Restarting the management agents on ESXi 4.1 / risk to restart the guest virtual machines?

Unfortunately, we must restart management on our HP ML350 G6 agents almost once a week (unable to connect with VSphere client more, customers are works well).

We know this Article -: restart on an ESX or ESXi Server management agents

and this article useful: http://www.vm-help.com/esx/esx3i/ESXi_restart_agents.php

How can we avoid the ESXi 4.1 command /sbin/services.sh restart will restart our execution of virtual machines (guests) as well?

(Our first test shows: it restarts the guests, but it is sure, or just happy coincidence?)

According to the second article linked:

... This will restart the agents installed in /etc/init.d/ and with a default installation that includes: pass, ntpd, sfcbd, sfcbd-watchdog, slpd and wsmand. He is aalso restarting the agent VMware HA (AAM) if that has been activated on the host...

Each of the agents of this is responsible for our virtual computer operation invited?

Do we ensure "auto start/stop" to each virtual machine too (as written in "Attention" on the first linked article)?

I think it was on ESX 3.5 U3 corected. In any case, this happened only if you autostart enabled for VMS - the ability of ESX Server to boot VMs when it starts itself. If you do not need, leave it disabled.

Each of the agents of this is responsible for our virtual computer operation invited?

Not at all. Virtual machines running in a very low level of the VMkernel. Even if you lose the COS (which is your main problem here - you should check it...) the VMs will still work.

Marcelo Soares

VMWare Certified Professional 310/410

Master virtualization technology

Globant Argentina

Review the allocation of points for "useful" or "right" answers.

Tags: VMware

Similar Questions

  • Restructuring of the pagefile to the guest virtual machines

    Is there a directive to rigtsize the pagefile for guest virtual machines (Windows)?

    With the hosted versions of VMware (Workstation and Server), I used to create a swap file whose size was RAM * 2 and place it on a different disk from the system disk / boot: can it be good for VMware ESXi as weel?

    Is there a better solution?

    Concerning

    Marius

    that will depend entirely on the Organization of the underlying hardware and the OS and app loading on these drives.  Having said that, generally, is not a big deal provided that machines have anyway enough RAM.

  • Use of memory in the guest virtual machines

    Hello

    I don't have any background on VMware technology, but we are using a facility of dynamic cloud based on vCloud Director to the company where I work.

    The group that operates the server publishes the offer with a 'pay per use model"of service.

    This is true for the CPU resources, as we pay only what we use in the month. But in my opinion, it does not for the use of the RAM.

    We have Linux VMs that allocate a 4 GB of RAM 'physical' and our monthly bills show that the RAM usage is at 100% for all the month. But when you look at the tools of memory for the virtual machine, you can see that most of the time 50% or even less is used by the guest operating system process.

    This means from a user perspective the RAM price per month should be half of what we are responsible now. Ask operators on the 'question', we always get the answer that the use of the RAM is not dynamic and we need to pay for the use of 100% from the point of view of the server host.

    Could someone share some light on this subject? Is there really no way to make use of the dynamic RAM as the CPU usage?

    Thank you

    Michael

    Evening,

    As a service provider, I'll let you in on the secret:

    It depends on

    vCloud takes care of three models of default allocation.  Your provide may use other

    -Pay for what you use. -pay as you use

    -Allowance - you get resources, but can burst more

    -Booking - you get allocated resources and is

    Chargeback depends on model (in out of the box chargeback... your provider might have something custom)

    Pay as you use - charged on the basis of the resources used

    Allowance - remuneration based on allocation and additional for burst

    Allocate pay based on booking-

    There in the last two models really no way to load except full allowance (i.e. 100% cpu and RAM)

    Statement, I think you have something along the lines of the compensation you are using you mixed with a price system vendor.

    From my experience the resource more used in ESXi clusters in the RAM.  That's really what I buy in clusters of resource... I always need more RAM.  I need no more CPU. In addition, distribution of the CPU to vcloud is made via Ghz is not CPU.   RAM is allocated by GB.   So those set up and I think you suppliers is betting that they will always have to RAM so they charge based on usage not allocation.  While CPU, they have tons while they charge by usage.

    Really, it's just my thoughts... What can charge you on:

    -Ghz processor

    -GB of RAM

    -GB of storage

    How can recharge you:

    -Used

    -Allocated

    So bottom line... they can charge by opportunity they choose not too...

    It will be useful,

    J

  • the specified virtual machine is currently not runing ID no: c1034a70 exchange System Manager

    I get this error when I want to mount the store point
    in exchange server 2003

    the specified virtual machine is currently not runing ID no: c1034a70 exchange System Manager

    Labor exchange in cluster

    Hi Mohamed Essa,

    The question you posted would be better suited in the Forums of Exchange Server. I would recommend posting your query in the TechNet Forums.

    TechNet (Exchange Server) Forum

    http://social.technet.Microsoft.com/forums/en-us/category/ExchangeServer/

    I hope this helps.

  • Manual removal placeholder VMs when the protected virtual machine is removed

    We run SRM 5.5. We had some VMs protected, we need more, so they have been removed from the drive (I can't). When I found out about it, I went in SRM and removed the protection for these deleted virtual machines. Most of the VMS placeholder have been correctly removed by this procedure, however, I see still 2 VMs of space reserved for virtual computers that no longer exist. No idea why only these 2 (about 6) were not cleaned.

    My question: can I remove manually just these VMs placeholder? They are not in any group of protection, or the stimulus plan. And the protected virtual machines which correspond to these placeholders no longer exist. It's as if they are zombie placeholders. :-)

    There is no problem with me removing the VMs placeholder manually, y at - it?

    Hello

    Yes, its safe to delete the place holder vm.  Make sure the source VM is not protected and no replication is configured. Restart the SRM service.

    Thank you

  • How to set up the generic dashboard for the specified virtual machines

    I have two question about "generic dashboard":

    1. I use the tag to classify certain virtual machines by role QA, DEV or PRD. If I select "Kind of resources", and then select all VM under a label, window "metrics selector" will be watch only "vCenter generated Operation Manager", it does not show metric VM. How can I see it?

    2. If I select the 'Resources' mode, then I'll be able to see the VM metrics on the window "metrics selector", but I must key in each VM name "Label to the area", it takes a lot of time if I have a lot of VM, is there no matter what symbol I can use to match this field with name column "resource"?

    Mode 'Resources like' or 'Resource', both ok for me, I want to just install a widget to display using the specified CPU VM real fast.

    Any information is appreciated!

    You will see that vC generated Ops metric because you are looking at the tag, which is a resourcekind. This switch of resources is "not" a filter label, these are the actual resources. If you want to see all the virtual machines, select the resourcekind 'virtual machine '.

    Generic dashboard is rarely used to show - all - kinds of resources, it is generally used for static selection for specific purposes. If you want to display all types of a resource type and apply a tag filter, you can select a different widget to display this information.

    There is no symbols that can act as wildcards, as you suggest. When you are 'kind of resources' mode, you need not fill out a label for the box.

  • Is it possible to transfer everything to a virtual hard disk that is smaller on a virtual hard disk that is larger in VMWare Workstation, using the same virtual machine?

    Hello

    I was wondering if it is possible to transfer everything to a virtual hard disk that is smaller on a virtual hard disk that is larger in VMWare Workstation, using the same virtual machine? In the affirmative, please could someone advise me on how to do it. Also utility management and disk transfer can be used in virtual machines to do this?

    Thank you very much

    Tim Auluck

    I'm not sure what exactly you are asking. Can you please clarify!

    Basically you can resize a virtual drive or add a second virtual disk to a virtual machine and transfer/copy files. You can also resize the virtual disk using VMware Converter for example.

    André

  • Could not create the Java virtual machine

    Hello

    I am using VM ware and installed Linux operating system.

    In my MedRecDomain of Weblogic Admin Server to start and also to start MedRecSvr1.

    But MedRecSvr2 & 3 do not stand up.

    Details:

    OS: Linux

    Details of the JVM: JAVA memory arguments: -Xms512m-Xmx512m

    WebLogic Version: 10.3.6

    Please see the Journal as shown below,

    [root@ram bin] #./startManagedWebLogic.sh MedRecSvr2 &

    [1] 3194

    [root@ram bin] #.

    .

    Arguments of memory in JAVA:-Xms512m-Xmx512m

    .

    Start mode of WLS = Production

    .

    CLASSPATH=/opt/labs/ows/wlshome/patch_wls1036/profiles/default/sys_manifest_classpath/weblogic_patch.jar:/opt/labs/ows/wlshome/patch_ocp371/profiles/default/sys_manifest_classpath/weblogic_patch.jar:/opt/labs/ows/wlshome/jrockit_160_29_D1.2.0-10/lib/tools.jar:/opt/labs/ows/wlshome/wlserver_10.3/server/lib/weblogic_sp.jar:/opt/labs/ows/wlshome/wlserver_10.3/server/lib/weblogic.jar:/opt/labs/ows/wlshome/modules/features/weblogic.server.modules_10.3.6.0.jar:/opt/labs/ows/wlshome/wlserver_10.3/server/lib/webservices.jar:/opt/labs/ows/wlshome/modules/org.apache.ant_ 1.7.1/lib/ant-all.jar:/opt/labs/ows/wlshome/modules/net.sf.antcontrib_1.1.0.0_1-0b2/lib/ant-contrib.jar:/opt/labs/ows/wlshome/wlserver_10.3/common/derby/lib/derbyclient.jar:/opt/labs/ows/wlshome/wlserver_10.3/server/lib/xqrl.jar

    .

    PATH=/opt/labs/ows/wlshome/wlserver_10.3/server/bin:/opt/labs/ows/wlshome/modules/org.apache.ant_1.7.1/bin:/opt/labs/ows/wlshome/jrockit_160_29_D1.2.0-10/jre/bin:/opt/labs/ows/wlshome/jrockit_160_29_D1.2.0-10/bin:/usr/lib/qt-3.3/bin:/usr/local/sbin:/usr/sbin:/sbin:/usr/local/bin:/usr/bin:/bin:/root/bin

    .

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

    * To start WebLogic Server, use a username and *.

    * password assigned to an administrator-level user.  For *.

    * server administration, using the WebLogic Server *.

    * console in http://hostname/ReportServer: port/console         *

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

    from weblogic with the Java version:

    Java version "1.6.0_29".

    Java (TM) SE Runtime Environment (build 1.6.0_29 - b11)

    Oracle JRockit (R) (build R28.2.0-79-146777-1.6.0_29-20111005-1807-linux-ia32, update mode)

    WLS starting with line:

    /opt/labs/ows/wlshome/jrockit_160_29_D1.2.0-10/bin/java - jrockit-Xms512m-Xmx512m-Dweblogic.Name=MedRecSvr2-Djava.security.policy=/opt/labs/ows/wlshome/wlserver_10.3/server/lib/weblogic.policy-Dweblogic.ProductionModeEnabled=true-Dweblogic.security.SSL.trustedCAKeyStore=/opt/labs/ows/wlshome/wlserver_10.3/server/lib/cacerts-da-Dplatform.home=/opt/labs/ows/wlshome/wlserver_10.3-Dwls.home=/opt/labs/ows/wlshome/wlserver_10.3/server-Dweblogic.home=/opt/labs/ows/wlshome/wlserver_10.3/server-Dweblogic.management.discover=false-Dweblogic.management.server= http://10.0.2.15:7020 -Dwlw.iterativeDev=false-Dwlw.testConsole=false-Dwlw.logErrorsToConsole=false-Dweblogic.ext.dirs=/opt/labs/ows/wlshome/patch_wls1036/profiles/default/sysext_manifest_classpath:/opt/labs/ows/wlshome/patch_ocp371/profiles/default/sysext_manifest_classpath weblogic. Server

    [ERROR] [memory] Could not commit 524288 KB memory heap at 0x95e09000.

    Could not create the Java virtual machine.


    Also, when the managed server stop also get the same question,

    [root@ram bin] #./stopManagedWebLogic.sh MedRecSvr2

    If you stop Weblogic Server...

    [ERROR] [memory] Could not commit 524288 KB memory heap at 0x95e09000.

    Could not create the Java virtual machine.

    Fact

    Stopping Server Derby...

    Derby server stopped.

    [root@ram ~] # cat/proc/meminfo

    MemTotal: 511620 kB

    MemFree: 5224 kB

    Stamps: 760 kB

    Caching: 20440 kB

    SwapCached: 109492 kB

    Active principles: 219424 kB

    Inactive: 220788 kB

    Active (Anon): 210272 kB

    Inactive (Anon): 210408 kB

    Active (leader): 9152 kB

    Inactive (queue): 10380 kB

    Unevictable: 0 kB

    Mlocked: 0 kB

    HighTotal: 0 kB

    HighFree: 0 kB

    LowTotal: 511620 kB

    LowFree: 5224 kB

    SwapTotal: 1048568 kB

    SwapFree: 1516 KB

    Sale: 12 KB

    Writeback: 512 KB

    AnonPages: 309740 kB

    Mapped: 12684 kB

    Shmem: 1628 kB

    Slab: 48500 kB

    SReclaimable: 12196 kB

    SUnreclaim: 36304 kB

    KernelStack: 2656 kB

    PageTables: 9104 kB

    NFS_Unstable: 0 kB

    Bounce: 0 kB

    WritebackTmp: 0 kB

    CommitLimit: 1304376 kB

    Committed_AS: 2210364 kB

    VmallocTotal: 505912 kB

    VmallocUsed: 5196 kB

    VmallocChunk: 486476 kB

    HugePages_Total: 0

    HugePages_Free: 0

    HugePages_Rsvd: 0

    HugePages_Surp: 0

    Hugepagesize: 2048 kB

    DirectMap4k: 6080 kB

    DirectMap2M: 518144 kB

    Please help identify the problem.

    Thnks in advance

    Kind regards

    Bouchra Arun

    Web Site: http://www.industryvertical.co.in/

    People,

    This issue has been resolved,

    Details of the problem:

    RAM 512 MB only, use JVM parameters

    "" Arguments of memory in JAVA: -Xms512m-Xmx512m ".

    If WL struck to use more memory for the start and the stop MedRecSvr2, 3 (managed server)

    Please read the following steps to correct,

    1. open Oracle VM Virtual box

    2. go to settings and select the system option (make sure that all OS turns off).

    3. go in card mother in this window select Base memory previously is changed to 1920 MB (1.875 GB) 512 MB.

    4. click on Ok.

    Then the allocated RAM will be more, for Weblogic doesn't get knocked because of insufficient RAM memory.


    Respect,

    Bouchra Arun

    Portrait of the industry

    Please Mark Helpful / good answer

  • is it a good idea to run the JAVA virtual machine inside the hypervisor?

    If I want to write a global user inside the hypervisor platform and you want to reuse some java lib, it's a good idea to install and run the JAVA virtual machine as part of global programs of the user inside the hypervisor? Thank you.

    Maybe you listen to the wrong people?

    The ESXi hypervisor is pretty well stripped of the linux console you are talking about, there is a thin busybox linux like environment, but you are strongly advised not to run anything in there.

    Performer of applications within this console may have all kinds of side effects such as severely affect the performance of your computer virtual guests.

    Why the hell would you want to run the JAVA virtual machine inside the hypervisor? What is the reason why you want to do this? If it's because you get a better return, then you're wrong, the console becomes just a little bit of resources available on your host computer.

    If this is to go directly to the features that the hypervisor provides then the VMA we're talking is the answer.

    So getting back to your question initial for EACH use case that I can think it's a very bad idea.

    --

    Wil

  • Installation BAR inside the Android virtual machine files.

    Hello everyone!

    Can someone explain to me the following problem:

    I have an Android app that can install any other Android Apps (think like that as a personal App Store for Android). Now, I want to convert to BlackBerry. What I have done.

    Now what I want to know, is how my application (BAR the file running inside the Android virtual machine) to install other files BAR? Y at - it any option (API?) that can escape from VM Android for QNX for I might be able to install a BAR?

    Thank you!

    Ricardo Teixeira

    This is not supported.  Third-party applications on BBX/BlackBerry Tablet OS are not able to install other third-party applications.

  • Cannot create an instance of the java virtual machine

    Hi when I try to launch my sql developer get this error

    Cannot create an instance of the java virtual machine located in the path C:\Program Files\Java\jdk1.6.0_38\jre\bin\client\jvm.dll, but I do not have the file in the location

    I'm trying to do

    BI CHAOS: SQL Developer: cannot create an instance of the Virtual Machine Java

    Cannot create an instance of the Virtual Machine Java

    but am not able to locate file product.conf in my machine

    OK, we lost time because you said never explicitly release of SQL Developer in use, and then, when I deduced this wrong, you went right with it, rather than correct me.  Remember the line from Blade Runner when Harrison Ford hits Rutger Hauer in the head with a pipe?  Than hurt!

    Anyway, since the last post (because of the osgi references in the file ide.conf) I see that you are using a version 4.0.x of SQL Developer.  You must update the file product.conf with the smallest Xmx value in order to create an instance of the JVM.  If you have not yet found the product.conf file on your system, this might help: http://www.thatjeffsmith.com/archive/2013/12/oracle-sql-developer-4-windows-and-the-jdk/

    Otherwise, ask someone in your organization COMPUTER help - keep in mind that AppData is one of those hidden Windows directories, so you'll have to tell explore to make it visible by the organize > Folder Options > view > hidden files and folders parameters.  In addition, you should use Java 7, not Java 6 - try to get that upgrade to the latest version of jdk1.7.0_xx (currently 60, I think).  Since you never said if your JDK is 32-bit or 64-bit, I can not say to use an AddVMOption, Add32VMOption or Add64VMOption.

    Best wishes

    Gary

    Post edited by: Gary Graham

  • High using the Virtuakbox virtual machine CPU load

    I have a MacBook Pro 2.5 GHz processor dual-core Intel Core i5 (Turbo Boost up to 3.1 GHz) with 3 MB L3 cache.

    I use VMware Fusion Pro 7.0.1 (2235595) to install and use Windows 8 to run the Simulator on my MacBook internetworking solution.

    These simulators come with Oracle VM VirtualBox and they used to run in the Windows virtual machine on my MacBook Pro with no problems.

    So far, no problem!

    However, after upgrading to OS X Yosmite 10.10.1 simulators no longer work due to the high utilization of the processor by VBoxHeadless.exe

    I deleted the virtual machine, I uninstalled VMware Fusion and reinstalled, then reinstalled the Windows operating system, makes all the possible update and finally installed solutions simulators without success to have any Simulator market, they used. The reason is always high utilization of the processor by VBoxHeadless.exe as seen in the following images:

    2014-11-20_11-38-20.png2014-11-20_11-39-34.png

    I then moved to bootcamp and all do it again. Now simulators all run as before but on windows on the bootcamp partition.  Now I can even run a lot of devices in simulators like in the following photos with no problems:

    20141119_194835.jpg

    20141119_194908.jpg

    Please, do you know how I can fix this problem? I prefer to use VMware fusion because it gives me more room for what I want to achieve compared to the bootcamp.

    There is a difference between the hosts with regard to the information provided by windows on the ability of the virtualization of the MacBook.

    I thank very you much in advance!

    Kind regards

    Hello

    sramdane wrote:

    I am really a bit disappointed by the support of this forum.

    Please note that the forum is not taken in formal charge and that this is a forum where users help other users.

    Your problem unfortunately was not something that other people recognized as a problem they had seen or they simply didn't have the time to answer.

    The majority of the Fusion users will not also have installed VirtualBox.

    However, I am glad to hear that you found a solution (in a restaurant) and please let us know.

    --

    Wil

  • Two POS devices back up the same virtual machines

    Hello

    So my goal is to have weekly backups to a unit of POS, and monthly backups go to another device of POS, on a separate data store.  I'm stuck on VDP v5.1.1 because since I can not upgrade my vCenters yet.  What I'm trying to do, is to protect my backups to a store of data or corruption VDP OS - which have already occurred in the past.

    My question is, I'll be able to simply deploy a 2nd POS device and tell it to save the same virtual machines that are already underway by the 1st POS device?  I wonder what that complications would be, i.e. If the 2nd POS only runs a monthly backup, it will be smart enough get all changes since the last monthly backup or he'll come CBT files since the last daily backup of the 1st POS device?

    Thank you!

    You can make backups in this way with 2 POS devices and CBT is treated properly. Monthly backup will take more time because of the higher number of changes. Look at the time of the departure of the backups, they will take parallel steps.

    Paul

  • Problem fixing existing vmdk to the new virtual machine

    Hello

    Just upgraded to Fusion 6 and things seem ok (except for the dreaded license bug, but it's ok now).

    I'm having a problem attaching a form of virtual disk existing a VM - operation - existing. Basically, all the vmdk files are grayed out and non-eligible, regardless of the option to share / copy / to support the disc.

    I checked that I have full R/W access to the fie (and it works fine if I fire up the other virtual machine).

    Any idea?

    Well I ended up adding another player new to the virtual machine, I wanted to and the drafting of the description of disk (Disk DescriptorFile #)

    in the manually - vmdk file works great...

  • Not configured VM prevents the recovery of the configured virtual machines?

    I try to test a recovery plan. There is a problem with one of the virtual machines on the data store. It has a disc attached to a volume that is not replicated.

    When I run the plan, it logs the error below for all of the active steps after "Suspend Non-critical VMs on recovery Site".

    Error - virtual "name VM" machines that
    are not protected by this protection group use warehouses "Volume name"
    that are in this protection group.

    I understand that I can't protect this particular virtual machine. The question is if it will prevent the other virtual machines being restored or is something else going on?

    If you have a data store in a protection group then * everything * VMs on this data store must be configured for protection, otherwise it will fail.

    That is, if it is not protected SRM is not aware, cannot stop it and which will not be able to remove the data store, etc.

Maybe you are looking for