Traditional Backup Methodology
I'm just wondering what would be the best way to implement something in VEEAM. Preferably without killing our disk space.
I want to do a full backup. Then daily Incrementals, weekly differential, occasional full (monthly). That part is ok, I effectively get that in VEEAM with appropriate options.
but what if, one of our users wants a file that was deleted 3 months ago. Most recent full isn't going to have that file, and I think we only keep so many restore points on disk and only a couple weeks of VM retention. Traditionally with disk/tape backups, can go to a specific backup file, and just retrieve that file. Or maybe a restore to a specific point in time as opposed to what the system looks like today or yesterday.
Also, there really isn't a way to keep monthly backups in a separate location. In order to control the backup correctly, everything must be controlled from a single job. So if I do a monthly, and want to offload it to a separate disk for archival purposes and data retention, I can't really do that. I could kind of do it as a separate job, but it seems like I loose a bit of cohesiveness with the backup if I start scheduling separate jobs to backup the same VM for different reasons.
Maybe I'm missing something obvious??
I want to do a full backup. Then daily Incrementals, weekly differential, occasional full (monthly). That part is ok, I effectively get that in VEEAM with appropriate options.
but what if, one of our users wants a file that was deleted 3 months ago. Most recent full isn't going to have that file, and I think we only keep so many restore points on disk and only a couple weeks of VM retention. Traditionally with disk/tape backups, can go to a specific backup file, and just retrieve that file. Or maybe a restore to a specific point in time as opposed to what the system looks like today or yesterday.
Also, there really isn't a way to keep monthly backups in a separate location. In order to control the backup correctly, everything must be controlled from a single job. So if I do a monthly, and want to offload it to a separate disk for archival purposes and data retention, I can't really do that. I could kind of do it as a separate job, but it seems like I loose a bit of cohesiveness with the backup if I start scheduling separate jobs to backup the same VM for different reasons.
Maybe I'm missing something obvious??
-
- VP, Product Management
- Posts: 27364
- Liked: 2794 times
- Joined: Mar 30, 2009 9:13 am
- Full Name: Vitaliy Safarov
- Contact:
Re: Traditional Backup Methodology
Hello Carl,
Could you please tell my why do you want to control everything from the single backup job? If you search the forums, you will see that many customers are already offloading backup files to external NAS devices for the archival purposes.
Also I would say that keeping backup files onsite and offsite is a pretty good idea and you should consider it as a best practice.
Thanks.
Could you please tell my why do you want to control everything from the single backup job? If you search the forums, you will see that many customers are already offloading backup files to external NAS devices for the archival purposes.
Also I would say that keeping backup files onsite and offsite is a pretty good idea and you should consider it as a best practice.
Thanks.
Re: Traditional Backup Methodology
I don't necessarily want to control everything from a central job, but it seems VEEAM wants to do that. If I want to run full backups once a month of a job, I can schedule that within the same job (periodic active fulls). I can't seem to create a "Do full backup job" (just one VBK file) and store it in a separate location (maybe even a separate location based on the month, and upload to offsite location as well). I have to have an incremental or reversed incremental with periodic full in ANY one given backup job.
Maybe I'm missing something fundamental, VM backups are a bit new to me I've dealt with more traditional backup methods to disk and tape and am trying to wrap my head around things.
If I have a separate job to backup the SAME VMs (say for monthly), my standard daily/weekly jobs would get out of sync.
Here's a typical backup scenario. Say we have a full backup in January. We do daily incrementals, and a weekly full (synthetic full) with rollback so there's only one VBK. The next full runs (February). New incrementals and weeklys. If a user creates AND accidentally deletes a file between the weekly backups, and wants it back 3 months later (this is more common than you may think), there is no way to get that file back unless you have a LOT of storage space and keep snapshots for more than 1 week, or at least employ the use of additional tape backups so we can keep backups around for 1 year say.
Am I missing something? Am I making sense?
Maybe I'm missing something fundamental, VM backups are a bit new to me I've dealt with more traditional backup methods to disk and tape and am trying to wrap my head around things.
If I have a separate job to backup the SAME VMs (say for monthly), my standard daily/weekly jobs would get out of sync.
Here's a typical backup scenario. Say we have a full backup in January. We do daily incrementals, and a weekly full (synthetic full) with rollback so there's only one VBK. The next full runs (February). New incrementals and weeklys. If a user creates AND accidentally deletes a file between the weekly backups, and wants it back 3 months later (this is more common than you may think), there is no way to get that file back unless you have a LOT of storage space and keep snapshots for more than 1 week, or at least employ the use of additional tape backups so we can keep backups around for 1 year say.
Am I missing something? Am I making sense?
-
- Veteran
- Posts: 282
- Liked: 26 times
- Joined: Nov 10, 2010 6:51 pm
- Full Name: Seth Bartlett
- Contact:
Re: Traditional Backup Methodology
You could do a monthly job, a daily job and a weekly job if you are wanting to conserve space and have a far back retention policy. You can have separate jobs back up the same VMs and nothing gets out of sync. You would just want to ensure that the jobs never run at the same time(two jobs cannot backup the same VM at the same time). If a user deletes a file between the weekly backups and wants it back 3 months later, there is no way you can get it back without keeping every restore point for every day you run the job.
Skype: Sethbartlett88 - Make sure to label who you are and why you want to add me
Twitter: @sethbartlett
If my post was helpful, please like it. Sometimes twitter is quicker to hit me up if you need me.
Twitter: @sethbartlett
If my post was helpful, please like it. Sometimes twitter is quicker to hit me up if you need me.
-
- Enthusiast
- Posts: 35
- Liked: 5 times
- Joined: May 26, 2011 4:42 pm
- Full Name: Brett
- Contact:
Re: Traditional Backup Methodology
I have a dedupe device so I want to keep my monthly backups in the same location as my normal backups. I want to keep monthly backups for a year. If I scheduled a job that only ran once per month and wanted it do only an active full backup each time (no incremental), do I just set the active full schedule to match the job schedule? Will that force it do a full every time it runs? And if I set it to keep 12 restore points on disk then that will keep 12 full monthly backups? Am I understanding this correctly?
-
- Chief Product Officer
- Posts: 31780
- Liked: 7280 times
- Joined: Jan 01, 2006 1:01 am
- Location: Baar, Switzerland
- Contact:
Re: Traditional Backup Methodology
That's right.
Who is online
Users browsing this forum: Bing [Bot], Semrush [Bot] and 80 guests