Re: vmware server reboot and in process image updates
Hi Gary, Prior to the reboot, the VCL site should go into maintenance mode to prevent a user from creating a reservation while this reboot procedure happens. You'll want to write a script that enumerates the VMs running on the VM host prior to shutdown, then bring those VMs back up after reboot. Once the VM host is back online, bring VCL out of maintenance mode. I know this would be pretty straightforward with ESXi, but I'm not sure about VMWare Server 2. I just did a quick Google search and found this: http://www.cyberciti.biz/tips/start-stop-vmware-virtualization-vms-command.html Regards, Mike -- *Mike Haudenschild* Education Systems Manager Longsight Group (740) 599-5005 x809 m...@longsight.com www.longsight.com On Thu, Jun 14, 2012 at 6:22 PM, Gary Trail tra...@rpi.edu wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Thanks Mike. you are correct - rebooting during pulling image would be avoided if at all possible. To elaborate a bit. Our objective is to run an unattended regular (weekly/monthly) vmware server restart/reboot without impacting the integrity of VCL. Of course, users would be disconnected when their guest vm shuts down but, upon restart vmware server (version 2) would restart any previously running vm's and users could just re-connect using the same IP address, user name etc and the VCL infrastructure would resume in its pre-reboot state and reservations would time out s if nothing happened. In our pilot project, the VCL management node and vmware server2 services are running on the same physical hardware while the VCL database and VCL front-end components are on two different hardware hosts. VM images currently reside on DAS disks on the management/vmware node hardware. Is this all possible/reasonable? How are others accomplishing this? Thanks Gary Mike Haudenschild wrote the following on 6/14/2012 5:02 PM: Hi Gary, I'm assuming you do NOT mean shutting down the VM host while VCL is actively pulling an image... That would surely be bad. I think that the reservation should still remain active in the database. Shut down the client image cleanly, then bring down the management node and front end cleanly, and reboot the VM host. Bring up the management node and front end. You'll have to manually power on your image VM on via the vSphere Client, but as long as the reservation hasn't expired it should be okay. Should be the same credentials to login to the image VM, too. Regards, Mike On Thursday, June 14, 2012, Gary Trail wrote: Quick question - We're in the middle of a pilot VCL installation (3 machines - front-end, database and management node (VCL management and vmware server on one management node). What will happen to any in process image updates if the vmware server reboots? Will the image updates survive the reboot? Thanks Gary Gary Trail Rensselaer Hartford Graduate Center 860-548-7869 tra...@rpi.edu javascript:; - -- Gary Trail Director of Network Services Rensselaer Hartford Graduate Center 860-548-7869 tra...@rpi.edu -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.9 (MingW32) iQEcBAEBAgAGBQJP2mQyAAoJENG9onuHOfZFsEMIAKbQV0w9qxLraoiFkl6gIc0F sGmZ5a0QaZIXEUyryrlXgxuvJkp+/4XEHDNi8UVXAFam0gNo8jlbEKcTh84Ix6uk 9p+Llmw2RTiQtqGXGF/cufkIDnw8lZB4KaU8zjjfdNnAcW6x1YmLOlfzMBVbxxzH gC9uU5UR0fZvwTxDM2HKwNsmg3Zugu7VhbxqXgDyFBNwS2AR2175pZT0PFP+wEAw 1bNM77/8TCM/fsmHfAqejgEFWxqmwlxK3M+nf/vRCPc2KhWL9/mkTF4zKMy06iY8 MHXylps30IGXHdH7nwsX/pZoc3NGxXN2vHcmhDsKXYASnYG2ofaPBxL9t4SSh30= =+CP6 -END PGP SIGNATURE-
Re: Image capture succeeds, then fails
Thanks Mike. Checking the storage profile for this VM host, Data Store Path is /vmfs/volumes/T2_6045_dev_VCL_ds01. That directory is present and read/write on our ESXi machine. VM Path is set to (empty). That's what we want if we plan to store our VM disks alongside our other VM data, right? VM Disk is set to localdisk, in case that matters. Repository Path is an NFS volume. We want that to be read-only on the hypervisor, read-write on the management node, but for the moment due to a configuration oversight it's read-write everywhere. I don't know if that's significant for the present case or not. On Fri, Jun 15, 2012 at 07:13:42PM -0400, Mike Haudenschild wrote: Hi Michael, It looks like the problem starts here: |32473|14:14|reload| 2012-06-15 15:31:21|32473|14:14|reload|VMware.pm:load(395)|failed to prepare vmx file for vcl-linux-template-2-bak on VM host: vcl0001 This is the second warning message in the log snippet. The first warning can be ignored. This is VCL writing the VMX file used to tell VMWare about the VM during the reload process. After the image is collected, it brings it up on that computer in read only mode. That process appears to have failed here. Check that the VM path is configured correctly for the VM host and that the directory is writable by te VM host. I'm writing from an airport on an ipad, so I can't fully review your log easily. Regards, Mike On Friday, June 15, 2012, Michael Jinks wrote: I think I may have just run my first successful image capture. There's an image directory now in my templates directory, and when I browse to Manage Images and Edit Image Profiles, my base image is there. Hooray! ...But, shortly after the capture completed, our admin address received the message below. I'm forwarding the whole thing because reading it myself I don't see anything that looks like a statement of a problem, except for the first reservation failed line. Any idea what went wrong? Or, maybe nothing went wrong and this is a spurious error? Where would I look for clues? Thanks, --Michael - Forwarded message from [1]r...@pilot-lab.uchicago.edu - reservation failed on vcl-linux-template-2-bak: process failed after trying to load or make available time: 2012-06-15 15:31:21 caller: State.pm:reservation_failed(213) ( 0) State.pm, reservation_failed (line: 213) (-1) [2]new.pm, process (line: 298) (-2) vcld, make_new_child (line: 568) (-3) vcld, main (line: 346) management node: [3]pilot-lab.uchicago.edu reservation PID: 32473 parent vcld PID: 2506 request ID: 14 reservation ID: 14 request state/laststate: reload/image request start time: 2012-06-15 15:31:12 request end time: 2012-06-15 16:01:12 for imaging: no log ID: none computer: vcl-linux-template-2-bak computer id: 7 computer type: virtualmachine computer eth0 MAC address: undefined computer eth1 MAC address: undefined computer private IP address: 10.50.84.15 computer public IP address: 10.50.84.15 computer in block allocation: no provisioning module: VCL::Module::Provisioning::VMware::VMware vm host: vcl0001 vm host ID: 1 vm host computer ID: 1 vm profile: VMware ESXi - IT Services vm profile VM path: /vmfs/volumes/T2_6045_dev_VCL_ds01 vm profile repository path: /vmfs/volumes/nfs-vcl-templates vm profile datastore path: /vmfs/volumes/T2_6045_dev_VCL_ds01 vm profile disk type: localdisk image: vmwarelinux-Linuxtemplate2v218-v0 image display name: Linux template 2v2 image ID: 18 image revision ID: 16 image size: 30720 MB use Sysprep: yes root access: yes image owner ID: 1 image owner affiliation: Local image revision date created: 2012-06-15 15:29:23 image revision production: yes OS module: VCL::Module::OS::Linux user: vclreload user name: vcl reload user ID: 2 user affiliation: Local RECENT LOG ENTRIES FOR THIS PROCESS: 2012-06-15 15:31:19|32473|14:14|reload|VMware.pm:get_image_size_bytes(3841)|ret rieved the size of the image from the datastore on the VM host: 32,212,255,246 2012-06-15 15:31:19|32473|14:14|reload|VMware.pm:get_image_size_bytes(3851)|ima ge size retrieved from both the image repository and VM host datastore: |32473|14:14|reload| image repository: 2,397,636,084 |32473|14:14|reload| VM host
Re: Image capture succeeds, then fails
A little more information, maybe relevant, maybe not: I notice that VCL *is* getting to the point of creating a VM directory in that path. On the hypervisor: # pwd /vmfs/volumes/T2_6045_dev_VCL_ds01 # ls |grep 218 vmwarelinux-Linuxtemplate2v218-v0 That directory name appears to have been generated by VCL. In the directory we have two files: # ls vmwarelinux-Linuxtemplate2v218-v0-flat.vmdk vmwarelinux-Linuxtemplate2v218-v0.vmdk But, as Mike suggested, the vmx file isn't appearing. What might account for the fact that VCL can write one set of files but then fails when it tries to write something else? On Mon, Jun 18, 2012 at 12:58:25PM -0500, Michael Jinks wrote: Thanks Mike. Checking the storage profile for this VM host, Data Store Path is /vmfs/volumes/T2_6045_dev_VCL_ds01. That directory is present and read/write on our ESXi machine. VM Path is set to (empty). That's what we want if we plan to store our VM disks alongside our other VM data, right? VM Disk is set to localdisk, in case that matters. Repository Path is an NFS volume. We want that to be read-only on the hypervisor, read-write on the management node, but for the moment due to a configuration oversight it's read-write everywhere. I don't know if that's significant for the present case or not. On Fri, Jun 15, 2012 at 07:13:42PM -0400, Mike Haudenschild wrote: Hi Michael, It looks like the problem starts here: |32473|14:14|reload| 2012-06-15 15:31:21|32473|14:14|reload|VMware.pm:load(395)|failed to prepare vmx file for vcl-linux-template-2-bak on VM host: vcl0001 This is the second warning message in the log snippet. The first warning can be ignored. This is VCL writing the VMX file used to tell VMWare about the VM during the reload process. After the image is collected, it brings it up on that computer in read only mode. That process appears to have failed here. Check that the VM path is configured correctly for the VM host and that the directory is writable by te VM host. I'm writing from an airport on an ipad, so I can't fully review your log easily. Regards, Mike On Friday, June 15, 2012, Michael Jinks wrote: I think I may have just run my first successful image capture. There's an image directory now in my templates directory, and when I browse to Manage Images and Edit Image Profiles, my base image is there. Hooray! ...But, shortly after the capture completed, our admin address received the message below. I'm forwarding the whole thing because reading it myself I don't see anything that looks like a statement of a problem, except for the first reservation failed line. Any idea what went wrong? Or, maybe nothing went wrong and this is a spurious error? Where would I look for clues? Thanks, --Michael - Forwarded message from [1]r...@pilot-lab.uchicago.edu - reservation failed on vcl-linux-template-2-bak: process failed after trying to load or make available time: 2012-06-15 15:31:21 caller: State.pm:reservation_failed(213) ( 0) State.pm, reservation_failed (line: 213) (-1) [2]new.pm, process (line: 298) (-2) vcld, make_new_child (line: 568) (-3) vcld, main (line: 346) management node: [3]pilot-lab.uchicago.edu reservation PID: 32473 parent vcld PID: 2506 request ID: 14 reservation ID: 14 request state/laststate: reload/image request start time: 2012-06-15 15:31:12 request end time: 2012-06-15 16:01:12 for imaging: no log ID: none computer: vcl-linux-template-2-bak computer id: 7 computer type: virtualmachine computer eth0 MAC address: undefined computer eth1 MAC address: undefined computer private IP address: 10.50.84.15 computer public IP address: 10.50.84.15 computer in block allocation: no provisioning module: VCL::Module::Provisioning::VMware::VMware vm host: vcl0001 vm host ID: 1 vm host computer ID: 1 vm profile: VMware ESXi - IT Services vm profile VM path: /vmfs/volumes/T2_6045_dev_VCL_ds01 vm profile repository path: /vmfs/volumes/nfs-vcl-templates vm profile datastore path: /vmfs/volumes/T2_6045_dev_VCL_ds01 vm profile disk type: localdisk image: vmwarelinux-Linuxtemplate2v218-v0 image display name: Linux template 2v2 image ID: 18 image revision ID: 16 image size: 30720 MB use Sysprep: yes root access: yes image owner ID: 1 image owner affiliation: Local image revision date
