mongodb-user
[Arriba] [Todas las Listas]

Re: [mongodb-Usuario] mongodb config server produse hight IO en disco

To: mongodb-user <mongodb-user@xxxxxxxxxxxxxxxx>
Subject: Re: [mongodb-Usuario] mongodb config server produse hight IO en disco
From: Alexander <al.al.platonov@xxxxxxxxx>
Date: Tue, 15 Sep 2015 05:23:37 -0700 (PDT)
Delivery-date: Tue, 15 Sep 2015 08:35:53 -0400
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlegroups.com; s=20120806; h=date:from:to:message-id:in-reply-to:references:subject:mime-version :content-type:x-original-sender:reply-to:precedence:mailing-list :list-id:x-spam-checked-in-group:list-post:list-help:list-archive :sender:list-subscribe:list-unsubscribe; bh=N9gHsjoT9giHKCv8Vos77+5w2vitqq4NNkSZyZZbsh0=; b=MLuLRZsdTMPM2vxI/Tbs5a050oSkNL8bBcNeyI1aLaSFlYozPeZM35YgH8ffoQIfa7 qamGcwgvkS+4TdkwpM6FrKyCLABbhJMQ7yRPlSHP7I1OBn8EEDtZTKL7W5hE4HqiMLNV H9L2uDwYfkKy+O3gR0LArXxuOA6NxV46AgRse7UUreIu2DxrQsjU0l+n4HgQ6bPhw+OX f6kvfreNI288DZq0PfRpPI+HNGYVkHkiSMezkT5MOuirQXVoBXq3hTTx1ir9eNb4TIpQ RzS9LAhctPBvdOpxTVVM7teeHdnR9pdWCDIeGSk5KTEMFlVtRC2k1DMfaSEteqnKRDYt wV4A==
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=date:from:to:message-id:in-reply-to:references:subject:mime-version :content-type:x-original-sender:reply-to:precedence:mailing-list :list-id:x-spam-checked-in-group:list-post:list-help:list-archive :sender:list-subscribe:list-unsubscribe; bh=N9gHsjoT9giHKCv8Vos77+5w2vitqq4NNkSZyZZbsh0=; b=fD5XpfOOxTxPhkzPSyD+mVGI+7IcI/Ejs/zi6bSZp+ujOTc2cyKom++QFH3IAW+0L5 xv9PFT6Tv1eWz+IR1pMYHYTm30s7Sor700ObOxXx/g5Uu/F7cyzDRse/Z3L53QDwcWgY r1gUO8cX/v9lCyfldV64f1g1c8/Tj2eLNUn9hk1O3HJpG2NMFs2bx3UWgFF6ilLTDelt h3N/00+UyCg228zGvelSqmDlMl56/nX9Qd2a+84XMQTDLGD18JoWDAEhd+KnMI7Nvxmx lFuh21osazdWMTF+SYNmNAF9hZiJxRYBIVyQHfEQbf0G1ObE0hC/qnxaXklLnh/bejb8 sU6g==
Envelope-to: traductor@xxxxxxxxxxx
In-reply-to: <CAOe6dJDsx=kuDPO+qQt4KuJk09QO1PJau3ht4uRqnv_sw5NLpg@mail.gmail.com>
List-archive: <http://groups.google.com/group/mongodb-use>
List-help: <http://groups.google.com/support/>, <mailto:mongodb-user+help@googlegroups.com>
List-id: <mongodb-user.googlegroups.com>
List-post: <http://groups.google.com/group/mongodb-user/post>, <mailto:mongodb-user@googlegroups.com>
List-subscribe: <http://groups.google.com/group/mongodb-user/subscribe>, <mailto:mongodb-user+subscribe@googlegroups.com>
List-unsubscribe: <mailto:googlegroups-manage+1044811755470+unsubscribe@googlegroups.com>, <http://groups.google.com/group/mongodb-user/subscribe>
Mailing-list: list mongodb-user@xxxxxxxxxxxxxxxx; contact mongodb-user+owners@xxxxxxxxxxxxxxxx
References: <11fcc8a1-82b7-4fca-99f7-c21d25a4c4b2@googlegroups.com> <CAOe6dJDsx=kuDPO+qQt4KuJk09QO1PJau3ht4uRqnv_sw5NLpg@mail.gmail.com>
Reply-to: mongodb-user@xxxxxxxxxxxxxxxx
Sender: mongodb-user@xxxxxxxxxxxxxxxx

 No piensas que quizás esto es normal?  Cuándo el *balancer está corriendo, 
es *logging #varios información sobre lo que está haciendo al *config *server 
(entre otras cosas).

Pienso que esto es normal cuándo *i tiene carga de trabajo o cuándo *i *whait algún tiempo 
después de carga de trabajo de la parón. *Maby No es normal cuándo *i crea grupo nuevo, 
*shard *database, *presplit colecciones, *mongorestore dato y no dar carga 
de trabajo para un tiempo largo.

Tan *i soy un preocupado sobre él e intentar entender por qué esto pasa, *maby *i 
 *smth mal.

понедельник, 14 сентября 2015 г., 18:22:47 *UTC+3 пользователь *Asya *Kamsky 
написал:
>
> > cuándo *i parón *balancer carga de disco de mi *config *server como 0, entonces *i 
> inicio *balancer y carga de disco alto otra vez
>
>  no piensas que quizás esto es normal?  Cuándo el *balancer está corriendo, 
> es *logging #varios información sobre lo que está haciendo al *config *server 
> (entre otras cosas).
>
> Cuándo el *balancer no está corriendo, hay ninguna carga de disco...
>
> *Asya
>
>
> En *Mon, *Sep 14, 2015 en 7:16 AM, Alexander <al.al.p...@xxxxxxxxx 
> <*javascript:>> escribió:
>
>> *Hi.
>>
>> Intento entender por qué mi *mongodb *config *server producto *hight IO en 
>> disco. Mi cuestión parece como disco Muy alto IO en *Mongodb *Config *servers 
>> <*http://*serverfault.*com/Cuestiona/539167/muy-alto-disco-*io-encima-*mongodb-*config-*servers> 
>> tema.
>>
>> Típico *iostat -*xnt 3 producción:
>> Dispositivo:         *rrqm/*s   *wrqm/*s     *r/*s     *w/*s    *rMB/*s    *wMB/*s *avgrq-*sz 
>> *avgqu-*sz   espera *r_esperar *w_esperar  *svctm  %*util
>> *vda               0,00    11,67    0,00  205,00     0,00     0,24     2,
>> 39     1,21    3,51    0,00    3,51   2,58  52,80
>>
>> Esto es un entorno de prueba y *i tiene ningún problema hasta *i  algunas pruebas 
>> de carga.
>>
>> Hago unos cuantos experimentos y *definetly seguro aquello:
>> - mi *config *server el disco carga independiente de carga de trabajo
>> - cuándo *i parón *balancer carga de disco de mi *config *server como 0, entonces *i 
>> inicio *balancer y carga de disco alto otra vez
>> - cuándo *balancer *enebled su estado es *blinked 0/1/2/0/1/2 todos cronometran
>>
>> tengo grupo con 6 *shards y sólo un *sharded *database - *apikeys con 99 
>> *sharded colecciones, todas las  colecciones son incluso:
>> *sh.Estado()
>> *apikeys.*hourly_Contrarresta99
>> *shard llave: "#unknown{^*s" : 1 }
>>                *chunks:
>>                 *ps-*rs2    2
>>                 *ps-*rs3    2
>>                 *ps-*rs4    1
>>                 *ps-*rs5    1
>>                 *ps-*rs6    1
>>                 *ps-*rs1    2 "#unknown{^*s" : "$#unknown{^*minKey" : 1 } } -- "#unknown{^*s" : 2 } encima : *ps-*rs2 
>> *Timestamp(2, 0) "#unknown{^*s" : 2 } -- "#unknown{^*s" : 3 } encima : *ps-*rs3 *Timestamp(3, 0) "#unknown{^*s" : 3 } -- "#unknown{^*s" : 4 } encima : *ps-*rs4 *Timestamp(4, 0) "#unknown{^*s" : 4 } -- "#unknown{^*s" : 5 } encima : *ps-*rs5 *Timestamp(5, 0) "#unknown{^*s" : 5 } -- "#unknown{^*s" : 6 } encima : *ps-*rs6 *Timestamp(6, 0) "#unknown{^*s" : 6 } -- "#unknown{^*s" : 7 } encima : *ps-*rs2 *Timestamp(7, 0) "#unknown{^*s" : 7 } -- "#unknown{^*s" : 8 } encima : *ps-*rs3 *Timestamp(8, 0) "#unknown{^*s" : 8 } -- "#unknown{^*s" : 9 } encima : *ps-*rs1 *Timestamp(8, 1) "#unknown{^*s" : 9 } -- "#unknown{^*s" : "$#unknown{^*maxKey" : 1 } } encima : *ps-*rs1 
>> *Timestamp(1, 16) 
>>
>> Cada hora *i escribir dato a una colección nueva y hacer algunos *statistic cuenta. 
>> El Dato parece este
>> *db.*hourly_Contrarresta1.Encuentra() "_#unknown{^*id" : 1, "*c" : 1, "*s" : 1, "*v" : 0 "_#unknown{^*id" : 2, "*c" : 1, "*s" : 2, "*v" : 0 "_#unknown{^*id" : 3, "*c" : 1, "*s" : 3, "*v" : 0 "_#unknown{^*id" : 4, "*c" : 1, "*s" : 4, "*v" : 0 "_#unknown{^*id" : 5, "*c" : 1, "*s" : 5, "*v" : 0 "_#unknown{^*id" : 12, "*c" : 1, "*s" : 12, "*v" : 0 "_#unknown{^*id" : 6, "*c" : 1, "*s" : 6, "*v" : 0 "_#unknown{^*id" : 7, "*c" : 1, "*s" : 7, "*v" : 0 "_#unknown{^*id" : 21, "*c" : 2, "*s" : 3, "*v" : 0 "_#unknown{^*id" : 22, "*c" : 2, "*s" : 4, "*v" : 0 }
>>
>> *s la variable puede ser 0 - 18 (*i sabe aquello no es perfeccionar *sharded).
>>
>> Tengo un *log archivo con *systemlog *verbosity: 5. *Default *verbosity Espectáculo 
>> sólo *fsync mensaje y no informativo.
>> He *mongodb 2.6.11 y *i prueba 2.6.4, 2.6.5 versiones. Yo *completly 
>> *recreate grupo y restaurar dato.
>>
>> Si *u tiene cualesquier cuestiones, *welcome :)
>> Y *thx para ayuda.
>>
>> -- 
>> Recibiste este mensaje porque eres *subscribed al *Google Grupos 
>> "*mongodb-grupo"
>> de usuario.
>>  
>> Para otro *MongoDB opciones de apoyo técnico, ve: 
>> *http://www.mongodb.org/sobre/apoyo/.
>> --- 
>> Recibiste este mensaje porque eres *subscribed al *Google Grupos 
>> "*mongodb-grupo" de usuario.
>> A *unsubscribe de este grupo y la parón que recibe *emails de él, enviar un 
>> *email a *mongodb-user...@xxxxxxxxxxxxxxxx <*javascript:>.
>> A correo a este grupo, envía *email a mongod...@xxxxxxxxxxxxxxxx 
>> <*javascript:>.
>> Visita este grupo en *http://grupos.*google.*com/Grupo/*mongodb-usuario.
>> Para ver esta discusión en la visita de web 
>> *https://grupos.*google.*com/*d/*msgid/*mongodb-Usuario/11*fcc8#uno1-82*b7-4*fca-99*f7-*c21*d25#uno4*c4*b2%40*googlegroups.*com 
>> <*https://Grupos.*google.*com/*d/*msgid/*mongodb-Usuario/11*fcc8#uno1-82*b7-4*fca-99*f7-*c21*d25#uno4*c4*b2%40*googlegroups.*com?*utm_Medio=*email&*utm_fuente=*footer>
>> .
>> Para más opciones, visita *https://grupos.*google.*com/*d/*optout.
>>
>
>

-- 
Recibiste este mensaje porque eres *subscribed al *Google Grupos "*mongodb-grupo"
de usuario.

Para otro *MongoDB opciones de apoyo técnico, ve: *http://www.mongodb.org/sobre/apoyo/.
--- 
Recibiste este mensaje porque eres *subscribed al *Google Grupos "*mongodb-grupo" de usuario.
A *unsubscribe de este grupo y la parón que recibe *emails de él, enviar un *email a *mongodb-usuario+unsubscribe@xxxxxxxxxxxxxxxx.
A correo a este grupo, envía *email a *mongodb-user@xxxxxxxxxxxxxxxx.
Visita este grupo en *http://grupos.*google.*com/Grupo/*mongodb-usuario.
Para ver esta discusión en la visita de web *https://grupos.*google.*com/*d/*msgid/*mongodb-Usuario/6263*e0*b1-552un-447*c-8089-2*d82*e1*ece915%40*googlegroups.*com.
Para más opciones, visita *https://grupos.*google.*com/*d/*optout.

Don't you think that maybe this is normal?  When the balancer is running, 
it's logging various information about what it's doing to the config server 
(among other things).

I think this is normal when i have work load or when i whait some time 
after stop work load. Maby it is not normal when i create new cluster, 
shard database, presplit collections, mongorestore data and not give work 
load for a long time.

So i am a worried about it and try to understand why this happens, maby i 
do smth wrong.

понедельник, 14 сентября 2015 г., 18:22:47 UTC+3 пользователь Asya Kamsky 
написал:
>
> > when i stop balancer disk load from my config server like 0, then i 
> start balancer and disk load high again
>
> Don't you think that maybe this is normal?  When the balancer is running, 
> it's logging various information about what it's doing to the config server 
> (among other things).
>
> When the balancer is not running, there is no disk load...
>
> Asya
>
>
> On Mon, Sep 14, 2015 at 7:16 AM, Alexander <al.al.p...@xxxxxxxxx 
> <javascript:>> wrote:
>
>> Hi.
>>
>> I try to understand why my mongodb config server produce hight IO on 
>> disk. My question seems like Very high disk IO on Mongodb Config servers 
>> <http://serverfault.com/questions/539167/very-high-disk-io-on-mongodb-config-servers> 
>> topic.
>>
>> typical iostat -xnt 3 output:
>> Device:         rrqm/s   wrqm/s     r/s     w/s    rMB/s    wMB/s avgrq-sz 
>> avgqu-sz   await r_await w_await  svctm  %util
>> vda               0,00    11,67    0,00  205,00     0,00     0,24     2,
>> 39     1,21    3,51    0,00    3,51   2,58  52,80
>>
>> This is a test environment and i have no troubles until i do some load 
>> tests.
>>
>> I make a few experiments and definetly sure that:
>> - my config server disk load independent from work load
>> - when i stop balancer disk load from my config server like 0, then i 
>> start balancer and disk load high again
>> - when balancer enebled his status is blinked 0/1/2/0/1/2 all time
>>
>> I have cluster with 6 shards and only one sharded database - apikeys with 
>> 99 sharded collections, all collections are even:
>> sh.status()
>> apikeys.hourly_counters99
>> shard key: { "s" : 1 }
>>                chunks:
>>                 ps-rs2    2
>>                 ps-rs3    2
>>                 ps-rs4    1
>>                 ps-rs5    1
>>                 ps-rs6    1
>>                 ps-rs1    2
>>             { "s" : { "$minKey" : 1 } } -->> { "s" : 2 } on : ps-rs2 
>> Timestamp(2, 0) 
>>             { "s" : 2 } -->> { "s" : 3 } on : ps-rs3 Timestamp(3, 0) 
>>             { "s" : 3 } -->> { "s" : 4 } on : ps-rs4 Timestamp(4, 0) 
>>             { "s" : 4 } -->> { "s" : 5 } on : ps-rs5 Timestamp(5, 0) 
>>             { "s" : 5 } -->> { "s" : 6 } on : ps-rs6 Timestamp(6, 0) 
>>             { "s" : 6 } -->> { "s" : 7 } on : ps-rs2 Timestamp(7, 0) 
>>             { "s" : 7 } -->> { "s" : 8 } on : ps-rs3 Timestamp(8, 0) 
>>             { "s" : 8 } -->> { "s" : 9 } on : ps-rs1 Timestamp(8, 1) 
>>             { "s" : 9 } -->> { "s" : { "$maxKey" : 1 } } on : ps-rs1 
>> Timestamp(1, 16) 
>>
>> Every hour i write data to a new collection and do some statistic count. 
>> Data look like this
>> db.hourly_counters1.find()
>> { "_id" : 1, "c" : 1, "s" : 1, "v" : 0 }
>> { "_id" : 2, "c" : 1, "s" : 2, "v" : 0 }
>> { "_id" : 3, "c" : 1, "s" : 3, "v" : 0 }
>> { "_id" : 4, "c" : 1, "s" : 4, "v" : 0 }
>> { "_id" : 5, "c" : 1, "s" : 5, "v" : 0 }
>> { "_id" : 12, "c" : 1, "s" : 12, "v" : 0 }
>> { "_id" : 6, "c" : 1, "s" : 6, "v" : 0 }
>> { "_id" : 7, "c" : 1, "s" : 7, "v" : 0 }
>> { "_id" : 21, "c" : 2, "s" : 3, "v" : 0 }
>> { "_id" : 22, "c" : 2, "s" : 4, "v" : 0 }
>>
>> s variable can be 0 - 18 (i know that's not perfect sharded).
>>
>> I have a log file with systemlog verbosity: 5. Default verbosity show 
>> only fsync message and not informative.
>> I have mongodb 2.6.11 and i try 2.6.4, 2.6.5 versions. I completly 
>> recreate cluster and restore data.
>>
>> If u have any questions, welcome :)
>> And thx for help.
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "mongodb-user"
>> group.
>>  
>> For other MongoDB technical support options, see: 
>> http://www.mongodb.org/about/support/.
>> --- 
>> You received this message because you are subscribed to the Google Groups 
>> "mongodb-user" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to mongodb-user...@xxxxxxxxxxxxxxxx <javascript:>.
>> To post to this group, send email to mongod...@xxxxxxxxxxxxxxxx 
>> <javascript:>.
>> Visit this group at http://groups.google.com/group/mongodb-user.
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/mongodb-user/11fcc8a1-82b7-4fca-99f7-c21d25a4c4b2%40googlegroups.com 
>> <https://groups.google.com/d/msgid/mongodb-user/11fcc8a1-82b7-4fca-99f7-c21d25a4c4b2%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups "mongodb-user"
group.

For other MongoDB technical support options, see: http://www.mongodb.org/about/support/.
--- 
You received this message because you are subscribed to the Google Groups "mongodb-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mongodb-user+unsubscribe@xxxxxxxxxxxxxxxx.
To post to this group, send email to mongodb-user@xxxxxxxxxxxxxxxx.
Visit this group at http://groups.google.com/group/mongodb-user.
To view this discussion on the web visit https://groups.google.com/d/msgid/mongodb-user/6263e0b1-552a-447c-8089-2d82e1ece915%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
<Anterior por Tema] Tema Actual [Siguiente por Tema>