Joakim Koed Dec 18 4:09PM 2017
As always, thanks for a great piece of software!
I currently run the following from a Linux VM in crontab:
0 2 6,13,20,27 * * plink -ssh -2 -pw ********* email@example.com "/vmfs/volumes/datastore1/verticalbackup/vertical backup -e FreeNAS --threads 4 --email"
0 3 6,13,20,27 * * plink -ssh -2 -pw ********* firstname.lastname@example.org "/vmfs/volumes/datastore1/verticalbackup/vertical prune -k 0:30"
I looked at the guide but I was a bit unsure of how to make this happen 100%. Does prun need to a cron schedule as well?
I would really like to be able to run cron and prun from within ESXi if you can help me with this.
Also I will soon buy this for my company's ESXi free version, if I can just get this last ting to work perfectly.
gchen Dec 18 10:16PM 2017
The cron command can only support one cron entry. However, you can set up a cron backup command and then use a post-prune script to run the prune command after the backup. The User Guide has a section on how to set this up, but basically you just need to create a
.verticalbackup/scripts/post-backup script that runs the prune command.
The only downside of doing this is that when the prune command is running, the backup process is still there, and because ESXi only allows a limited amount of memory to host processes, this may run into a memory issue sooner than running the prune command from your Linux VM.
Another option is to run the Duplicacy command line directly on your Linux VM for the prune operation. And Duplicacy is free for this purpose.
Joakim Koed Dec 19 9:29AM 2017
Thanks for your answer.
One thing I'm still unsure about:
You talk about memory issues when running prune. Is that only when running the prune script you talked about, or will my current solution (where the Linux VM just sends an SSH-command to ESXi) also mean memory issues?
gchen Dec 19 1:19PM 2017
Running the prune command on ESXi via ssh may still run into memory issues if you have too many backups in the storage, but it will certainly be less severe than running it as a post-backup script.
The recommended way is to run Duplicacy to prune on a VM or even a physical machine where you can have much more memory at your disposal.
Joakim Koed Dec 27 2:58PM 2017
Hey again :)
Looking into Duplicacy.
How many VMs do you think before verticalbackup prune is not good enough? I currently have my backup-storage on NFS though FreeNAS connected to same virtual switch in ESXi (so over 10Gbit) - Running with over 40GB of Free RAM.
gchen Dec 27 9:38PM 2017
A process running on the ESXi host can only have a limited amount of memory, no matter how much RAM the host has. My estimate is that the prune command may run out of memory once the total size of the storage is more than 1TB.
Henrik Dec 29 12:41PM 2017
If I want to run the prune job from another machine, how should the repository be initialized? Should I initialize it using the same ID as on the ESXi host?
Nice product by the way. Only issue I've come across so far is that SFTP does not work on other ports than 22. I made sure that the firewall was configured but still go the error "UnboundLocalError: local variable 'hostAndPort' referenced before assignment". I can make a new topic about this if needed.
gchen Dec 29 11:34PM 2017
The 1.1.5 build (sha256: 221045d73ddfcb4978b44d93fafe450b46541d3f6642aaa6b1150b1355fd4514) should fix this SFTP port issue.
gchen Dec 31 6:06PM 2017
Henrik, sorry I forgot to answer your question. If the repository is used only for the prune job then the ID doesn't matter. You can use any ID.
Henrik Jan 13 3:56AM 2018
Thank you, I got the prune job running from another physical machine. I haven't tested if the SFTP issue has been fixed but I will at some point.
Is there a reason why you don't update the binary link on the Guide page at the same time? When I downloaded the update from https://acrosync.com/esxi/vertical the binary was still the old one in http://verticalbackup.com/esxi/vertical.
gchen Jan 13 2:41PM 2018