[ClusterLabs] Antw: Re: Antw: Re: Q: Resource balancing opration

Ulrich Windl Ulrich.Windl at rz.uni-regensburg.de
Thu Apr 21 10:23:21 UTC 2016


>>> Tomas Jelinek <tojeline at redhat.com> schrieb am 21.04.2016 um 10:20 in Nachricht
<57188D6B.5020001 at redhat.com>:
> Dne 21.4.2016 v 08:56 Ulrich Windl napsal(a):
>>>>> Ken Gaillot <kgaillot at redhat.com> schrieb am 20.04.2016 um 16:44 in Nachricht
>> <571795E5.4090004 at redhat.com>:
>>> On 04/20/2016 01:17 AM, Ulrich Windl wrote:
>>>> Hi!
>>>>
>>>> I'm wondering: If you boot a node on a cluster, most resources will go to
>>> another node (if possible). Due to stickiness configured, those resources
>>> will stay there.
>>>> So I'm wondering whether or how I could cause a rebalance of resources on
>>> the cluster. I must admit that I don't understand the details of stickiness
>>> related to other parameters. In my understanding stickiness should be 
> related
>>> to a percentage of utilization dynamically, so that a resource running on a
>>> node that is "almost full" should dynamically lower its stickiness to allow
>>> resource migration.
>>>>
>>>> So if you are going to implement a manual resource rebalance operation,
>>> could you dynamically lower the stickiness for each resource (by some amount
>>> or some factor), wait if something happens, and then repeat the process 
> until
>>> resources look balanced. "Looking balanced" should be no worse as if all
>>> resources are started when all cluster nodes are up.
>>>>
>>>> Spontaneous pros and cons for "resource rebalancing"?
>>>>
>>>> Regards,
>>>> Ulrich
>>>
>>> Pacemaker gives you a few levers to pull. Stickiness and utilization
>>> attributes (with a placement strategy) are the main ones.
>>>
>>> Normally, pacemaker *will* continually rebalance according to what nodes
>>> are available. Stickiness tells the cluster not to do that.
>>>
>>> Whether you should use stickiness (and how much) depends mainly on how
>>> significant is the interruption that occurs when a service is moved. For
>>
>> We agree on this: What I was asking for was a "manually triggered automatic 
> rebalance" that would temporarily override the stickiness parameters set for 
> the resources. Manually means people will blame me, not the cluster (at least 
> for starting the operation ;-)) if something bad happens.
> 
> Hi Ulrich,
> 
> You may want to take a look at "pcs resource relocate" commands which 
> are available in pcs-0.9.143 and newer (pcs-0.9.144 contains an 
> additional fix for dealing with remote nodes).
> 
> What pcs basically does is:
> * simulate what would pacemaker do if stickiness was set to 0 on resources
> * create constraints to move resources based on the simulation outcome
> * wait for the resources to move
> * remove the constraints
> 
> It is also possible to just display migrated resources or run the whole 
> rebalancing in a dry-run mode. You can specify which resources you want 
> to be rebalanced.

Hi!

That sounds great, but unfortunately that command is not available with crm shelland SLES11 SP4.

Ulrich

> 
> Regards,
> Tomas
> 
>>
>>> a large database supporting a high-traffic website, stopping and
>>> starting can take a long time and cost a lot of business -- so maybe you
>>> want an infinite stickiness in that case, and only rebalance manually
>>> during a scheduled window. For a small VM that can live-migrate quickly
>>> and doesn't affect any of your customer-facing services, maybe you don't
>>> mind setting a small or zero stickiness.
>>>
>>> You can also use rules to make the process intelligent. For example, for
>>> a server that provides office services, you could set a rule that sets
>>> infinite stickiness during business hours, and small or zero stickiness
>>> otherwise. That way, you'd get no disruptions when people are actually
>>> using the service during the day, and at night, it would automatically
>>> rebalance.
>>
>> Could you give a concrete example for this?
>>
>>>
>>> Normally, pacemaker's idea of "balancing" is to simply distribute the
>>> number of resources on each node as equally as possible. Utilization
>>> attributes and placement strategies let you add more intelligence. For
>>> example, you can define the number of cores per node or the amount of
>>> RAM per node, along with how much each resource is expected to use, and
>>> let pacemaker balance by that instead of just counting the number of
>>> resources.
>>
>> Knew that; I was specifically talking about the imbalance that occurs after 
> one node was down for service: If capacity allows the remaining nodes will 
> run all the services for the downed node, and they will stay there even if 
> the node is up again.
>>
>> Usually I want to avoid moving the resources, e.g. when one resource does 
> down (or up), causing an imbalance, causing other resources to be moved in 
> turn. Specifically if you know the resource will be back up (down) soon.
>>
>> I guess it's not possible to delay the rebalancing effect.
>>
>> Regards,
>> Ulrich
>>
>>>
>>> _______________________________________________
>>> Users mailing list: Users at clusterlabs.org 
>>> http://clusterlabs.org/mailman/listinfo/users 
>>>
>>> Project Home: http://www.clusterlabs.org 
>>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
>>> Bugs: http://bugs.clusterlabs.org 
>>
>>
>>
>>
>> _______________________________________________
>> Users mailing list: Users at clusterlabs.org 
>> http://clusterlabs.org/mailman/listinfo/users 
>>
>> Project Home: http://www.clusterlabs.org 
>> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
>> Bugs: http://bugs.clusterlabs.org 
>>
> 
> _______________________________________________
> Users mailing list: Users at clusterlabs.org 
> http://clusterlabs.org/mailman/listinfo/users 
> 
> Project Home: http://www.clusterlabs.org 
> Getting started: http://www.clusterlabs.org/doc/Cluster_from_Scratch.pdf 
> Bugs: http://bugs.clusterlabs.org 







More information about the Users mailing list