Thank you, Marco! I will try splitting the backup of the large system into several smaller jobs and staggering it. If that is feasible with the number/size of files in the production storage, it sounds like an excellent solution!
Patrick On Wed, Oct 21, 2015 at 6:58 AM, Marco Weiß <[email protected]> wrote: > Hi Patrick, > > after posting i realized that my answer is not right ;) You are already > doing a virtual full. Ok > > Now your problem is that you do not have enough time to get a virtual full > out of your backups and you do not have enough storage to get a virtual > full because bareos copies the data first into the new full and deletes the > old after that right? > > If i'm right i don't think that this functionality is implemented in > Bareos but maybe can enlight us here ;) > > We had the same problem with a few TByte and resolved it by splitting the > backups into a few jobs. > With more jobs you can distribute your full backups on different days. > We do every day a full backup of different storage slices so you don't > have to do a full backup of all your data in one backup window > > Regards Marco > > > On Wednesday, October 21, 2015 at 12:40:07 PM UTC+2, Marco Weiß wrote: > > Hi Patrick, > > > > nice to hear that you got it running! And thank you for sharing your > knowledge! > > > > For your the new question i think "virtual full" is the solution you are > searching for. > > Have look into the documentation. > > > > With that option you can do an virtual full backup based on the data on > your backup storage without getting the data from the client again. > > > > Regards Marco > > > > On Tuesday, October 20, 2015 at 7:01:56 PM UTC+2, Patrick Glomski wrote: > > > Hey, Marco. Apologies on taking so long to respond. I dug into the > manual more, specifically the section dealing with Migration jobs. > Apparently bareos will not allow you to do a virtual full backup from one > storage device to the same storage device (the manual even mentioned the > need for different media types). > > > > > > > > > I created a new pool (Brick01Storage_virt) and assigned it as the > 'Next Pool' in my other pool definitions (I've appended sanitized pool and > storage definitions). Virtual Full backups will now run successfully. > However, now I have a serious problem with both backup time and storage > capacity. I was looking for functionality like "Merge everything before > this date into a new Full backup to use as a baseline for incrementals and > differentials". The functionality I'm currently getting is "Make a merged > copy of the data in another storage pool". As my production system contains > a Petabyte of clustered storage, the model I have implemented isn't > reasonable for daily backups. > > > > > > > > > The goal is currently to keep 10 days of incrementals. I can't do a > "Full" backup of this much data every ten days and I also can't copy/merge > a petabyte (and growing) of data in that time (nor do I have the 2x space > needed to do this). Is there any option in bareos for an in-place (or near > to in-place) merge of old jobs? > > > > > > > > > If not, is there any other solution you can think of for this type of > problem? > > > > > > > > > > > > I appreciate your time, > > > > > > > > > Patrick > > -- > You received this message because you are subscribed to the Google Groups > "bareos-users" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To post to this group, send email to [email protected]. > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "bareos-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. For more options, visit https://groups.google.com/d/optout.
