You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@cassandra.apache.org by Peng Xiao <25...@qq.com> on 2017/07/24 09:58:02 UTC
tolerate how many nodes down in the cluster
Hi,
Suppose we have a 30 nodes cluster in one DC with RF=3,
how many nodes can be down?can we tolerate 10 nodes down?
it seems that we are not able to avoid the data distribution 3 replicas in the 10 nodes?,
then we can only tolerate 1 node down even we have 30 nodes?
Could anyone please advise?
Thanks
Re: 回复： tolerate how many nodes down in the cluster
Posted by Anuj Wadehra <an...@yahoo.co.in.INVALID>.
Hi Brooke,
Very nice presentation:https://www.youtube.com/watch?v=QrP7G1eeQTI !! Good to know that you are able to leverage Racks for gainingoperational efficiencies. I think vnodes have made life easier.
I still see some concerns with Racks:
1. Usually scaling needs are driven by business requirements. Customerswant value for every penny they spend. Adding 3 or 5 servers (because you haveRF=3 or 5) instead of 1 server costs them dearly. It's difficult to justify theadditional cost as fault tolerance can only be improved but not guaranteed with racks.
2. You need to maintain mappings of Logical Racks (=RF) andphysical racks (multiple of RFs) for large clusters.
3. Using racks tightlycouples your hardware (rack size, rack count) / virtualization decisions (VMSize, VM count per physical node) with application RF.
Thanks
Anuj
On Tuesday, 25 July 2017 3:56 AM, Brooke Thorley <br...@instaclustr.com> wrote:
Hello Peng.
I think spending the time to set up your nodes into racks is worth it for the benefits that it brings. With RF3 and NTS you can tolerate the loss of a whole rack of nodes without losing QUORUM as each rack will contain a full set of data. It makes ongoing cluster maintenance easier, as you can perform upgrades, repairs and restarts on a whole rack of nodes at once. Setting up racks or adding nodes is not difficult particularly if you are using vnodes. You would simply add nodes in multiples of <num racks> to keep the racks balanced. This is how we run all our managed clusters and it works very well.
You may be interested to watch my Cassandra Summit presentation from last year in which I discussed this very topic: https://www.youtube.com/watch?v=QrP7G1eeQTI (from 4:00)
If you were to consider changing your rack topology, I would recommend that you do this by DC migration rather than "in place".
Kind Regards,Brooke ThorleyVP Technical Operations & Customer Servicessupport@instaclustr.com | support.instaclustr.com
Read our latest technical blog posts here.This email has been sent on behalf of Instaclustr Pty. Limited (Australia) and Instaclustr Inc (USA).This email and any attachments may contain confidential and legally privileged information. If you are not the intended recipient, do not copy or disclose its content, but please reply to this email immediately and highlight the error to the sender and then immediately delete the message.
On 25 July 2017 at 03:06, Anuj Wadehra <an...@yahoo.co.in.invalid> wrote:
Hi Peng,
Three things are important when you are evaluating fault tolerance and availability for your cluster:
1. RF2. CL3. Topology - how data is replicated in racks.
If you assume that N nodes from ANY rack may fail at the same time, then you can afford failure of RF-CL nodes and still be 100% available. E. g. If you are reading at quorum and RF=3, you can only afford one (3-2) node failure. Thus, even if you have a 30 node cluster, 10 node failure can not provide you 100% availability. RF impacts availability rather than total number of nodes in a cluster.
If you assume that N nodes failing together will ALWAYS be from the same rack, you can spread your servers in RF physical racks and use NetworkTopologyStrategy. While allocating replicas for any data, Cassandra will ensure that 3 replicas are placed in 3 different racks E.g. you can have 10 nodes in 3 racks and then even a 10 node failure within SAME rack shall ensure that you have 100% availability as two replicas are there for 100% data and CL=QUORUM can be met. I have not tested this but that how the rack concept is expected to work. I agree, using racks generally makes operations tougher.
ThanksAnuj
On Mon, 24 Jul 2017 at 20:10, Peng Xiao<25...@qq.com> wrote: Hi Bhuvan,From the following link,it doesn't suggest us to use RAC and it looks reasonable.http://www.datastax.com/dev/ blog/multi-datacenter- replication
Defining one rack for the entire cluster is the simplest and most common implementation. Multiple racks should be avoided for the following reasons: • Most users tend to ignore or forget rack requirements that state racks should be in an alternating order to allow the data to get distributed safely and appropriately. • Many users are not using the rack information effectively by using a setup with as many racks as they have nodes, or similar non-beneficial scenarios. • When using racks correctly, each rack should typically have the same number of nodes. • In a scenario that requires a cluster expansion while using racks, the expansion procedure can be tedious since it typically involves several node moves and has has to ensure to ensure that racks will be distributing data correctly and evenly. At times when clusters need immediate expansion, racks should be the last things to worry about.
------------------ 原始邮件 ------ ------------发件人: "Bhuvan Rawal";<bh...@gmail.com>;发送时间: 2017年7月24日(星期一) 晚上7:17收件人: "user"<user@cassandra. apache.org>; 主题: Re: tolerate how many nodes down in the cluster
Hi Peng ,
This really depends on how you have configured your topology. Say if you have segregated your dc into 3 racks with 10 servers each. With RF of 3 you can safely assume your data to be available if one rack goes down.
But if different servers amongst the racks fail then i guess you are not guaranteeing data integrity with RF of 3 in that case you can at max lose 2 servers to be available. Best idea would be to plan failover modes appropriately and letting cassandra know of the same.
Regards,Bhuvan
On Mon, Jul 24, 2017 at 3:28 PM, Peng Xiao <25...@qq.com> wrote:
Hi,
Suppose we have a 30 nodes cluster in one DC with RF=3,how many nodes can be down?can we tolerate 10 nodes down?it seems that we are not able to avoid the data distribution 3 replicas in the 10 nodes?,then we can only tolerate 1 node down even we have 30 nodes?Could anyone please advise?
Thanks
Re: 回复： tolerate how many nodes down in the cluster
Posted by Brooke Thorley <br...@instaclustr.com>.
Hello Peng.
I think spending the time to set up your nodes into racks is worth it for
the benefits that it brings. With RF3 and NTS you can tolerate the loss of
a whole rack of nodes without losing QUORUM as each rack will contain a
full set of data. It makes ongoing cluster maintenance easier, as you can
perform upgrades, repairs and restarts on a whole rack of nodes at once.
Setting up racks or adding nodes is not difficult particularly if you are
using vnodes. You would simply add nodes in multiples of <num racks> to
keep the racks balanced. This is how we run all our managed clusters and
it works very well.
You may be interested to watch my Cassandra Summit presentation from last
year in which I discussed this very topic:
https://www.youtube.com/watch?v=QrP7G1eeQTI (from 4:00)
If you were to consider changing your rack topology, I would recommend that
you do this by DC migration rather than "in place".
Kind Regards,
*Brooke Thorley*
*VP Technical Operations & Customer Services*
support@instaclustr.com | support.instaclustr.com
<https://www.instaclustr.com/>
<https://www.facebook.com/instaclustr> <https://twitter.com/instaclustr>
<https://www.linkedin.com/company/instaclustr>
Read our latest technical blog posts here
<https://www.instaclustr.com/blog/>.
This email has been sent on behalf of Instaclustr Pty. Limited (Australia)
and Instaclustr Inc (USA).
This email and any attachments may contain confidential and legally
privileged information. If you are not the intended recipient, do not copy
or disclose its content, but please reply to this email immediately and
highlight the error to the sender and then immediately delete the message.
On 25 July 2017 at 03:06, Anuj Wadehra <an...@yahoo.co.in.invalid>
wrote:
> Hi Peng,
>
> Three things are important when you are evaluating fault tolerance and
> availability for your cluster:
>
> 1. RF
> 2. CL
> 3. Topology - how data is replicated in racks.
>
> If you assume that N nodes from ANY rack may fail at the same time, then
> you can afford failure of RF-CL nodes and still be 100% available. E. g.
> If you are reading at quorum and RF=3, you can only afford one (3-2) node
> failure. Thus, even if you have a 30 node cluster, 10 node failure can not
> provide you 100% availability. RF impacts availability rather than total
> number of nodes in a cluster.
>
> If you assume that N nodes failing together will ALWAYS be from the same
> rack, you can spread your servers in RF physical racks and use
> NetworkTopologyStrategy. While allocating replicas for any data, Cassandra
> will ensure that 3 replicas are placed in 3 different racks E.g. you can
> have 10 nodes in 3 racks and then even a 10 node failure within SAME rack
> shall ensure that you have 100% availability as two replicas are there for
> 100% data and CL=QUORUM can be met. I have not tested this but that how the
> rack concept is expected to work. I agree, using racks generally makes
> operations tougher.
>
>
> Thanks
> Anuj
>
>
>
> On Mon, 24 Jul 2017 at 20:10, Peng Xiao
> <25...@qq.com> wrote:
> Hi Bhuvan,
> From the following link,it doesn't suggest us to use RAC and it looks
> reasonable.
> http://www.datastax.com/dev/blog/multi-datacenter-replication
>
> Defining one rack for the entire cluster is the simplest and most common
> implementation. Multiple racks should be avoided for the following reasons:
> • Most users tend to ignore or forget rack requirements that state racks
> should be in an alternating order to allow the data to get distributed
> safely and appropriately.
> • Many users are not using the rack information effectively by using a
> setup with as many racks as they have nodes, or similar non-beneficial
> scenarios.
> • When using racks correctly, each rack should typically have the same
> number of nodes.
> • In a scenario that requires a cluster expansion while using racks, the
> expansion procedure can be tedious since it typically involves several node
> moves and has has to ensure to ensure that racks will be distributing data
> correctly and evenly. At times when clusters need immediate expansion,
> racks should be the last things to worry about.
>
>
>
>
>
> ------------------ 原始邮件 ------------------
> *发件人:* "Bhuvan Rawal";<bh...@gmail.com>;
> *发送时间:* 2017年7月24日(星期一) 晚上7:17
> *收件人:* "user"<us...@cassandra.apache.org>;
> *主题:* Re: tolerate how many nodes down in the cluster
>
> Hi Peng ,
>
> This really depends on how you have configured your topology. Say if you
> have segregated your dc into 3 racks with 10 servers each. With RF of 3 you
> can safely assume your data to be available if one rack goes down.
>
> But if different servers amongst the racks fail then i guess you are not
> guaranteeing data integrity with RF of 3 in that case you can at max lose 2
> servers to be available. Best idea would be to plan failover modes
> appropriately and letting cassandra know of the same.
>
> Regards,
> Bhuvan
>
> On Mon, Jul 24, 2017 at 3:28 PM, Peng Xiao <25...@qq.com> wrote:
>
> Hi,
>
> Suppose we have a 30 nodes cluster in one DC with RF=3,
> how many nodes can be down?can we tolerate 10 nodes down?
> it seems that we are not able to avoid the data distribution 3 replicas
> in the 10 nodes?,
> then we can only tolerate 1 node down even we have 30 nodes?
> Could anyone please advise?
>
> Thanks
>
>
>
回复： 回复： tolerate how many nodes down in the cluster
Posted by Peng Xiao <25...@qq.com>.
as per Brooke suggests,RACs a multipile of RF.
https://www.youtube.com/watch?v=QrP7G1eeQTI
if we have 6 machines with RF=3,then we can set up 6 RACs or setup 3RACs,which will be better?
Could you please further advise?
Many thanks
------------------ 原始邮件 ------------------
发件人: "我自己的邮箱";<25...@qq.com>;
发送时间: 2017年7月26日(星期三) 晚上7:31
收件人: "user"<us...@cassandra.apache.org>;
抄送: "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
主题: 回复： 回复： tolerate how many nodes down in the cluster
One more question.why the # of racks should be equal to RF?
For example,we have 4 machines,each virtualized to 8 vms ,can we set 4 RACs with RF3?I mean one machine one RAC.
Thanks
------------------ 原始邮件 ------------------
发件人: "我自己的邮箱";<25...@qq.com>;
发送时间: 2017年7月26日(星期三) 上午10:32
收件人: "user"<us...@cassandra.apache.org>;
抄送: "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
主题: 回复： 回复： tolerate how many nodes down in the cluster
Thanks for the remind,we will setup a new DC as suggested.
------------------ 原始邮件 ------------------
发件人: "kurt greaves";<ku...@instaclustr.com>;
发送时间: 2017年7月26日(星期三) 上午10:30
收件人: "User"<us...@cassandra.apache.org>;
抄送: "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
主题: Re: 回复： tolerate how many nodes down in the cluster
Keep in mind that you shouldn't just enable multiple racks on an existing cluster (this will lead to massive inconsistencies). The best method is to migrate to a new DC as Brooke mentioned.
回复： 回复： tolerate how many nodes down in the cluster
Posted by Peng Xiao <25...@qq.com>.
One more question.why the # of racks should be equal to RF?
For example,we have 4 machines,each virtualized to 8 vms ,can we set 4 RACs with RF3?I mean one machine one RAC.
Thanks
------------------ 原始邮件 ------------------
发件人: "我自己的邮箱";<25...@qq.com>;
发送时间: 2017年7月26日(星期三) 上午10:32
收件人: "user"<us...@cassandra.apache.org>;
抄送: "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
主题: 回复： 回复： tolerate how many nodes down in the cluster
Thanks for the remind,we will setup a new DC as suggested.
------------------ 原始邮件 ------------------
发件人: "kurt greaves";<ku...@instaclustr.com>;
发送时间: 2017年7月26日(星期三) 上午10:30
收件人: "User"<us...@cassandra.apache.org>;
抄送: "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
主题: Re: 回复： tolerate how many nodes down in the cluster
Keep in mind that you shouldn't just enable multiple racks on an existing cluster (this will lead to massive inconsistencies). The best method is to migrate to a new DC as Brooke mentioned.
回复： 回复： tolerate how many nodes down in the cluster
Posted by Peng Xiao <25...@qq.com>.
Thanks for the remind,we will setup a new DC as suggested.
------------------ 原始邮件 ------------------
发件人: "kurt greaves";<ku...@instaclustr.com>;
发送时间: 2017年7月26日(星期三) 上午10:30
收件人: "User"<us...@cassandra.apache.org>;
抄送: "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
主题: Re: 回复： tolerate how many nodes down in the cluster
Keep in mind that you shouldn't just enable multiple racks on an existing cluster (this will lead to massive inconsistencies). The best method is to migrate to a new DC as Brooke mentioned.
Re: 回复： tolerate how many nodes down in the cluster
Posted by kurt greaves <ku...@instaclustr.com>.
Keep in mind that you shouldn't just enable multiple racks on an existing
cluster (this will lead to massive inconsistencies). The best method is to
migrate to a new DC as Brooke mentioned.
回复： 回复： tolerate how many nodes down in the cluster
Posted by Peng Xiao <25...@qq.com>.
Thanks All for your reply.We will begin using RACs in our C* cluster.
Thanks.
------------------ 原始邮件 ------------------
发件人: "kurt greaves";<ku...@instaclustr.com>;
发送时间: 2017年7月25日(星期二) 上午6:27
收件人: "User"<us...@cassandra.apache.org>; "anujw_2003@yahoo.co.in"<an...@yahoo.co.in>;
抄送: "Peng Xiao"<25...@qq.com>;
主题: Re: 回复： tolerate how many nodes down in the cluster
I've never really understood why Datastax recommends against racks. In those docs they make it out to be much more difficult than it actually is to configure and manage racks.
The important thing to keep in mind when using racks is that your # of racks should be equal to your RF. If you have keyspaces with different RF, then it's best to have the same # as the RF of your most important keyspace, but in this scenario you lose some of the benefits of using racks.
As Anuj has described, if you use RF # of racks, you can lose up to an entire rack without losing availability. Note that this entirely depends on the situation. When you take a node down, the other nodes in the cluster require capacity to be able to handle the extra load that node is no longer handling. What this means is that if your cluster will require the other nodes to store hints for that node (equivalent to the amount of writes made to that node), and also handle its portion of READs. You can only take out as many nodes from a rack as the capacity of your cluster allows.
I also strongly disagree that using racks makes operations tougher. If anything, it makes them considerably easier (especially when using vnodes). The only difficulty is the initial setup of racks, but for all the possible benefits it's certainly worth it. As well as the fact that you can lose up to an entire rack (great for AWS AZ's) without affecting availability, using racks also makes operations on large clusters much smoother. For example, when upgrading a cluster, you can now do it a rack at a time, or some portion of a rack at a time. Same for OS upgrades or any other operation that could happen in your environment. This is important if you have lots of nodes. Also it makes coordinating repairs easier, as you now only need to repair a single rack to ensure you've repaired all the data. Basically any operation/problem where you need to consider the distribution of data, racks are going to help you.
Re: 回复： tolerate how many nodes down in the cluster
Posted by kurt greaves <ku...@instaclustr.com>.
I've never really understood why Datastax recommends against racks. In
those docs they make it out to be much more difficult than it actually is
to configure and manage racks.
The important thing to keep in mind when using racks is that your # of
racks should be equal to your RF. If you have keyspaces with different RF,
then it's best to have the same # as the RF of your most important
keyspace, but in this scenario you lose some of the benefits of using racks.
As Anuj has described, if you use RF # of racks, you *can* lose up to an
entire rack without losing availability. Note that this entirely depends on
the situation. *When you take a node down, the other nodes in the cluster
require capacity to be able to handle the extra load that node is no longer
handling. *What this means is that if your cluster will require the other
nodes to store hints for that node (equivalent to the amount of writes made
to that node), and also handle its portion of READs. You can only take out
as many nodes from a rack as the capacity of your cluster allows.
I also strongly disagree that using racks makes operations tougher. If
anything, it makes them considerably easier (especially when using vnodes).
The only difficulty is the initial setup of racks, but for all the possible
benefits it's certainly worth it. As well as the fact that you can lose up
to an entire rack (great for AWS AZ's) without affecting availability,
using racks also makes operations on large clusters much smoother. For
example, when upgrading a cluster, you can now do it a rack at a time, or
some portion of a rack at a time. Same for OS upgrades or any other
operation that could happen in your environment. This is important if you
have lots of nodes. Also it makes coordinating repairs easier, as you now
only need to repair a single rack to ensure you've repaired all the data.
Basically any operation/problem where you need to consider the distribution
of data, racks are going to help you.
Re: 回复： tolerate how many nodes down in the cluster
Posted by Anuj Wadehra <an...@yahoo.co.in.INVALID>.
Hi Peng,
Three things are important when you are evaluating fault tolerance and availability for your cluster:
1. RF2. CL3. Topology - how data is replicated in racks.
If you assume that N nodes from ANY rack may fail at the same time, then you can afford failure of RF-CL nodes and still be 100% available. E. g. If you are reading at quorum and RF=3, you can only afford one (3-2) node failure. Thus, even if you have a 30 node cluster, 10 node failure can not provide you 100% availability. RF impacts availability rather than total number of nodes in a cluster.
If you assume that N nodes failing together will ALWAYS be from the same rack, you can spread your servers in RF physical racks and use NetworkTopologyStrategy. While allocating replicas for any data, Cassandra will ensure that 3 replicas are placed in 3 different racks E.g. you can have 10 nodes in 3 racks and then even a 10 node failure within SAME rack shall ensure that you have 100% availability as two replicas are there for 100% data and CL=QUORUM can be met. I have not tested this but that how the rack concept is expected to work. I agree, using racks generally makes operations tougher.
ThanksAnuj
On Mon, 24 Jul 2017 at 20:10, Peng Xiao<25...@qq.com> wrote: Hi Bhuvan,From the following link,it doesn't suggest us to use RAC and it looks reasonable.http://www.datastax.com/dev/blog/multi-datacenter-replication
Defining one rack for the entire cluster is the simplest and most common implementation. Multiple racks should be avoided for the following reasons: • Most users tend to ignore or forget rack requirements that state racks should be in an alternating order to allow the data to get distributed safely and appropriately. • Many users are not using the rack information effectively by using a setup with as many racks as they have nodes, or similar non-beneficial scenarios. • When using racks correctly, each rack should typically have the same number of nodes. • In a scenario that requires a cluster expansion while using racks, the expansion procedure can be tedious since it typically involves several node moves and has has to ensure to ensure that racks will be distributing data correctly and evenly. At times when clusters need immediate expansion, racks should be the last things to worry about.
------------------ 原始邮件 ------------------发件人: "Bhuvan Rawal";<bh...@gmail.com>;发送时间: 2017年7月24日(星期一) 晚上7:17收件人: "user"<us...@cassandra.apache.org>; 主题: Re: tolerate how many nodes down in the cluster
Hi Peng ,
This really depends on how you have configured your topology. Say if you have segregated your dc into 3 racks with 10 servers each. With RF of 3 you can safely assume your data to be available if one rack goes down.
But if different servers amongst the racks fail then i guess you are not guaranteeing data integrity with RF of 3 in that case you can at max lose 2 servers to be available. Best idea would be to plan failover modes appropriately and letting cassandra know of the same.
Regards,Bhuvan
On Mon, Jul 24, 2017 at 3:28 PM, Peng Xiao <25...@qq.com> wrote:
Hi,
Suppose we have a 30 nodes cluster in one DC with RF=3,how many nodes can be down?can we tolerate 10 nodes down?it seems that we are not able to avoid the data distribution 3 replicas in the 10 nodes?,then we can only tolerate 1 node down even we have 30 nodes?Could anyone please advise?
Thanks
回复： tolerate how many nodes down in the cluster
Posted by Peng Xiao <25...@qq.com>.
Hi Bhuvan,
From the following link,it doesn't suggest us to use RAC and it looks reasonable.
http://www.datastax.com/dev/blog/multi-datacenter-replication
Defining one rack for the entire cluster is the simplest and most common implementation. Multiple racks should be avoided for the following reasons:
• Most users tend to ignore or forget rack requirements that state racks should be in an alternating order to allow the data to get distributed safely and appropriately.
• Many users are not using the rack information effectively by using a setup with as many racks as they have nodes, or similar non-beneficial scenarios.
• When using racks correctly, each rack should typically have the same number of nodes.
• In a scenario that requires a cluster expansion while using racks, the expansion procedure can be tedious since it typically involves several node moves and has has to ensure to ensure that racks will be distributing data correctly and evenly. At times when clusters need immediate expansion, racks should be the last things to worry about.
------------------ 原始邮件 ------------------
发件人: "Bhuvan Rawal";<bh...@gmail.com>;
发送时间: 2017年7月24日(星期一) 晚上7:17
收件人: "user"<us...@cassandra.apache.org>;
主题: Re: tolerate how many nodes down in the cluster
Hi Peng ,
This really depends on how you have configured your topology. Say if you have segregated your dc into 3 racks with 10 servers each. With RF of 3 you can safely assume your data to be available if one rack goes down.
But if different servers amongst the racks fail then i guess you are not guaranteeing data integrity with RF of 3 in that case you can at max lose 2 servers to be available. Best idea would be to plan failover modes appropriately and letting cassandra know of the same.
Regards,
Bhuvan
On Mon, Jul 24, 2017 at 3:28 PM, Peng Xiao <25...@qq.com> wrote:
Hi,
Suppose we have a 30 nodes cluster in one DC with RF=3,
how many nodes can be down?can we tolerate 10 nodes down?
it seems that we are not able to avoid the data distribution 3 replicas in the 10 nodes?,
then we can only tolerate 1 node down even we have 30 nodes?
Could anyone please advise?
Thanks
Re: tolerate how many nodes down in the cluster
Posted by Bhuvan Rawal <bh...@gmail.com>.
Hi Peng ,
This really depends on how you have configured your topology. Say if you
have segregated your dc into 3 racks with 10 servers each. With RF of 3 you
can safely assume your data to be available if one rack goes down.
But if different servers amongst the racks fail then i guess you are not
guaranteeing data integrity with RF of 3 in that case you can at max lose 2
servers to be available. Best idea would be to plan failover modes
appropriately and letting cassandra know of the same.
Regards,
Bhuvan
On Mon, Jul 24, 2017 at 3:28 PM, Peng Xiao <25...@qq.com> wrote:
> Hi,
>
> Suppose we have a 30 nodes cluster in one DC with RF=3,
> how many nodes can be down?can we tolerate 10 nodes down?
> it seems that we are not able to avoid the data distribution 3 replicas
> in the 10 nodes?,
> then we can only tolerate 1 node down even we have 30 nodes?
> Could anyone please advise?
>
> Thanks
>