On Sat, Dec 23, 2023 at 10:49 Brock Palen <bro...@mlds-networks.com> wrote:
> I do this and it works but it has some bugs that are reported but not had > progress in few years. > > My setup is: > > Disk Pools: > AI-Incremental > AI-Consolidated > > Tape Pool: > LTO > Offsite > > I run the AI setup on the disk pools but I use a Migration Job to migrate > AI-Consolidated jobs to Offsite. > This is where one bug shows up ’sometimes’ Bareos will correctly > understand that the job it needs is on tape but if you run parallel jobs it > won’t correctly wait for the tape drive to free up if already busy. It will > instead fail. It’s easy to work around you just run it again when that > happens. > > Sometimes it also gets hungup with multiple disk ‘devices’ where it won’t > swap the needed disk volume even though it’s not being used. You can avoid > all of these by forcing serial jobs, and all ‘bugs’ are more inconvenient > than show stoppers. > > > I also use the Offsite pool (which for me is a second tape drive but not > part of an autoloader) to write monthly offsite copies using VirutalFulls > > https://docs.bareos.org/TasksAndConcepts/AlwaysIncrementalBackupScheme.html#long-term-storage-of-always-incremental-jobs > > This is not AI related, but regular copy jobs are not reocmended because > the consolidate pulls in those jobs. So this is more of an ‘emergency get > it back no more than a month old’ offsite copy. > > > The migration job runs a script that checks and truncates all pruned > volumes to free disk space rather than waiting for them to expire by age. > Again this setup I find requires few TB of disk (need to run full backups > to disk) and requires some watching. To make sure I understand, this requires enough disk space for a full backup? I can't write out each tape's worth of data to tape while the full backup is running? > #!/bin/bash > POOL=$1 > for x in `echo "list volumes pool=${POOL}" | bconsole | grep -v "list > volumes" | grep $POOL | awk -F\| '{print $3}'` > do > echo "prune volume=$x yes" > done | bconsole > > # actaully free up disk space > echo "truncate volstatus=Purged pool=$POOL yes" \ > | bconsole > > > Pool { > Name = AI-Incremental > Pool Type = Backup > Recycle = yes # Bareos can automatically recycle > Volumes > Auto Prune = yes # Prune expired volumes > Volume Retention = 6 months # How long should jobs be kept? > Maximum Volume Bytes = 10G # Limit Volume size to something > reasonable > Label Format = "AI-Incremental-" > Volume Use Duration = 7d > Storage = File > Next Pool = AI-Consolidated # consolidated jobs go to this pool > Action On Purge=Truncate > Migration High Bytes = 500G > Migration Low Bytes = 300G > } > > Pool { > Name = AI-Consolidated > Pool Type = Backup > Recycle = yes # Bareos can automatically recycle > Volumes > Auto Prune = yes # Prune expired volumes > Volume Retention = 6 months # How long should jobs be kept? > Maximum Volume Bytes = 50G # Limit Volume size to something > reasonable > Label Format = "AI-Consolidated-" > Volume Use Duration = 2 days > Storage = File > Next Pool = Longterm # copy jobs write to this pool > Action On Purge=Truncate > Migration Time = 7 days > Migration High Bytes = 600G > Migration Low Bytes = 300G > } > > > Job { > Name = "Migrate-To-Offsite-AI-Consolidated-size" > Client = myth-fd > Type = Migrate > Purge Migration Job = yes > Pool = AI-Consolidated > Level = Full > Next Pool = LTO > Schedule = WeeklyCycleAfterBackup > Allow Duplicate Jobs = no > Priority = 4 #before catalog dump > Messages = Standard > Selection Type = PoolOccupancy > Spool Data = No > Selection Pattern = "." > RunAfterJob = "sudo /usr/local/bin/prune.sh AI-Consolidated" > Enabled = no > } > > Brock Palen > bro...@mlds-networks.com > www.mlds-networks.com > Websites, Linux, Hosting, Joomla, Consulting > > > > > On Dec 17, 2023, at 7:38 PM, Russell Harmon <eatnumb...@gmail.com> > wrote: > > > > Hi there, > > > > I see the note in > https://docs.bareos.org/TasksAndConcepts/AlwaysIncrementalBackupScheme.html#storages-and-pools > about "at least two storages are needed" for Always Incremental, but is > there a way to make this work with a *temporary* secondary storage? > > > > I have just one tape drive, and while I can temporarily use local disk > as a secondary storage, I don't have enough for a full backup on disk... > only enough for one tape's worth of data. > > > > Is there any way to spool a consolidated job to disk, then let me swap > in a tape for the consolidated pool (therefore removing the incremental > pool's tape), despool it, then resume consolidating? > > > > Thanks, > > Russ Harmon > > > > -- > > 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 bareos-users+unsubscr...@googlegroups.com. > > To view this discussion on the web visit > https://groups.google.com/d/msgid/bareos-users/6addbe69-f3a4-4bb4-b10d-849d713d5723n%40googlegroups.com > . > > -- 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 bareos-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/bareos-users/CA%2BzrezSxi_eLzbQ%2BZvaikD69%3DFcF2Vtv%2BBRrLawBNvJYk5jXcw%40mail.gmail.com.
