Solved

VMWare Snapshot questions

Posted on 2009-06-30
14
805 Views
Last Modified: 2012-05-07
I am running a Novell Server on VMWare 3.5.  I took a snapshot when I originally set it up.  Recently i wanted to take another snapshot but I did not have enough free space to so I used the VI Client and went to snapshot manager and selected delete all snapshots.  

After I submitted the delete the VI client was disconnected.  Since the Novell server was still running I let i run overnight and tried to connect in the morning.  I still could not connect and since the Novel server was still running I waited until evening to start looking into it again.  

Tonight I am able to connect using the VI client but when I look at the storage there is still no available space.   When i deleted the snapshot shouldn't that have freed up the space?

Isn't a snapshot just a backup the server state at a specific time?  Does deleting a snapshot have any effect on the server itself?
0
Comment
Question by:qvfps
  • 7
  • 4
  • 2
  • +1
14 Comments
 
LVL 90

Expert Comment

by:John Hurst
ID: 24751069
I believe that a snapshot is a copy of the VM at a point in time, but that is from the perspective of VMware Workstation where I know that to be true. Did you just delete the file? Or did you use VMware to remove (delete) the snapshot?  ... Thinkpads_User
0
 

Author Comment

by:qvfps
ID: 24751076
I used the VI Client to delete the snapshot.  Inventory->snapshot manager and selected delete all.
0
 
LVL 90

Expert Comment

by:John Hurst
ID: 24751085
I wonder if your host machine still has the deleted snapshot in shadow storage? Otherwise I am not sure. ... Thinkpads_User
0
 
LVL 24

Expert Comment

by:ryder0707
ID: 24751103
By right you should not have that problem, esx should be able to know how to delete snapshots especially if you are using VIC.
Have you checked the vm folder in datastore, can you see any delta files?
0
 

Author Comment

by:qvfps
ID: 24751303
There is a server-flat.vmdk file which is larger the server itself and a small server.vmdk file.
0
 
LVL 24

Expert Comment

by:ryder0707
ID: 24751552
that is normal, server.vmdk is the disk descriptor for the flat disk file
how about folder for other vms? no snapshot files as well
0
 

Author Comment

by:qvfps
ID: 24751661
I am looking in vmfs/volumes/storage/server and the only large file there is the vmdk file.  The second server I have running on this machine is a small workstation and there are no large files in the corresponding folder.  

Is there another folder i should be looking in?  

When I use he VI Client and check storage under under the configuration tab it shows that I have used approximately 2 times the disk size for server 1 plus the disk size of server 2.  

The amount of free space did not change after i deleted snapshots.
0
IT, Stop Being Called Into Every Meeting

Highfive is so simple that setting up every meeting room takes just minutes and every employee will be able to start or join a call from any room with ease. Never be called into a meeting just to get it started again. This is how video conferencing should work!

 
LVL 24

Expert Comment

by:ryder0707
ID: 24751831
hmmm what about log files? try check all log files under /var/log
0
 

Author Comment

by:qvfps
ID: 24754829
I found this in the hostd-6.log.   Is there anything specific I should be looking for?

[2009-06-30 03:29:51.972 'TaskManager' 74410928 info] Task Created : haTask-80-vim.VirtualMachine.removeAllSnapshots-21
[2009-06-30 03:29:51.982 'vm:/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC.vmx' 74410928 info] State Transition (VM_STATE_ON -> VM_STATE_REMOVEALL_SNAPSHOT)
[2009-06-30 03:29:52.019 'vm:/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC.vmx' 74410928 info] Deleting snapshot: before PO
[2009-06-30 03:29:52.079 'BaseLibs' 48909232 info] DISKLIB-VMFS : "/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC-000003-delta.vmdk" : open successful (21) size = 268435456000, hd = 0. Type 8
[2009-06-30 03:29:52.080 'BaseLibs' 48909232 info] DISKLIB-VMFS : "/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC-000003-delta.vmdk" : closed.
[2009-06-30 03:29:52.277 'BaseLibs' 48909232 info] DISKLIB-VMFS : "/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC-flat.vmdk" : open successful (21) size = 268435456000, hd = 0. Type 3
[2009-06-30 03:29:52.277 'BaseLibs' 48909232 info] DISKLIB-VMFS : "/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC-flat.vmdk" : closed.
[2009-06-30 03:29:52.373 'BaseLibs' 48909232 info] DISKLIB-VMFS : "/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC-000002-delta.vmdk" : open successful (21) size = 268435456000, hd = 0. Type 8
[2009-06-30 03:29:52.374 'BaseLibs' 48909232 info] DISKLIB-VMFS : "/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC-000002-delta.vmdk" : closed.
[2009-06-30 03:44:38.473 'BaseLibs' 48909232 info] Disconnect check in progress: /vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC.vmx
[2009-06-30 03:44:38.474 'vm:/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC.vmx' 102611888 info] Question info: The VMware Tools poweron script did not run successfully in this virtual machine. If you have configured a custom poweron script in this virtual machine, please inspect it for errors; otherwise, please submit a support request.
, Id: 0 : Type : 2, Default: 0, Number of options: 1
[2009-06-30 03:44:38.474 'BaseLibs' 48909232 info] Disconnect check in progress: /vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC.vmx
[2009-06-30 12:44:48.034 'vm:/vmfs/volumes/47d1e040-3498377e-60f4-001f29c4a22c/DDC/DDC.vmx' 74410928 info] State Transition (VM_STATE_REMOVEALL_SNAPSHOT -> VM_STATE_ON)
[2009-06-30 12:44:48.034 'TaskManager' 74410928 info] Task Completed : haTask-80-vim.VirtualMachine.removeAllSnapshots-21
0
 
LVL 24

Expert Comment

by:ryder0707
ID: 24759893
well from the log all snapshots have been removed successfully
so your free space problem must be caused by something else
what about the size of all log files? check the log file size for each vm as well.
0
 

Author Comment

by:qvfps
ID: 24760042
I am missing approximaely 250 Gig.   I checked both VMs and there are no folders or logs anywhere near this size.
0
 

Author Comment

by:qvfps
ID: 24760043
Is there any particular folder I should be looking in?
0
 

Author Comment

by:qvfps
ID: 24760218
I just connected through the VI Client and suddenly I the 250 gig of space is suddenly free.  It was not there the last time I connected but it is now.   I have not done anything but review the logs since the last time I checked it.
0
 
LVL 10

Accepted Solution

by:
dnilson earned 250 total points
ID: 24768516
VI is disconnected a bit from the real - time state of VMware.

If the snapshot is large, the commit time is long, if its longer than the VI timeout the disconencts begin (italisized VM names).  The commit of the VM is still in process which can be verified from a putty session using hte commandline tools.

You are lucky - you waited, and it all came back.  I've seen folks reboot the host at the point you were at and end up with hash.

The BEST way to roll back a snapshot is from the commandline so you can actually SEE when the rollback is DONE, no just that VI gave up on it.

The VERY BEST way to prevent this is to rememeber that sanpshots are NOT backup, the y ARE NOT, "a copy of the VM in time."

A snapshot is where the host is witing changes to the VHD right now.  After taking a snapshot, the bse disk (the flatfile) is frosen to read only and checksumemd to detect change (changes to the base disk are very bad aaif a snapshot exists and will render the VM unbootable).

All writes are done to the snapshot file which grows with time.

All reads are done from the base disk, and all of the snapshots so the most recent data is available.  Obviously this is very slow.

So the answer to your question is -
1 )the disk space wasnt freed until the delete (commit) of the snapshot was completed.
2) the commit wasnt done for a long time after I gave up on it (ive seen 8hours or more when snapshots were 30 days old)
3) the freeing of the idsk space wasnt even begun until after the cmmit was complete.
4) VI could not help you know when these events occurred, becasue by design, if timeouts on commits after a certain time,
5) It times out after such a short time because , operationally, one should never plan to leave a snapshot in place for more than a few hours on a busy server.

0

Featured Post

Maximize Your Threat Intelligence Reporting

Reporting is one of the most important and least talked about aspects of a world-class threat intelligence program. Here’s how to do it right.

Join & Write a Comment

When we have a dead host and we lose all connections to the ESXi, and we need to find a way to move all VMs from that dead ESXi host.
David Varnum recently wrote up his impressions of PRTG, based on a presentation by my colleague Christian at Tech Field Day at VMworld in Barcelona. Thanks David, for your detailed and honest evaluation!
Teach the user how to install ESXi 5.5 and configure the management network System Requirements: ESXi Installation:  Management Network Configuration: Management Network Testing:
Advanced tutorial on how to run the esxtop command to capture a batch file in csv format in order to export the file and use it for performance analysis. He demonstrates how to download the file using a vSphere web client (or vSphere client) and exp…

762 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question

Need Help in Real-Time?

Connect with top rated Experts

23 Experts available now in Live!

Get 1:1 Help Now