No, there's nothing in the .vmdk file that references the .vmxf file.

Kelly


>>> Aaron Peeler <aaron_pee...@ncsu.edu> 10/07/2009 11:44 AM >>>

We're not run into this yet. From what we've experienced on our end this 
file is getting created when registering the the vm.

Is there any mention of this vmxf file in your <image_name>.vmdk file?

Aaron

--On October 6, 2009 1:25:18 PM -0400 Kelly Robinson 
<isg...@langate.gsu.edu> wrote:

>
> after doing some more testing it appears that the hypervisor also wants
> the .vmxf file to be located in the "[VCL]/inuse/<vm computer name>" file
> in order to register the machine.  This file is not copied over in the
> vcl code.  Is there a work around?
>
> Kelly
>
>
>
> Kelly P. Robinson
> Software Systems Engineer, Lead
> Advanced Campus Services
> Information Systems and Technology
> P.O. Box 3994
> Atlanta, Ga 30302-3994
> (404) 413-4757
> krobinso...@gsu.edu 
>
>
>
>>>> "Kelly Robinson" <isg...@langate.gsu.edu> 10/06/2009 11:49 AM >>>
> Thanks Josh!  Your suggestions worked!
>
> However, I have a new issue.  Once the reservation is scheduled, the
> request stays in pending.  I checked the vcld.log and it appears that the
> vm is not being registered on the hypervisor correctly.  Its giving a
> "virtual machine not found" message. When I look on the hypervisor that
> the vm is trying to run on, it has a vm registered named
> "Unknown(invalid)", not the one that's expected (vg1-1 in this case).
> However, when I run the command to register the vm straight from the
> command line, it works.  Any ideas on what the issue could be?  An
> excerpt from the log is listed below.
>
> (Note: the ip addresses and username/passwords have been "XXX" out.)
> ==========================================
> .
> .
> .
> 2009-10-06
> 11:36:55|6919|124:187|new|utils.pm:run_scp_command(6896)|attempt 1/3:
> executing SCP command: /usr/bin/scp -B -i /etc/vcl/winxp_blade.key -P 22
> -p -r /tmp/vg1-1.vmx
> XXX.XX.XX.XXX:/install/vmware_images/inuse/vg1-1/vmwareinxp-base7-v1.vmx
> 2>&1
> 2009-10-06 11:36:55|6919|124:187|new|utils.pm:run_scp_command(6942)|scp
> successful: attempt 1/3, exit status: 0, output: none
> 2009-10-06 11:36:55|6919|124:187|new|esx.pm:load(336)|Register Command:
> /usr/lib/vmware-viperl/apps/vm/vmregister.pl --server 'VCLXXX.gsu.edu'
> --username vcl --password 'XXXXXXX' --vmxpath
> '[VCL]/inuse/vg1-1/vmwareinxp-base7-v1.vmx' --operation register --vmname
> vg1-1 --pool Resources --hostname 'VCLXXX.gsu.edu' --datacenter
> 'ha-datacenter'
> 2009-10-06 11:36:57|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:36:57
> 2009-10-06 11:37:02|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:02
> 2009-10-06 11:37:07|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:07
> 2009-10-06 11:37:12|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:12
> 2009-10-06 11:37:17|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:17
> 2009-10-06 11:37:22|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:22
> 2009-10-06 11:37:27|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:27
> 2009-10-06 11:37:32|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:32
> 2009-10-06 11:37:37|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:37
> 2009-10-06 11:37:39|6919|124:187|new|esx.pm:load(339)|Registered:
>| 6919|124:187|new| Register of VM 'vg1-1' successfully completed under
>| host VCLXXX.gsu.edu
> 2009-10-06 11:37:39|6919|124:187|new|esx.pm:load(348)|Power on command:
> /usr/lib/vmware-viperl/apps/vm/vmcontrol.pl --server 'VCLXXX.gsu.edu'
> --vmname vg1-1 --operation poweron --username vcl --password 'XXXXXX'
> 2009-10-06 11:37:41|6919|124:187|new|esx.pm:load(351)|Powered on: Virtual
> Machine vg1-1 not found.
> 2009-10-06 11:37:43|32759|vcld:main(166)|lastcheckin time updated for
> management node 3: 2009-10-06 11:37:42
> /usr/lib/sendmail: option requires an argument -- f
> 2009-10-06 11:37:43|6919|124:187|new|utils.pm:mail(1301)|SUCCESS --
> Sending mail To: , PROBLEM -- esx.pm
>
>| 6919|124:187|new| ---- CRITICAL ----
>| 6919|124:187|new| 2009-10-06
>| 11:37:43|6919|124:187|new|esx.pm:load(359)|Could not query for VM in VI
>| PERL API 6919|124:187|new| ( 0) utils.pm, notify (line: 691)
>| 6919|124:187|new| (-1) esx.pm, load (line: 359)
>| 6919|124:187|new| (-2) new.pm, reload_image (line: 664)
>| 6919|124:187|new| (-3) new.pm, process (line: 266)
>| 6919|124:187|new| (-4) vcld, make_new_child (line: 600)
>| 6919|124:187|new| (-5) vcld, main (line: 347)
>
>
>| 6919|124:187|new| ---- WARNING ----
>| 6919|124:187|new| 2009-10-06
>| 11:37:43|6919|124:187|new|new.pm:reload_image(669)|vmwareinxp-base7-v1
>| failed to load on vg1-1, returning 6919|124:187|new| ( 0) utils.pm,
>| notify (line: 691)
>| 6919|124:187|new| (-1) new.pm, reload_image (line: 669)
>| 6919|124:187|new| (-2) new.pm, process (line: 266)
>| 6919|124:187|new| (-3) vcld, make_new_child (line: 600)
>| 6919|124:187|new| (-4) vcld, main (line: 347)
>
> 2009-10-06
> 11:37:43|6919|124:187|new|utils.pm:insertloadlog(5324)|inserted
> computer=14, loadimagefailed, vmwareinxp-base7-v1 failed to load on vg1-1
>
>
> .
> .
> ==================================================
>
>
> Kelly
>
>>>> Josh Thompson <josh_thomp...@ncsu.edu> 09/28/2009 4:06 PM >>>
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> (Added vcl-dev since there is a similar thread on that list.)
>
> Kelly,
>
> It sounds like you have 2 problems.  The first has to do with something
> keeping the scheduling algorithm from selecting a computer to run the
> image
> on (this is the isAvailable function in utils.php).  The second is that
> when
> you are manually inserting entries into the request and reservation
> tables,
> you are inserting an imaging reservation.  An imaging reservation tells
> vcld
> to capture the image currently on the computer.  Effectively, you're
> telling
> vcld to capture an image, but telling it to use the same name as an
> existing
> image.
>
> I'll try to enumerate what you should check (I'll leave out the specifics
> for
> the parts that have already been mentioned).
>
> - -image is in an image group
> - -image group is mapped to a computer group
> - -computer group contains computers in the "available" state
> - -you have an active management node (it is updating lastcheckin in the
> management node table)
> - -management node is in a management node group
> - -management node group maps to a computer group containing specified
> computer
> from above
> - -schedule for computer is available during the selected time
> - -platform for image matches platform of computer
> - -privileges set up in Privileges section of site to allow you to access
> image
> and computer
> - -specs of image (min RAM, min CPU speed, etc) are at or less than specs
> of
> computer
> - -no other entries in the reservation table for the computer that
> overlap with
> the selected time period
>
> Let us know if you are still having trouble after checking the above list.
>
> Josh Thompson
> VCL Developer
>
> On Monday September 28, 2009, Kelly Robinson wrote:
>> ..or maybe a better question would be:
>>
>> After an image is created, how do you get make it available for
>> requesting? I followed these directions to create a base Windows XP
>>  image and created the entry in the vcl database.
>> http://cwiki.apache.org/VCL/creating-a-windows-xp-base-image-on-vmware.h 
>> tml
>>
>> As stated previously, the vcld.log gives the critical error that the
>> image already exists in the repository and then it goes on to say image
>> creation failed and begins to exit the reservation process.
>>
>> Is there something else that I'm supposed to do?
>>
>> >>> "Kelly Robinson" <isg...@langate.gsu.edu> 09/28/2009 9:15 AM >>>
>>
>> I am testing a base Windows XP image (no apps) and have its image
>> requirements set for the following:
>>
>> minram: 512
>> minprocnumber: 1
>> minprocspeed: 1024
>> minnetwork: 100
>>
>> I've also tried to make sure the image is grouped correctly and mapped to
>> the correct server, however it still gives the error.
>>
>> Does the message in the vcld.log of "image already exists in the
>> repository" after I submit the sql statements to manually place a request
>> an image have anything to do with the error?
>>
>> Thanks,
>> Kelly
>>
>> >>> Nathan Dawson <ndaw...@email.wcu.edu> 09/28/2009 8:57 AM >>>
>>
>> Kelly,
>> One other thing that it may be is if you set the image requirements too
>> high (min CPU, min ram etc), if your servers or VMs aren't set high
>> enough to meet those criteria it can also give you that message.
>>
>> Nathan Dawson, MCP, MCDST, ACHDS
>> Director of Networking and Information Systems
>> The Kimmel School
>> Western Carolina University
>> 828.227.2748
>>
>>
>>
>> From:Kelly Robinson [mailto:isg...@langate.gsu.edu] 
>> Sent: Monday, September 28, 2009 8:50 AM
>> To: vcl-u...@incubator.apache.org 
>> Subject: VCL Reservation Problem
>>
>>
>>
>> How do you get VCL to allow an image to be requested?  I've added the
>> image to the database and would now like to request it through the web
>> interface, however when you click "New Reservation" and select the image
>> from the drop down menu, it gives a message that states "Selection Not
>> Currently Available".
>>
>> I've tried to manually insert a request into the database using the
>> following sql statements, however the vcld.log gives error messages
>> stating that the image already exists in the repository.  What am I
>> missing?
>>
>> =================================
>> INSERT INTO `vcl`.`request` (
>> `id` ,
>> `stateid` ,
>> `userid` ,
>> `laststateid` ,
>> `logid` ,
>> `forimaging` ,
>> `test` ,
>> `preload` ,
>> `start` ,
>> `end` ,
>> `daterequested` ,
>> `datemodified`
>> )
>> VALUES (
>> '1' , '16', '1', '16', '', '1', '0', '0', NOW( ) , TIMESTAMPADD(MINUTE,
>> 120, NOW()), NOW( ) , NULL );
>> ===============================================
>>
>> ===============================================
>> INSERT INTO `vcl`.`reservation` (
>> `id` ,
>> `requestid` ,
>> `computerid` ,
>> `imageid` ,
>> `imagerevisionid` ,
>> `managementnodeid` ,
>> `remoteIP` ,
>> `lastcheck` ,
>> `pw`
>> )
>> VALUES (
>> NULL , '1', '2', '7', '7', '1', NULL , NULL , NULL
>> );
>> ===========================================
>>
>>
>> Thanks,
>> Kelly
>>
>>
>>
>>
>>
>> Kelly P. Robinson
>>
>> Software Systems Engineer, Lead
>>
>> Advanced Campus Services
>>
>> Information Systems and Technology
>>
>> P.O. Box 3994
>> Atlanta, Ga 30302-3994
>> (404) 413-4757
>>
>> krobinso...@gsu.edu 
> - --
> - -------------------------------
> Josh Thompson
> Systems Programmer
> Advanced Computing | VCL Developer
> North Carolina State University
>
> josh_thomp...@ncsu.edu 
> 919-515-5323
>
> my GPG/PGP key can be found at pgp.mit.edu
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.6 (GNU/Linux)
>
> iD8DBQFKwRc+V/LQcNdtPQMRAt2AAJ46y3d6w36ykrx26tjf7aiQI0prxgCffKUt
> W6qIWQeZ3FOTuXWukVW/It4=
> =c+W1
> -----END PGP SIGNATURE-----



Aaron Peeler
OIT Advanced Computing
College of Engineering-NCSU
919.513.4571
http://vcl.ncsu.edu

Reply via email to