ghettoVCB.sh - Free alternative for backing up VM's for ESX(i) 3.5, 4.x, 5.x, 6.x & 7.x

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

vlho · ‎09-16-2009

Hi winterg

your solution is works (only ESXi 4), thanks !

For ESXi 3.5 you usage:

kill `pidof crond` ; crond -b

vlho · ‎09-16-2009

Hi all,

next option for restart crond

kill -HUP `pidof crond`

i.e. without crond -b

I think that this is better and cleaner than the last...

winterg · ‎09-16-2009

Either works fine to kill the running process, I'm just glad it works. It has run two nights in a row without fail. I restart crond without the -b for two reasons:

- The -b is already the default behavior without the switch

- the original process that is started doesn't use it and I wanted it to mirror what VMware was doing at start-up.

Hopefully lamw will update the article text for the crontab time entry (to include the minute) and the restart of crond so that others don't suffer the same issues.

Now on to my next issue... Does anyone have compression working on ESXi 4.0 free? It fails every time I enable compression. I haven't looked into it yet as I needed to fix the bigger problems first and wondered if anyone else had it working.

scowse · ‎09-16-2009

I can confirm that compression works without problems in esxi 4.0.0 181792 but I do not use it because of the extra time it adds to the backup process. You don't say what the error is?

lamw · ‎09-16-2009

winterg, thanks for all the feedback and work you've put in, I'll update the document tonight if I get some time.

Regarding compression, it should work, I know few people out there that are using it including my colleague in UCSB What issues are you having? Can you elaborate?

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

winterg · ‎09-16-2009

So every time I try to backup I get the following. There is plenty of storage available according to df -k.

"################ Taking backup snapshot for hostname.temp-dom.local ... ################"

Start time: Thu Sep 17 03:36:16 UTC 2009

Destination disk format: VMFS thin-provisioned

Cloning disk '/vmfs/volumes/SAS-Datastore1/hostname.temp-dom.local/hostname.temp-dom.local.vmdk'...

Clone: 100% done.

Removing snapshot from hostname.temp-dom.local ...

Compressing VM backup "/vmfs/volumes/External-eSata1/host-backups/hostname.temp-dom.local/hostname.temp-dom.local-2009-09-17.gz"...

tar: cannot store file 'hostname.temp-dom.local-2009-09-17/hostname.temp-dom.local-flat.vmdk' of size 107389255680, aborting

End time: Thu Sep 17 03:49:09 UTC 2009

"#################### Completed backup for hostname.temp-dom.local! ####################"

Start time: Thu Sep 17 03:36:13 UTC 2009

End time: Thu Sep 17 03:49:09 UTC 2009

Duration : 12.93 Minutes

Completed backing up specified Virtual Machines!

I cleaned out the destination directory to make sure no other backups where in the system.

lamw · ‎09-16-2009

I would suggest in the future if you post output to wrap it in code tags, makes it much more readable

Regarding thin provisioning, take a look at my response back on Aug 6:

I think there has been quite a bit of confusion on Thin Provisioning, not only with this script which has supported this feature for awhile but also in the general forums.
Within the guestOS, it may show free/unsused space but it does not mean that blocks have not been allocated in the past that could have used up more space than it's showing. Remember, when you delete something, you're just not referencing it anymore, it's still there. Blocks within the OS needs to be zeroed out and this is explained in detail from Mr. 007 himself Duncan Epping at: http://www.yellow-bricks.com/2009/07/31/storage-vmotion-and-moving-to-a-thin-provisioned-disk/
Thin provisioning does work using vmkfstools but you need to ensure you've zeroed out any unused blocks before actually see this when trying to backup or even convert a VMDK.
Hopefully this clears things up

Now onto the compression, it looks like the compression is in fact starting, but it fails to store the backup. If you still have the backup, can you see if you can manually compress it? I've not seen this error before but it may be complaining about the size of the file, perhaps a limitation in the version of tar in ESX (would not surprise me). Secondly, are you backing this up to a VMFS volume? What is the blocksize you've set, remember that blocksize on a given volume will dictate how a big a file can be and you want to make sure you're not exceeding any of those limits.

I should be updating the documentation regarding the cron information but unfortunately our development system is down at moment for some new hardware upgrades and I want to confirm a few things before posting the changes, depending if it's up later tonight, the post may be postponed till tomorrow.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

kanngoo · ‎09-19-2009

hello,

i use esxi 4.0 and want to set up the cronejob correct. when i use the script manual it woks fine.

when i make a cronjob now an then add the necessary lines in the /etc/rc.local is this example correct?

~ # cat /etc/rc.local

#! /bin/ash

export PATH=/sbin:/bin

log() {

echo "$1"

logger init "$1"

}

#execute all service retgistered in /etc/rc.local.d

if http:// -d /etc/rc.local.d; then

for filename in `find /etc/rc.local.d/ | sort`

do

if && ; then

log "running $filename"

$filename

fi

done

fi

/bin/echo "* 0 * * 1-5 /vmfs/volumes/simplejack-local-storage/ghettoVCB.sh /vmfs/volumes/simplejack-local-storage/backuplist > /vmfs/volumes/simplejack-local-storage/ghettoVCB-backup-\$(date +
%s).log" >> /var/spool/cron/crontabs/root

/bin/busybox crond -b

kill `cat /var/run/crond.pid` ; crond

lamw · ‎09-19-2009

Please always use code tags to wrap your console output, makes reading much easier.

I haven't had time to update the document with the latest changes mentioned in the last few replies, I'm currently out of town.

Regarding the changes for cron to work, you need to first kill the cron daemon first ... then issue the updates and restart (without -b)

kill $(cat /var/run/crond.pid)
/bin/echo "* 0 * * 1-5 /vmfs/volumes/simplejack-local-storage/ghettoVCB.sh /vmfs/volumes/simplejack-local-storage/backuplist > /vmfs/volumes/simplejack-local-storage/ghettoVCB-backup-\$(date +
%s).log" >> /var/spool/cron/crontabs/root
/bin/busybox crond

You may also want to specify the full path to where kill is located it at, I believe it might be in /bin

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

chayawatv · ‎09-21-2009

I try to use this scipt to do backup with compression enabled.

Howerver, I cannot uncompress "filename".gz as it show "tar: Invalid tar magic".

Currently, I use ESX 3.5i U4 free edition and backup it to local drive of ESX server itself.

Any thought about the issue?

Thank you

lamw · ‎09-21-2009

Can you please provide the full command you're using to uncompress the backup?

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

chayawatv · ‎09-21-2009

1.) # gunzip < /vmfs/volumes/datastore1/backup/vm1/vm1-2009-09-22--1.gz | tar xvf - -C /vmfs/volumes/datastore1/restore

and

2.) # gunzip < /vmfs/volumes/nfs/backup/vm1/vm1-2009-09-22--1.gz | tar xvf - -C /vmfs/volumes/datastore

I try to backup and uncompress in 2 scenarios 1.) on local hard disk, 2.) on nfsshare

both give me "tar: Invalid tar magic"

Thank you

lamw · ‎09-22-2009

Perhaps tar changed between ESXi 3.5u4 and ESXi 4.0 .... I would need to replicate this in our dev environment and see what I get. I'll get back to you on this one, need to find some free cycles to test.

Thanks for reporting this.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

kanngoo · ‎09-22-2009

hi,

backup and cronjob with esxi 4.0 works great.

thanks

kanngoo

lamw · ‎09-22-2009

Thanks for the comments and glad it's working!

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

lamw · ‎09-22-2009

Interesting, it works for me on ESXi 3.5u4

~ # vmware -v
VMware ESX Server 3i 3.5.0 build-153875

~ # gunzip < /vmfs/volumes/datastore1/WILLIAM_BACKUPS/test/test-2009-09-22--1.gz | tar xvf - -C /vmfs/volumes/datastore1/
test-2009-09-22/
test-2009-09-22/test.vmx
test-2009-09-22/test-flat.vmdk
test-2009-09-22/test.vmdk

I executed this at the root of Busybox and it works perfectly fine for me.

What version of gunzip and tar is displaying for your version of ESXi?

~ # gunzip -v
gunzip: invalid option -- v
BusyBox v1.2.1 (2008.06.26-17:58+0000) multi-call binary

Usage: gunzip [OPTION]... FILE

Uncompress FILE (or standard input if FILE is '-').

Options:
        -c      Write output to standard output
        -f      Force read when source is a terminal
        -t      Test compressed file integrity

~ # tar -v
BusyBox v1.2.1 (2008.06.26-17:58+0000) multi-call binary

Usage: tar -[czxtvO] [-f TARFILE] [-C DIR] [FILE(s)] ...

Create, extract, or list files from a tar file.

Options:
        c               create
        x               extract
        t               list

Archive format selection:
        z               Filter the archive through gzip

File selection:
        f               name of TARFILE or "-" for stdin
        O               extract to stdout
        C               change to directory DIR before operation
        v               verbosely list files processed

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

Kabs123 · ‎09-23-2009

Hi William,

thank you for the nice script. There is just something wrong with my settings. Can you help me please?

I get the following message, what does it mean:

DEVEL_MODE=0

printUsage() {

SCRIPT_PATH=$(basename $0)

echo -e "\nUsage: ${SCRIPT_PATH} \n"

/vmfs/volumes/4a3bec10-5b0ef122-e3cb-001777642eb1/backupscript # ./ghettoVCB.sh vmbackups

(vim.fault.AlreadyExists) {

dynamicType = ,

faultCause = (vmodl.MethodFault) null,

name = "/vmfs/volumes/b22ca6b3-931cd49f",

msg = "The specified key, name, or identifier already exists.",

}

Snapshot found for xvm3_1, backup will not take place

lamw · ‎09-23-2009

It looks like the error is:

Snapshot found for xvm3_1, backup will not take place

Please take a look at FAQ #10

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

cookieme · ‎09-23-2009

Is it possible to remotely kick-off the script from an Ubuntu VM with vCLI installed instead of enabling SSH through the unsupported mode? The reason I'm asking is that with your ghettoVCBg2 script you are using the vMA which has vi-fastpass for authentication and so does not require SSH. So without vMA and fastpass is there a command in the vCLI that can run the script located on the local datastore or shared storage although it might prompt for credentials?

Thanks

chayawatv · ‎09-23-2009

Hi,

here is my VMWare Version

~ # vmware -v

VMware ESX Server 3i 3.5.0 build-153875

Here is my tar and gunzip version

~ # tar -v

BusyBox v1.2.1 (2008.06.26-17:58+0000) multi-call binary

Usage: tar -[czxtvO] ...

Create, extract, or list files from a tar file.

Options:

c create

x extract

t list

Archive format selection:

z Filter the archive through gzip

File selection:

f name of TARFILE or "-" for stdin

O extract to stdout

C change to directory DIR before operation

v verbosely list files processed

~ # gunzip -v

gunzip: invalid option -- v

BusyBox v1.2.1 (2008.06.26-17:58+0000) multi-call binary

Usage: gunzip ... FILE

Uncompress FILE (or standard input if FILE is '-').

Options:

-c Write output to standard output

-f Force read when source is a terminal

-t Test compressed file integrity

This is error what I get:

~ # gunzip < /vmfs/volumes/datastore2/backup/vm01/vm01-2009-09-24--1.gz | tar xvf - -C /vmfs/volumes/datastore2/

vm01-2009-09-24/

vm01-2009-09-24/vm01.vmx

vm01-2009-09-24/vm01-0-flat.vmdk

tar: Invalid tar magic

I set disk format as "VMFS Thick". Not sure whether it relates to the issue.

Thank you very much for you time on this matter.

lamw · ‎09-23-2009

Thought I share with everyone that ghettoVCB just hit 100,000+ views! Thanks to everyone for their contributions, comments and use of the script!

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "he

lamw · ‎09-23-2009

You're posting in the wrong forum, this is not ghettoVCBg2 forum and No you can not. The script is written specifically to work with vMA and vi-fastpass.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

lamw · ‎09-23-2009

Nope, the VMDK format does not matter with respect to extracting the contents. Curious, how big is the VM that you backed up and how big is the file after compression?

Not sure what to tell you, it works fine on our system ... the only thing I can think of is there might be a size limitation when trying to extract from within Busybox and it would not surprise me at all if this was a tar limitation within Busybox confines.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

chayawatv · ‎09-23-2009

tar size is 8 GB

backup size after compression is around 750 MB

lamw · ‎09-23-2009

Let me create a dummy VM that is 8GB using eagerzeroedthick and try backing it up with compression enabled and see if I run into the same issue. Post back shortly

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

chayawatv · ‎09-23-2009

Hi,

I just recreate dummy VM with 2 GB size do backup with compression. it is running smooth without a problem. So, I think the problem as you mentioned about tar file size limitation.

I will try to find server to put VMWare ESX 4 and try whether will give me the same issue. And, will let you know the result.

Thank you very much for your assistance.

Kabs123 · ‎09-24-2009

Unfortunately the backup does not work properly for me.

The only file that gets copied by backup process is the *.vmx file.

The second problem is, that there will no further backup take place when there already exists one *.vmx copy.

The error message i get:

Error: failed to lookup testvm!

It would be great if you could help me.

lamw · ‎09-24-2009

Please take a look at FAQ #9 and also edit the script and look for the variable DEVEL_MODE and change that from 0 to a 1 and provide the output and remember to wrap the output from the console using code tags to make it readable.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

Kabs123 · ‎09-24-2009

(vim.fault.DuplicateName) {
dynamicType = <unset>,
faultCause = (vmodl.MethodFault) null,
name = "nfs1-vol1",
object = 'vim.Datastore:nfs1.cad:/vol/vol1/data/esx-datastore',
msg = "The name 'nfs1-vol1' already exists.",
}
##########################################
Virtual Machine: testvm
VM_ID: 80
VMX_PATH: /vmfs/volumes/datastore2/testvm/testvm.vmx
VMX_DIR: /vmfs/volumes/datastore2/testvm
VMX_CONF: testvm/testvm.vmx
VMFS_VOLUME: datastore2
##########################################
Start time: Thu Sep 24 16:57:42 UTC 2009
End time: Thu Sep 24 16:57:43 UTC 2009
Duration : 1 Seconds
Completed backing up specified Virtual Machines!

ralfoertner · ‎09-24-2009

Excuse me if that isn't the right way to ask such a simple question but I did not find another way.

The only downloadable version of ghettovcb on this page is dated on 11/01/2008.

Is that truely the latest version esx-i 4?

I've been searching high and but could not find an newer one to download.

Once again I hope tah this was the right place to ask such a question

Sincerely

Ralf Oertner

DSTAVERT · ‎09-24-2009

The download link is for the latest version.

ralfoertner · ‎09-24-2009

Thank you for your quick reply.

The link is located before the known issues section.

Underneath the "Attachment" Section.

Is that the right link?

Sorry my question may sound stupid but I am truely confused about that.

Ralf Oertner

DSTAVERT · ‎09-24-2009

There is only one download link.

lamw · ‎09-24-2009

So nothing from the output is showing me there were any errors and if you look at the duration it was pretty quick to finish and I can see that all the params were validated. So, my question, is there an actual VMDK that is valid and attached to this VM?

Also when I mentioned code tags, there were specific ones for the Jive forums, take a look at this page: http://communities.vmware.com/markuphelpfull.jspa

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

Kabs123 · ‎09-25-2009

Problem cleared, i have deleted accidently a " in the source code and a part of the script has been commented out.

Thank you for support.

DSTAVERT · ‎09-25-2009

I think I have found an issue with huge notes in Annotations. I downloaded the ntop appliance and the output from vim-cmd vmsvc/getallvms generates a bunch of junk lines. I haven't been able to test the script and it may not affect anything. The output looks like:


"112";"ntop";"[storage]";"ntop/ntop.vmx"
"       VirtualAppliances.net NTOP Appliance for realtime network monitoring.";"";""
"       To monitor your network's traffic usage, simply start this appliance.";"";""
"       It will automatically begin collecting traffic statistics from your";"";""
"       network.  To view statistics, simply use your web browser to visit";"";""
"       the NTOP interface at http://ntop.local.:3000 or the IP Address given";"";""
"       on the console.  NTOP is excellent for collecting statistics for ";"";""
"       trouble shooting your network as well as longer range monitoring and";"";""
"       planning activities.  Visit http://ntop.org for more information about";"";""
"       the ntop application. Community support available at http://forums.virtualappliances.net ";"";""

lamw · ‎09-25-2009

Yes, annotations with the VM will cause the input to be captured when dumping out all VM(s), it'll pretty much fail on the lines where it contains the annotation since each line is suppose to represent a VM entry.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

DSTAVERT · ‎09-26-2009

Did finally get to test it and at least in my test environment it doesn't appear to affect anything. Screen output doesn't seem to be affected either. Sorry to have caused unnecessary concern.

gbras · ‎09-27-2009

Hi,

I need to do a small modification to the script, to skip both virtual and Physical RDMs (I need the VM backed up but not the RDMs).

I'm going to change this bit of code:

\{code\}

#support for vRDM and deny pRDM

grep "vmfsPassthroughRawDeviceMap" "$" > /dev/null 2>&1 \{code\} with this: \{code\} #support for vRDM and deny pRDM grep "RawDeviceMap" "$" > /dev/null 2>&1

\{code\}

Is this OK?

Thank you

Guido

lamw · ‎09-27-2009

You can do that OR you can just deny both types of RDMs: physical and virtual by doing the following:

grep -E '(vmfsPassthroughRawDeviceMap|RawDeviceMap)' "${SOURCE_VMDK}" > /dev/null 2>&1

This is an extended grep which allows you to use an OR type statement to look for multiple strings and in this case, you want to deny any disks that are pRDM or vRDM.

When you display code on the forums, you don't need to escape the "{" when wrapping your output using the code tags.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

chayawatv · ‎09-27-2009

run backup with this script on ESX 4.0.

Have no problme so far.

Thanks

ralfoertner · ‎09-28-2009

Dear GhettoVCBackup Experts,

First of all many thanks for the excellent script!

I have a few questions regarding ghettoVCB on ESX-I 3.5 and ESX-I 4.0 both used with the "free" license.

I use the script with the "plink-method" but I am having trouble with the log file. I am using the following commnandline to start the script:

c:\programme\putty\plink.exe root@192.168.110.3 -pw password "cd /vmfs/volumes/store-VMs/ && ./ghettoVCB.sh servers > ghettobackup-server.log &"

The log file is created as expected but sometimes lines are missing or the logging stops before the scriptrun has ended. For that reason I am often unsure if the script is still running or not.

Is there any way to make sure the script is finished or not? How can I fix the "logging-problem"? Is the plink-method the best way or is it better to use crontab (maybe logging works better that way)? When I look at the backup-destination during the script is runnning the file sizes are not changing. They are set to the file size of the original vmdk as soon as the script is running. What happens if the script ist failing in the middle of the copy process. Do I end with a copy of a vmdk which is just "halffull" and the other half is filled probably with zeros.

A lot of questions I hope you can help and many thanks once again!

davidpyper · ‎09-28-2009

Hi William,

My client has 3 ESXi machines hosting 19 virtual servers, all Windows XP or 2003 machines. The ghettoVCB.sh script runs on the 3 ESXi machines and backs-up the 19 servers to one large NFS share.

The script regularly hangs while removing snapshots. The symptom of the problem is that the last line of the logfile reads "Removing snapshot from " (vnname) followed by no further activity. The manual workaround involves using vSphere to create then remove a snapshot, at which point the script continues working its charm.

The problem apparently stems from the following codeblock from ghettoVCB.sh (line numbers added for clarity):

385 #powered on VMs only w/snapshots

386 if [[ ! $ -eq 1 ]] && [; then

387 $ vmsvc/snapshot.remove $ > /dev/null 2>&1 388 389 #do not continue until all snapshots have been committed 390 echo "Removing snapshot from $ ..." 391 while ls "$" | grep -q delta; 392 do 393 sleep 3 394 done 395 fi As long as there are delta files in the VM's directory, lines 392 to 394 loop perpetually. My contribution to your excellent script is to suggest a modification to line 387 to use vmsvc's snapshot.removeall command instead of snashot.remove. Line 387 would then look like this: 387 $ vmsvc/snapshot.removeall $ > /dev/null 2>&1

This enhancement removes all snapshots in the form of delta files, and allows the script to continue its work.

Cheers!

Dave

lamw · ‎09-28-2009

Not sure about the plink solution, I don't use that personally with ghettoVCB, I know it works great for others. You also have another logging possiblity as you've mentioned is to setup a cronjob and have it log locally, take a look at Advanced Logging setup in this document.

Regarding your destination disk, you'll want to take a look at your *-flat.vmdk and not your .vmdk as that is just the descriptor and not the actual disk and the final output will also be determined based on how you're backing up your VM, whether that is thin/thick or any other supported format. If something were to error, you'll see something logged and you'll end up with a bad backup.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators

If you find this information useful, please award points for "correct" or "helpful".

lamw · ‎09-28-2009

Thanks for the comments, unfortunately your solution is not a valid work around. The script does not support any VM(s) with snapshots, but I also do not make the assumption that a user may or may not modify the script. Having said that, I will only remove the most recent snapshot which should have been generated by my script only and hence it will continue to loop to ensure that the *-delta-XXXXX.vmdk are gone. The other thing that I'm not 100% sure is if snapshot.removeall is a blocking function because snapshot.remove is not and you could run into a situation where you'll have multiple backups completing and committing multiple snapshots which could hinder performance. This will guarantee that each VM is back into a consistent state before moving onto the next VM if something were to go wrong.

I've not heard others having issues with removing the snapshots.

=========================================================================

William Lam

VMware vExpert 2009

VMware ESX/ESXi scripts and resources at: http://engineering.ucsb.edu/~duonglt/vmware/

VMware Code Central - Scripts/Sample code for Developers and Administrators