I’m attempting to create a reliable offsite backup system using DroboCopy. However, the first time I ran it took forever. I had to restart my machine so I quite it after 3 days, and I haven’t started it again yet.
My question regards how DroboCopy works. Is it somewhat like Rsync and TimeMachine in that the 1st copy takes forever, and future copies don’t take so long after that?
How does it work?
Yes exactly. It has to copy everything you tell it to at first, then after that it checks to see what was changed.
In my situation, it really doesn’t have to copy much, as both drobos are pretty much up to date. However, the file I set it to copy is almost 500GB. Should it still take forever if it’s just scanning? 99.8% of the files it’s checking are already there, so it’s not having to copy them over to it.
Any suggestions to speed up scan time?
you are copying from a drobo to a drobo?
dont forget in order to know that the files are identical it has to READ ALL of BOTH files before it can compare them, this is what is taking time, not the actual copying.
Thank you. Yes, I’m copying from a local Drobo to a remote Drobo over the internet using SFTP.
The from folder is 500GB. The to folder has 99.8% of the files there.
It takes over 3 days just to scan it.
Any suggestions on ways to speed that up?
Thanks for your help
yikes, erm, dont do it over the internet, since i suspect it is trying to push/pull that 500Gb over the internet just to compare it with the local copy in order to work out what it needs to update, i think maybe you need to use a different program rather than drobo copy.
there will be plenty of other remote sync options for you.
you dont mention what OS you are using
I’m using Mac OS X. Leopard on the local machine, Tiger on the remote machine.
Originally, we were doing it w/ the DroboShare, but that was slower than if I hand delivered them by foot.
All backup programs seem to be doing the same thing (DroboCopy, TimeMachine,Rsync,Carbon Copy). Therefore, I’m not sure if a different software would solve the problem or not.
Maybe a boost in our internet speed?
rsync works by checking the file size and modification times of files. If either of these have changed then it marks the file for transfer. This process should really not take that long.
I recommend you run rsync in verbose mode and with the --progress option and see at what part it is slow.
Thanks for your input ajspencer,
However, isn’t DroboCopy doing the exact same thing Rsync is? Just checking Date Modified and filesize? I wouldn’t think it would take very long, but then again I’m working over a 10Mbsp Download, 1Mbps upload speed (though the best I’ve ever tested it was 6Mbps).
I tend to think it’s more of a bandwidth issue than anything else.
However, all suggestions are welcome. I may try Rsync just to see if there’s any difference.
i think from the soudns of things its comparing the binary data of the files - by sending one to the other - you need something which hashes the two files locally and comapred the hash
i woudl alsmot suggest live sync or dropbox type services
There could be a difference in the transfer part.
Once rsync has identified the files to transfer it then breaks those files into chunks and checksums them. Then it transfers anything thats different. This way it only transfers part of a file that have changed instead of the whole file.
Not sure how DroboCopy handles this. If it just copies the whole files it may be slower when used over the Internet.