

--
Met vriendelijke groet,

Martijn van Groningen

--

I see. That's an interesting idea. Unfortunately, I cannot think of a
mechanism that would allow you to do something like this. You can configure
a set of nodes to use HDD, and you can configure another set of nodes to
use SAN. You can use Allocation Awarenesshttp://www.elasticsearch.org/guide/reference/modules/cluster.htmlto make sure that if one shard is allocated on HDD node, its replica would
be allocated on SAN and vis versa. You can start HDD nodes first to make
sure they get all primaries shards. But that's it. There is really no way
to reassign primaries back to HDD nodes if they will get moved to SAN
nodes, or to limit searches only to HDD nodes when primaries are no longer
there.

On Saturday, November 17, 2012 4:12:31 AM UTC-5, elasticuser wrote:

My goal is to save space on the HDD. In my case, I have 5 To on my cluster
but with replica shards just 2.5 To. So, I would like to keep 5 To for my
primary shards and store the replica shards on a SAN.
I am aware that if a node with primary shards down all corresponding
replicas will be promoted to primaries but just during the time to repair
the node.
Indeed, I would like to use nodes with HDD for requests and nodes with SAN
just in case the nodes with primary shard has a problem.

On Saturday, November 17, 2012 12:11:15 AM UTC+1, Igor Motov wrote:

So, the problem here is that if you will even manage to place all primary
shards to the right places, the moment one of the nodes with primary shards
goes away all corresponding replicas will be promoted to primaries.

Could you talk a little bit about how you want to take advantage of
having primaries on HDD and replicas on SAN. Is the idea to use nodes with
HDD for requests and keep nodes that are using SAN as hot standby? What are
you trying to achieve by such allocation?

On Friday, November 16, 2012 12:08:48 PM UTC-5, elasticuser wrote:

Hi Martijn,

Because I have 10 servers with each about 500Go of hard drive space. So,
for my data, hard drive space is limited.
I would like to control where the primary shard and the replica shard
are stored in order to put my primary shard on the hdd and the replica
shard on a SAN.

On Friday, November 16, 2012 5:44:05 PM UTC+1, Martijn v Groningen wrote:

Hi,

What is the reason that you control the primary and replica shard
placement independently?
By default the primary shards's replicas aren't placed on the same
node.

Martijn

On 16 November 2012 15:41, elasticuser merik200...@yahoo.fr wrote:

It's possible to control shard placement on nodes. But how we can
control
primary shard and replica shard placement independently ?

Thanks.

--

--
Met vriendelijke groet,

Martijn van Groningen

--

Thanks for your help Igor.

On Saturday, November 17, 2012 2:23:49 PM UTC+1, Igor Motov wrote:

I see. That's an interesting idea. Unfortunately, I cannot think of a
mechanism that would allow you to do something like this. You can configure
a set of nodes to use HDD, and you can configure another set of nodes to
use SAN. You can use Allocation Awarenesshttp://www.elasticsearch.org/guide/reference/modules/cluster.htmlto make sure that if one shard is allocated on HDD node, its replica would
be allocated on SAN and vis versa. You can start HDD nodes first to make
sure they get all primaries shards. But that's it. There is really no way
to reassign primaries back to HDD nodes if they will get moved to SAN
nodes, or to limit searches only to HDD nodes when primaries are no longer
there.

On Saturday, November 17, 2012 4:12:31 AM UTC-5, elasticuser wrote:

My goal is to save space on the HDD. In my case, I have 5 To on my
cluster but with replica shards just 2.5 To. So, I would like to keep 5 To
for my primary shards and store the replica shards on a SAN.
I am aware that if a node with primary shards down all corresponding
replicas will be promoted to primaries but just during the time to repair
the node.
Indeed, I would like to use nodes with HDD for requests and nodes with
SAN just in case the nodes with primary shard has a problem.

On Saturday, November 17, 2012 12:11:15 AM UTC+1, Igor Motov wrote:

So, the problem here is that if you will even manage to place all
primary shards to the right places, the moment one of the nodes with
primary shards goes away all corresponding replicas will be promoted to
primaries.

Could you talk a little bit about how you want to take advantage of
having primaries on HDD and replicas on SAN. Is the idea to use nodes with
HDD for requests and keep nodes that are using SAN as hot standby? What are
you trying to achieve by such allocation?

On Friday, November 16, 2012 12:08:48 PM UTC-5, elasticuser wrote:

Hi Martijn,

Because I have 10 servers with each about 500Go of hard drive space.
So, for my data, hard drive space is limited.
I would like to control where the primary shard and the replica shard
are stored in order to put my primary shard on the hdd and the replica
shard on a SAN.

On Friday, November 16, 2012 5:44:05 PM UTC+1, Martijn v Groningen
wrote:

Hi,

What is the reason that you control the primary and replica shard
placement independently?
By default the primary shards's replicas aren't placed on the same
node.

Martijn

On 16 November 2012 15:41, elasticuser merik200...@yahoo.fr wrote:

It's possible to control shard placement on nodes. But how we can
control
primary shard and replica shard placement independently ?

Thanks.

--

--
Met vriendelijke groet,

Martijn van Groningen

--