What is the use of Consolidating snapshots ?

[adrianych]

I have been reading up but I still do not really understand what is the use of this function.

Can someone kindly enlighten me ?

Also kindly correct if my train of thoughs is wrong :

i. Snapshots are not really backups

ii. Snapshots are supposed to be taken so that the VM can be recovered to that point in time (eg. before Windows Update)

iii. Snapshots are "integrated" with the VM's OSe (via VMware Tools), which means that we can perform snapshooting when the VMs are up.

- Windows 2008 ?

- Microsoft SQL ?

iv. Snapshots would take up space outside of the VMs

- The datastore should be much larger then the VMs (eg if VM is 200GB, datastore shd be 300GB to hold VMDKs + Snapshots)

- Snapshot sizes would grow as time passes (eg. snapshot taken 100 days ago may grow from 10GB to 20GB depending on changes made to VM)

v. Recommendations (per VM)

- Not to keep snapshots for extend periods of time

- Not to keep too many snapshots

- "consolidate" snapshots

[AWo]

adrianych schrieb:

I have been reading up but I still do not really understand what is the use of this function.

Can someone kindly enlighten me ?

Also kindly correct if my train of thoughs is wrong :

i. Snapshots are not really backups

Yes, they aren't. They only contain the changes since the snapshot was created and not all data.

ii. Snapshots are supposed to be taken so that the VM can be recovered to that point in time (eg. before Windows Update)

Yes.

iii. Snapshots are "integrated" with the VM's OSe (via VMware Tools), which means that we can perform snapshooting when the VMs are up.

- Windows 2008 ?

- Microsoft SQL ?

Snapshots can be taken without VMware tools and they are on the "virtual" hardware layer. Each operating system can be snapshotted while it is on or off. VMware Tools can help to create application consistent snapshots as they can trigger OS specific mechanisms like Windows Volume Shadow Servises.

iv. Snapshots would take up space outside of the VMs

- The datastore should be much larger then the VMs (eg if VM is 200GB, datastore shd be 300GB to hold VMDKs + Snapshots)

- Snapshot sizes would grow as time passes (eg. snapshot taken 100 days ago may grow from 10GB to 20GB depending on changes made to VM)

Yes. But AFAIK the snapshot can not grow beyond the size of the base vmdk. But I'm not sure on that, I will have a look if nobody else can answer that.

v. Recommendations (per VM)

- Not to keep snapshots for extend periods of time

- Not to keep too many snapshots

- "consolidate" snapshots

I agree, but sometimes requirements don't give you a choice.

AWo

[adrianych]

So Would it be recommended to have an offline or online snapshot of a Win2008 server running SQL ? Say for "pre-Windows Update" in case update then cannot bootup.

Is there anyway we can have like

- daily snapshot but later remove the odd dates ?

- Weekly snapshots but later remove week 2 & 4 ?

[AWo]

When S SQL is protected by VSS you can create an online snapshot, as VMware and VSS work together to keep the data consistent. AFAIK Exchange is using VSS, for example. I do not know it for MS SQL.

However, you can do online backups of MS SQL databases with the native Microsoft tool and schedule theses jobs. You'll get a consistent backup file of your database. That can be backed up by any tool while the server and the database is running.

What is the purpose of the daily and weekly snapshots? You can use the vCenter scheduler or Orchestrator to create such jobs.

AWo

[iw123]

The consolidate snapshots option commits all snapshots for a virtual machine to the original virtual disk, thereby deleting the snapshot vmdk files.

[adrianych]

Re: What is the use of Consolidating snapshots ?

The consolidate snapshots option commits all snapshots for a virtual machine to the original virtual disk, thereby deleting the snapshot vmdk files.

Does it mean that it reduces the "snapshot file" and turn it into a "pointer" so that it will "start to grow" ?

[ShirinKumar]

Hi ,

Please find the below Snapshot commitment and consolidation use,

• Base disk – 15GB
• Snapshot 1 – 1GB
• Snapshot 2 – 1GB
• Snapshot 3 – 1GB
• Snapshot 4 – 10GB

Snapshot 1 is copied in to Base disk, Snapshot 2 is copied in to Base disk, Snapshot 3 in to Base disk and Snapshot 4 in to your Base disk. After the copy of Snapshot 4 in to the Base disk all Snapshots will be deleted. Please note that the total amount of diskspace consumed before the “Delete All” was 28GB. Right before the final merge the consumed diskspace is still 28GB. Not only did VMware reduced the chances of running out of disk space, the time to commit the snapshot by using “delete all” has also been decreased using this new mechanism.

a snapshot isn’t a static situation like a clone is. A snapshot can best be compared to a redo log, although technically it isn’t because it’s just a bitmap of disk sectors that changed. When you create a snapshot you only create a small “differences” file (*.delta.vmdk) which will contain all the differences until you delete or revert. Please remember reverting(go to) doesn’t delete the differences file! And this file can grow very fast depending on how many changes are made on the disk.

[Kahonu84]

So what's the difference between "Consolidate" and "Delete All"??

[adrianych]

So what's the difference between "Consolidate" and "Delete All"??

I think you may have misconception of the snapshot....

I would use Windows update + service pack installation as an example.

VM with MS svr 2003 running....you may want to install SP1, then test, then install the Windows updates, then test.

- snapshot 1 (sn1) : Win2003 without updates & SP

- Installation of  Win2003 SP1 (with reboot etc)

- snapshot 2 (sn2) : Win 2003 SP1

- Install Windows updates on Win2003 SP1

- snapshot 3 (sn3) : Win 2003 SP1 with updates

- svr testing for 5 days (eg. install Acrobat reader, Microsoft office)

So u shd have 3 snapshots.....which all will "grow" as you test ur svr.

But ur VM will have 4 main file types (with regards to this context)

1. Base VM files (vmdk, vmx, VM's memory, swp files etc)

2. VM's snapshot file

3. VM's snapshot swp files (swap may not be correct name....but I cannot think of the proper name)

In logical terms....

- "sn1" is the base snapshot

- "sn2" is "sn1" + "data1"

- "sn3" is "sn1" + "data1" + data2"

- VM1 is sn1" + "data1" + data2" + un-named swap data

On the 5th day of testing....the VM will have files for

- "base VM", "sn1", " sn2", "sn3", "data1",  data2" and un-named swap data that is generated during the 5 days testing.

Then by right "consolidation" will convert the un-name swap into "data3" so that

- "sn1" is the base snapshot + "data3"

- "sn2" is "sn1" + "data1" + "data3"

- "sn3" is "sn1" + "data1" + data2" + "data3"

- VM1 is sn1" + "data1" + data2" + "data3"

In a layman way....VMs with snapshot are VMs that have have been split into many files.

When you delete snapshots (please do not try delete all), what you are doing is to merge sn1" + "data1" + data2" + un-named swap data or "data3" into a VM.

Sorry....I forgot to add...please do not house or keep VMware snapshots for extended periods of time.

Snapshots are not backups. They might snap then u have yourself shot......cheers....