vmware restore problem

[test...@ncc-1701a.de]

Hi,

i have a big issue.

I had a backup of a vm.

Now the vm is broke.

I thought, no prob, i have a backup.

But because of internet problem and vcenter on remote site, restore failed.

After i realized the internet problem i wanted to restore to localvmdk, but because of config error bareos overwrote 1 of the files with catalog.

It´s the second from 14 files.


Maybe someone has a hint to restore my files ?

[Erik Šolc]

Hi,
you can restore it to director. Run restore from bconsole and modify job, change the Client and Where to restore, so number 5 and 10.
OK to run? (yes/mod/no): mod
Parameters to modify:
1: Level
2: Storage
3: Job
4: FileSet
5: Restore Client
6: Backup Format
7: When
8: Priority
9: Bootstrap
10: Where
11: File Relocation
12: Replace
13: JobId
14: Plugin Options

So it will restore vmdk of the machine to the path you set in No. 10.

Erik

[test...@ncc-1701a.de]

Hi,


yes, i did that, but than i have a file with the data in it, but its vadp.

My thought was the same, just pipe this to the bareos_vadp, but this reads the json with the cbt data from the backupfile.

I gave the dumper a new json with other infos but it doesn´t wont to read it.

[test...@ncc-1701a.de]

the main problem is:

If i do that, i have file, but its not a vmdk and not a raw file. It has the vadp restore infos in it, thats seems to be my problem.

Question is, how can i remove this.

bareos_vadp_dumper seems not an option, because volume 2 from 14 is missing, vadp_dumper breaks here, because it can not find the next cbt entry.

As far as i understand the backup with vadp_dumper:

bareos_vadp_dumper read the drive info, put it in json and mark it with a magic-header

with the cbt i think is the same, magic-header, json with start and length, maybe magic-header, data.

next magic-header, json, data and so on.

But my problem is, after data is restored, it "jumps" to the next magic-header and want to read this, but because of missing volume, it should ignore the next and maybe more headers, but it didt´n.

I looked in the sources from bareos_vadp_dumper, i think i understand it, but my programming skills are not "so good" to change that behavior.

Am Montag, 15. April 2019 15:05:51 UTC+2 schrieb Erik Šolc:

[vol...@gmail.com]

I had restored VmWare VMs successfully, from Full backup definitely.
When it comes from Full+Diff+Inc+Inc, usually I got working VM with corrupted filesystem, that usually found next day after restore.

You can restore VMDK directly to VC, but it have to be same original VM, clone or similar new VM will not suit.

Usually, it is good to restore as local vmdk, then upload it to ESX.
Interesting fact, that the restored vmdk good for KVM, vmplayer, VirtualBox, but not for ESX where it taken. You have to run extra command on ESX itself to convert VMDK format.

I wrote http://www.voleg.info/bareos-restore-vmware.html article to memorize the procedure. Probably it will good for you too.