Question of Resource Pool

Similar Questions

• Question about Resource Pools

  Hi guys,.

  I studied pools of resources in my lab at home in preparation for the review of VCP 4.  I read the pdf pools resources management:

  http://www.VMware.com/PDF/vSphere4/R40/vsp_40_resource_mgmt.PDF

  I'm a bit confused, I hope someone can clear this up for me.  A practical question indicates that a virtual machine will always fail to turn on if reservation of the virtual machine cannot be guaranteed, however in the KB article below he states that admission control can be disabled.

  http://KB.VMware.com/selfservice/microsites/search.do?language=en_US & cmd = displayKC & externalId = 1007006

  It is, so I think I understand the fact of not being able to guarantee a reservation of the machine does not necessarily prevent machine market because admission control can be disabled.  Is this correct?

  Reservations must be guaranteed as directed by your comment. Which relates to resource pools.

  Admission control is specifically related to HA and ensure that a virtual machine cannot be turned on if it's availability can be ensured according to admission control policy. HA can use reservations VM more memory General fresh in these calculations.

  Disable the admission control would allow you to power on a virtual machine even if the availability of this virtual machine could not be provided by HA, that is a different situation when the virtual machine lives in a pool of resources.

  Just because you use HA does not necessarily mean that you use resource pools or DRS

  Scott.

• Question about Lab Manager and ESX Resource Pools

  Hello world

  I was wondering if I can get feedback from some of the members of the community.  We used Lab Manager very strongly in our support organizations and it has proved to be a valuable tool.  Recently, we collaborated with the Department of technical training and started hosting online seminars and classes for them using Lab Manager.  Last week we had a fairly large class of users (approx. 45), each duty deploy 2 rather Beefy VMs that were very intensive resources (Jboss, SQL, Mail to name a few).  My colleague and I went through a lot of trials and planning to ensure that our infrastructure can handle the load while allowing our users to date at the same speed and reliability, so that they are used.

  Our Lab Manager ESX server pools made up of the following:

  5 HP DL380 G5 servers with 2 x Quad Core 2.8 GHZ processors (8 cores each) with 32 GB of RAM each

  2 HP DL580 G5 servers with 4 x Quad Core 2.8 GHZ processors (16 cores each) with 128 GB of RAM each

  When the class is running and everyone has made their 2 machines, we noticed for some reason, he was all on ESX2 and 3 deployment (380 s).  Then came the ESX alarms, use of 90% of memory, then more than 100% bringing the disk page.  The CPU usage was at a all time high.

  I look at the 2 580 s and there are about 4 virtual machines deployed on them.

  So my question is...

  How does he know where to launch a Virtual Lab Manager machine?  There's no sense 2 servers have been brought to the red zone, and almost overloaded when the 2 most powerful servers in my farm are nearly dormant.  I noticed sometimes in the past, but not this bad.  Normally, we have about 45-50 computers deployed at some point and it seems to spread them out properly.

  This group of training has access to the same LUN and ESX server with Betclick as any other organization that we have.

  We have decided hosting several sessions of training may be greater than this one and would like to know that the virtual machines will be distributed properly.

  I would like to know your opinion on this

  First to answer your question, then make guess what happens.

  As indicated in response Ilya, LM distributed VMs differently with active DRS resource pools and pools of resources without active DRS.

  LM place VMs when the DRS is not used.

  When DRS is used, we use the DRS admission control to select the host to place virtual machines.

  DRS is turned off, LM uses its own placement routine:

  For each virtual machine, LM filter all managed servers that cannot run the virtual machine. A complete list of patterns are:

  • The managed server has not enough memory or quota for virtual machine to run the virtual machine.

  • The host is not connected to the data store that is the virtual machine.

  • The virtual machine has active state (suspended) and the CPU of the server managed by the waking state has been captured is not compatible with the managed server currently considered.

  • The virtual machine has been a guest on 64-bit and the managed server is not 64-bit capable.

  • The virtual machine has more processors than the managed server.

  • The managed server is not accessible or is set to "prohibit deployments." Not reachable may mean that the lm-agent or its not to ping queries is not answer (this is visible in the list of managed server page).

  Once we have the complete list of eligible servers, we place the virtual machines on the host including the smallest (MemoryReservation % + CPUReservation %). By default, we reserve not CPU on LM VMs, this internship will be largely due to the memory on the host computers if you have not changed the settings of reservation of CPU on your virtual machines.

  For closed deploy to the PF3, force us all virtual machines to go on the same host, so if any virtual machines have any special data store / CPU type specification, all virtual machines is forced to go on this host. If there are conflicting requirements between different virtual machines in the same configuration, deployment will fail. Closed VMs also have another condition to check that is rarely achieved in practice: number of closed networks available on the managed server. By default, this is 20 and you will need more than 20 different configurations closed running on a managed server before you who will strike.  (LM4 allows Cross fencing host, so this does not apply in LM4).

  To make a guess blind because of your distribution problem - if you use the saved state and fencing, it's probably the type of processor.  Processor type can be checked on the host properties page.

  If you not are not I would check that the images can deploy on the 580 s by disabling the 380s and trying to deploy them.

  If you use DRS (which you're not), it could be due to deployment of virtual machines too fast for the algorithm of control for the entrance of VC.  An easy solution for this is to extinguish the DRS on the cluster manager of laboratory.

  We tested load internal repeatedly, the product QA in different laboratories, we run (LM has been used in VMworld Labs for the past couple years and internal systems of training and demonstration for our Organization implemented SE), so no need to worry about this.  I assure you that the first 'real' performance problem you will encounter will be almost certainly due to an overload of your storage array (too many IOPS / s)... and for that you have to distribute content in data warehouses and use techniques of creating paths to balance the load.

  Kind regards

  Steven

  Another thread on this: http://communities.vmware.com/message/1258902.

• The list of all the Resource Pools that are part of a group of ESXi

  Hello

  I'm trying to get a piece of code work that will list all the resource pools that are present in a cluster of ESXi.

  • I know I can find what list of resources of a virtual machine resides in.  I don't want that.

  // Find a matching resource pool on an ESXi hosts
    var esxiCluster = esxiHost.parent;
    var esxiClusterResourcePools = esxiCluster.resourcePool_ResourcePool;
  
  
    System.log("what is the ESXi cluster: " + esxiCluster.name);
    System.log("vimtype: " + esxiCluster.vimType);
    System.log("what pools exists: " + esxiClusterResourcePools);
  

  So I know that the vimType will come back with what I expect (ClusterComputeResource).  The question I have is the object "resourcePool" is reference to the root resource list (in my case, it is 'resources').

  What I want is all DRS resource pools are configured.  I need something like "resourcePool_ResourcePool.children".

  Any help is appreciated.

  Thank you

  B

  Hello

  Check if this code example that works in your case:

  var esxiCluster = esxiHost.parent;
  var esxiClusterResourcePools = esxiCluster.resourcePool_ResourcePool;
  var pools = System.getModule("com.vmware.library.vc.resourcePool").getAllChildResourcePool(esxiClusterResourcePools);
  
  // dump names of child pools
  for each (var pool in pools) {
    System.log(pool.name);
  }
  

• pass fails with a missing resource pool alert

  Hi, I have a server dell power edge r510, with ESXi 5.0 is installed in it.

  in one of the virtual machines within this server, I installed Vcenter (I think that it is Version 5.1) everything was stable until the vcenter database gets 10 GB and crashed and I couldn't fix it. So I installed a new windows with vcenter 5.5 inside vps. and for this new vcenter reconnected to the server. so, I've reconfigured resource pools and VPS users again in this new vcenter.

  Watch ESXi host now as does not not to the vcenter. but all virtual machines inside the esxi server work without problems.  I can't reconnect to the host in vcenter, and I can't connect directly to the ESXi via vsphere client and root password.

  but I can log on to the esxi via ssh or console.

  from the console, I restarted the management services, but it does not help.

  the SSH, I ran the following commands:

  status /etc/init.d/vpxa #.

  vpxa is running

  status /etc/init.d/hostd #.

  pass is not running.

  When I restart pass, it is said has started, but after a few minutes still pass fails.

  l # /etc/init.d/hostd restart

  Watchdog - pass: PID of the file/var/run/vmware/watchdog-hostd. PID does not exist

  Watchdog - pass: impossible to end guard dog: no monitoring running processes to pass

  SH: you must specify which kill

  [421281] Begin ' pass ++ min = 0, swap, group = pass etc/vmware/hostd/config.xml', min-uptime = 60, max-fast-failures = 1, max-total-failures = 1000000, bg_pid_file = "

  pass began.

  **********************************************************

  status /etc/init.d/hostd #.

  pass is running.

  ***************************************************************

  but I can't use all the controls of esxcli.

  List of the allowedip of esxcli network firewall ruleset

  Connect to locahost failed: connection failure

  ************************************************************

  status /etc/init.d/hostd #.

  pass is not running.

  ***************************************************************

  I searched the problem on localhost, Ko, said check the CPU usage is below 90%, which is good in my esxtop, then says check network ports and pings, they were right, too, said to check/etc/hosts to 127.0.0.1 localhost entry and checking /etc/resolv.conf and they were right too. as I can ping localhost and localhost nslookup returns 127.0.0.1

  then I started reading the server logs and found this error output of vm-support: hostd.log

  [cntos58-64-Farzin - 105.vmx'] could not get the parameters of resource for a virtual machine under tension

  Award of the VM ' / vmfs/volumes/.../cntos58-64-Farzin-105.vmx' to the root resource pool - configured pool 'pool11' does not exist

  SysInfo error on return operation State: verification of Admission of cpu resource failure. See the VMkernel detailed error information log

  cntos58-64-Farzin - 105.vmx: vim.fault.InsufficientCpuResourcesFault

  *************************************************************************

  How can I change the list of resources of the VM to ssh when esxcli do not work?

  I am facing this problem recently, almost all the 5 days the server will respond is not in vcenter, restart the entire esxi server can solve the problem but it apears again.

  I think that resource in vcenter pools doesn't sync properly with the configuration of esxi, as when I connect to the server directly with root/vshpere it shows different configuration than what I see in vcenter!

  and I can't upgrade esxi host because there are about 10 VPSes important and stable running on this server which is my main server and I can not have time on this server or migrate virtual machines to a different server/IPs.

  because I'm not sure of the compatibility of the versions of drivers/bios nic with ESXi 5.5 and I fear that if I update the ESXi from 5.0 to 5.5 I face other questions strangers/downtime.

  I had the server up and running again, by editing this file: /etc/vmware/hostd/pools.xml

  then I found the VPS with id of the pool of resources 11 evil, changed for another pool of resources as the other virtual machines.

  2014 07-29 T 08: 38:52.64965Z

  2

  pool366

  and then made a

  restart the /etc/init.d/hostd

  all started working again without restarting the whole esxi server and turn off virtual machines, I have searched for about 1 day to solve this simple thing!

  even once when I made a few changes in the vcenter, and then restarted resource pool settings move it in ssh, it fails under vcenter changed the VM resource for a bad configuration pool.

  so I changed the resource pool manually xml and restarted spend, while vcenter can connect to the esxi host.

  then vcenter, again removed all the resource pools and created new resource pools. and this question of time is set.

  There was some resource pool configuration that was imported from esxi host to vcenter which was no longer available and has not been deleted in syncronizations between esxi and vcenter when changed resource pools. and the list of the missing resources caused problems.

• Configuration issue Resource Pool

  We do currently not pools of resources within our environment. What we seek to do is implement with reserves resource pools memory based on our VM sizes. For example, we will create a pool of resources of 4 GB for virtual machines with 4 GB of memory, a pool of resources for virtual machines with 8 GB of memory, 8 GB etc. We would then define the reserve of memory halfway (the virtual machine) for each pool. Thus, the 4 GB VM will have a reserve of 2GB. Our goal in doing this is to reduce the size of our swap files. We have a few virtual machines spread in our environment that are configured independently. For example, a 12 GB VM with a reserve of 6 GB has a 6 GB swap file. At the global level, it could save us a lot of space to store data.

  A few questions while doing this, however. What will happen for the VMs who have already activated bookings? The resource pool overrides the individual VM settings?

  And no one sees this as having harmful effects on an HA event? We had problems in the past in a different environment with reservations (big 38 GB +), and when we lost a multitude of virtual machines have been unable to turn on because there was no more available resources.

  Resource pool reservations have no effect on the reservation of the virtual machine in the pool - they are independent

  Difficulties with HA is a question - another is by assigning reservations to all VM you can come to the point where an ESXi host won't have enough memory to satisfy the reserve causing the VM does not start.

• DRS rules and Resource Pools

  Is it possible to assign the DRS rules to Resource Pools instead of creating groups of DRS and adding to the DRS group VMs and then creating a rule
  I ask this question because of a problem that I currently have
  I am far from being an expert as you'll see soon and back to the use of support groups
  If it I'll give you my scenario
  I have 4 ESX host in my test group
  2 hosts are connected to zFCP
  And all 4 are connected to independent contractors
  I have a resorce pool call FC and another called ISCSI
  We constantly evolve around VM between these 2 pools
  The highest availability VM remains on the rest on ISCSI FC
  I would like to create a rule that States if a virtual machine is in the RP FC then he must reside on host A and B, and if it resides on ISCSI it can reside on any host
  I realize that there is still other issues involved here as data themselves warehouses
  but it is another matter that he owns the rights
  We migrate VMs and the data store when we leave the virtual machine to another RP 1
  If the rule would be always true
  I would like to hear from you
  on this topic
  for now I just use rules and groups of DRS
  Thank you

  I don't think this is possible from the point of view of rule affinity DRS, as he has no relationship with resources or VM disk pools. This seems to be a single-use resource pools.

  You could probably do with a set of scripts PowerCLI who would check two resource pools and ensure that the virtual machines to the breast have their records on the correct data store and if not do a svMotion.

• Understand how resource pools really work

  In the research on how to properly nail using the resources for the resource pools in our groups, I found that there are nuances that are not specifically documented about 'Booking' vs 'Limit' that relate to Resource Pools. If someone is an expert Resource Pool please chime. I am specifically looking for validation of what we believe is true.

  Here's the deal:

  So based on my interpretation of the response and our conception of resource pools, it looks like the 'limit' is always an important value that decided the full amount of the memory of the virtual computer (reserved memory + swapped memory shared memory + virtual memory) in a pool of resources to all virtual machines and does not imply that memory is drawn from a pool of resources of the parent.

  In our environment, we allow 'extensible booking '. Our question was, 'the value limit of importance. " We assumed that with the 'extensible' NOT VERIFIED that the value reserve limit didn't matter, even if it is not grayed out. So if the limit is not grayed out, it allows resources to be charged to the parent? I understand that this is not the case. In our case, the limit does not mean that the resources drawn from the parent, but seems rather this sets the cap of the memory of all virtual machines in the pool resource can use (reserved memory, expanded memory, shared memory + virtual memory).

  Example (1): we have a pool of resources called "Exchange-RP' 10 GB reserved, not expandable, and a limit of 20 GB. If we have 11 VMs each configured with 2 GB of ram (individual virtual machines have no reservation but the virtual machine is configured to 2 GB).

  We can power on the first 10 virtual machines (10 x 2 = 20 GB) but the 11 will not because it exceeds the limit.

  In this example how the 10VMs become the 2 GB? I guess everyone gets 1 GB of physical memory and consists of 1 GB of virtual memory (swap, shared, etc., not physical ram). At no time is issued physical memory of the mother.

  Example (2): given the exact scenario above. If the unlimited checkbox is enabled (limit is grey = unlimited), it would mean that the VM 11 would now be allowed to be turned on and booking of physical memory it is just must be shared by all the virtual machines in the pool. It seems that once that we receive too many virtual machines powered that there is too little physical ram per virtual computer and we will begin to see the excessive ballooning and swapping then in turn affect performance.

  Please confirm the above examples is accurate of how memory is managed.

  We are looking to validate that the RP limit value is always important as a strict limit of all memory used by the VMs, not only physical (reserved memory). None of the VMware documents speak to this.

  Thank you

  Jase

  Welcome to the Forums - example 1: I understand, it's the list resources should act as a host that you oversommit the memory assigned to the list of resources - so you should be able to on this 11th VM - with 11 VM all who are in need of 2 GB and with the limit set at 20 GB - if all VMs using their 2 GB and no page sharing occurring I expect to see ballooning by VM vmkernel swap files - now used if you have 2 GB reservations you would be able to turn on 5 VMs in this example because with 5 virtual machines would use you all 10 GB of memory reserved - value and the vmkernel will not power on a virtual machine if it cannot guarantee its reserve-

  Example 2 - see above - even with alimit together, you will be able to oversommit the memory in the RP - the only thing you will earn if you have the limit set to unlimited is that you be Basel to put multiple virtual machines and impact on all users on the ESx host or cluster -

  If you find this or any other answer useful please consider awarding points marking the answer correct or useful

• Use of the Resource Pool with vSphere Essentials

  A silly question but I have to ask anyway:

  I have a license of vSphere Essentials with my two hosts ESXi4.  I knew that he was not in vSphere DRS.  What was not so obvious to me, it's that you need DRS activated in order to create a pool of resources to the cluster level vSphere without it.  (The option to create a new cluster is grayed out). http://kb.vmware.com/selfservice/microsites/search.do?language=en_US & cmd = displayKC & externalId = 1004098

  So my question is:

  Are there benefits to creating a cluster if you do not have DRS?  I now have to double and create pools of resources at the level of the host for two guests, who is an administrative burden.  In addition, one of the main reasons I have used resource pools was to facilitate the access rights based upon roles to specific virtual machines.   These two will now have to be defined for each host.

  We greatly appreciate your comments on this

  Other than the ability to use the VCA mode and setting up the location of the pagefile.  According to me, there isn't much more to a cluster if you use no HA or DRS.  Since you do not have DRS yor cannot create pools of resources that span multiple hosts within a cluster.

  You will need to create individual pools of resources on each host by taking your host from a cluster and have just in the data center.

  OBM

• How to get the name of my virtual machine to a named resource pool

  Based on the script ghettoVCBg2 on the platform VIMA, I realized a backup solution for my ESXi (free) servers (also an automatic mailing is integrated ). My next goal is to integrate an automatic selection of my computer virtual based on the name of the list of resources. To understand, I have mainly two categories '1_Productive' and '2_Test', from resource pools. In both categories, there is a resource pool 'SRV' and 'CLT '.

  for example: 1_Productive-SRV-VM

  ------CLT-VM

  At the time of the backup, I have to manually enter the name of the virtual machine in a file. My question is how to integrate an automatic selection for example my pool of resources "SRV." My first idea was to shoot the 'vidiscover.pl' tool and run it for each server separately. Subsequently I merge files into a single file for ghettoVCBg2. You should know I want only to mit backup s excluded '1_Productive\SRV' VM '2_TEST\SRV '. But the name in double "SRV" can be replaced. A really great solution could be a direct integration in ghettoVCBg2

  Could someone help because I'am not very firm with perl programming...

  Thank you very much.

  Take a look at this review of the script: listVMsInResourcePool.pl

  =========================================================================

  William Lam

  VMware vExpert 2009

  Scripts for VMware ESX/ESXi and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

  repository scripts vGhetto

  VMware Code Central - Scripts/code samples for developers and administrators

  150 VMware developer

  Twitter: @lamw

  If you find this information useful, please give points to "correct" or "useful".

• Issue of Resource Pools

  I'm relatively new to the VMware Infrastructure. I read a lot of documentation available, but I'm having a hard time getting my head around Resource Pools. I was wondering if someone could point me to documentation that includes examples of how and why the Resource Pools are used in a real production environment. In the meantime, if anyone could answer the following questions, I would appreciate it:

  It seems to me that the RPs have been created to distribute resources according to how many resources needs of an organizational unit. For example if a unit is in need of 5 machines running at 2 mhz each with 2 GB of ram, then a suitable unexapandable PR for this unit would include 10 mhz and 10 gigabytes of ram and if any machine would be created which depleted resources, the machine would not be able to start. Is this correct? If so, then how the RP can manage how much mhz CPU using each machine? If all 5 machines were launched at the same time, would not be the limit of 10 mhz be quickly exhausted? The machine would be stopped / stop? Or would the machine just run with > less < mhz?

  RPs affect DRS or HA? How can he? Because the DRS moves a machine so she uses a fixed amount of resources on a machine, ESX, how RPs come into play if they are universal to the cluster?

  Thank you very much

  psynophile,

  Relevant issues. RPs are used "Undernet world", Yes, but I've not seen too real implementations using it. But everything I've seen, they work very well for purposes of limitation and booking. Answer your questions:

  "any machine would be created which depleted resources, the machine would not be able to start": it depends. The limitation will be strong for the RAM (the sixth VM fuel not on), but for CPU the ESX will try to share the 10 GHz (I guess than the typo in the MHz) with all the virtual machines on the RP - always count you can provide each machine with a reserve of CPU, so if the reservation is greater than the limit , then you may not be able to power on the nth VM.

  If so, then how the RP can manage how much mhz CPU using each machine? If all 5 machines were launched at the same time, would not be the limit of 10 mhz be quickly exhausted? : it depends on the use of each virtual computer processor. The performance will be degraded IF the limit of 10 GHz is not sufficient to them.if you 5 database within this RP servers, they will probably present a performance degradation. But will never halt/stop.

  RPs affect DRS or HA? : as the RPs are on a group level, they do not. DRS will change virtual machines one ESX to another depending on the host CPU/RAM usage. HA won't be affected, only that your reservations can reduce the amount of servers for failover, you have set up for it.

  Several docs on this subject: http://www.vmware.com/pdf/vi3_35/esx_3/r35u2/vi3_35_25_u2_resource_mgmt.pdf

  Hope this helps,

  Marcelo Soares

  VMWare Certified Professional 310

  Technical Support Engineer

  Chief Executive Officer of the Linux server

• Actions and resource pools

  Hello

  We meet the following "problem":

  We have created 2 resource pools: one with the actions 'CPU and memory' defined as the default value counter 'low', the other with the default value is 'normal '.

  So far, no problem.

  The question is this:

  When we look at the actions for each virtual machine, on the tab of the resource allocation to the title of the list of resources, we see a lot of action, defined as 'custom '. But we never put these customs.

  So my question:

  How can we explain all our virtual machines should only be managed by our pools of two resources (low and normal), and we see "customs" for many of them?

  Thank you.

  I confirm that I saw the same thing when I upgraded from 3.0.2 to 3.5.  The upgrade keeps the value of the action at the time and applied to the virtual machine, rather than keeping the parent value.  If you do some research, you will see the posts for this purpose.  A bit boring.  In fact, I used a powershell script to run through all the vm and put them back to normal.

  -KjB

  VMware vExpert

• VM / resource pool Configuration

  Now that I've answered my own question, and the little light in my head is enabled, I have a better understanding of the relationship between the virtual machines and Resource Pools.  But a follow-up question that I hope that experts can answer that for me is this: after I have create a pool of resources and define the limits of the resources, what I have to change anything on the virtual machines?  In other words, if the virtual machines are configured for unlimited CPU and memory, they are limited by the amount of correct pool?  What would be the reasoning to restrict the VMs and the memory, if I can control it by a pool of resources?  At least for me and please correct me if I'm wrong, if I put limits on virtual machines why I would put in resource pools?

  limits/actions/reservations for the virtual machines are now applied to resources being provided by the resource pool - for example - if you have 10 GB limit set for the pool of resources out of 64 GB of the bunch, it's-if you have a limited memory set to unlimited for the virtual machine in this resource pool the maximum amount of memory for the virtual machine will never be able to get 10 GB is However if the resource pool memory limit has the unlimited value of the virtual machine will be able to go to 64 GB

  If you find this or any other answer useful please consider awarding points marking the answer correct or useful

• Resource pools &amp; reservations?

  Hello

  My question pools of resources. I'll cut right to hunt.

  If a resource pool is created and a reserve of memory is set to say that this means-4069 MB each VM on the host is allocated physical memory 4 GB when it starts?

  When its given a reserve of say 4 GB it never will give the host it's memory if it is not needed?

  What happens if a guest is only given 2 GB of memory, but the reserve is set to 4 GB. It will only take 2GB or take all 4 GB host?

  Here is my scenario:

  An ESX host is given to 8 GB memory.

  4 VMs were created and given 2 GB of memory each.

  A Resource Pool is created and a reserve memory is set to 4 GB of memory.

  This VM average only 2 can be powered on at the same time? Or all 4 machines can be powered on?

  I wonder because someone went and created a bunch of pools of resources like this to my work. I'm new to the technology, but it seems as if they were incorrect installation. My guess is that they were created after the virtual machine has been powered...

  If a resource pool is created and a reserve of memory is set to say that this means-4069 MB each VM on the host is allocated physical memory 4 GB when it starts?

  N ° this means that the POOL reserves 4 GB to apply to these virtual machines.  ALL VM sharing the resource in a POOL.  Up to 4 GB is shared between the virtual machine in this pool.

  When its given a reserve of say 4 GB it never will give the host it's memory if it is not needed?

  NOPE.  Not once it is used, it is reserved.  If you want to share, just use stock instead of reservations for the swimming POOL.  Increase the priority and this POOL will have a higher priority.

  What happens if a guest is only given 2 GB of memory, but the reserve is set to 4 GB. It will only take 2GB or take all 4 GB host?

  YES.  Once reservations are applied, they don't get until they are not necessary.  Once a virtual computer in that pool gets hold of the memory, he won't give it back unless it is turned off.  But since you apply the reserve to a POOL it is immediate, so if your cluster has 32GB of memory, this POOL booked 4GB, leaving 28 GB to share the rest of the cluster, but only this POOL keeps at least 4 GB of memory to ensure that enough memory is given to its members.  This will be guaranteed to members of this POOL of memory.

  This VM average only 2 can be powered on at the same time? Or all 4 machines can be powered on?

  Well what is happening, is that the booking of the pool takes these and its guarnateed to get memory.  Other virtual machines are going to starve (if not left) until this pool is granted to ALL memory.  Once the pool is enough resource, reset of struggle of pools/VM without reservation on the remaining memory.

  Until you are absolutely certain that your VM do NOT get what they need, I would turn off reservations.  They should be used with caution if you don't know what you're doing, you can really tilt performance and cause problems.  So change the settings of the shared pool, back not book memory and monitor the performance of the virtual machine.  If there seems to be some conflict later, you can always reactivate.

• Is there a default resource pool?

  All - I read the white paper on esxtop that was written by Scott Drummond and can be found here - & gt; http://communities.VMware.com/docs/doc-9279 In this one, page 9, under the definition of 'State', he said:

  While the state memory of the host is not used to determine if the memory should be retrieved from virtual machines (which
  decision is made at the level of the resource pool), it may affect what mechanisms are used to retrieve the memory if necessary.
  In the States high and soft, ballooning is favored over permutation. In States hard and low, swapping is favored by report
  balloon flight

  So my question is, what happens if I do not defined resource pools?  Is there a default pool?  If not, are there any rehabilitation of the memory available for my guests/guests?

  "Every independent host and each DRS cluster has a pool of resources of root (invisible) which brings together the resources of the host or a cluster. The resource pool root does not appear because the resources of the host (or cluster) and the root resource pool are always the same. "- extract from http://pubs.vmware.com/vi35u2/resmgmt/wwhelp/wwhimpl/common/html/wwhelp.htm?context=resmgmt&file=vc_resource_pools.5.2.html