Had a search for “daily limits” and most, if not all appear to relate to external services; Google for example.
My case, no such exists: everything inside a DC. We are running RCLONE v 1.57.0-1.el8.x86 on a new (4 months old) Windows server VM, 4vCPU, 128GB RAM. RCLONE will send copies of objects through a firewall to a storage array compatible with the protocol we’re using. All the holes are open, protocols correct, connectivity testing done etc etc
Apart from the vNIC in the VM the rest of the pipework is at least 10Gb. Thats just to test with, we intend to shift it to a physical box which will have 2 x 10Gb “inbound”interfaces and the same outbound, RHEL + lacp into an ACI fabric so will get full bore 20Gb (minus some overheads, obv)
Firewalls not involved in physical as using ACLs. Long story.
One of our techies is claiming “there’s a 4TB limit in 24 hours using RCLONE” but fails to substantiate this. Doubts creep in, uncertainty, sporadic violence in review meetings etc…
It seems mad there should be any limits as far as the application is concerned; does this get unlocked if we pay for support or something ?
I really don’t want to have to go and buy StorageX which will cost an absolute fortune to run in the manner I wish to use RCLONE…
strange, that your tech would state that with facts?
the most important hard limits are usually your internet connection, storage provider and how you tweak rclone's 100+ flags.
there are no rclone limits, nothing to unlock. rclone is free and open-source
tho, you are most welcome to donate or https://rclone.org/sponsor/
4TB/24 hours is 370 Mbit/s continuous - what were you hoping for? I'd check resource usage on the sender and receiver (network, disk io, cpu) to see if you can spot a bottle neck.
Where is the data coming from? Where is it going to?
Which protocol are you using? Some protocols are a lot faster than others.
Also that version of rclone quite old. The current version is 1.70.2
If you would like paid for support with rclone then you can drop us a note to sales@rclone.con
Some updates and answers to some questions, I have had to avoid answering them all.
the problem seems to be more fundamental than I’d thought. They are saying I can only run one instance of RCLONE per OS image… We all thought “Scotty, set five jobs running and let it rip” because we have hods of network capacity. If the problem is true, can it be made to behave itself running in a container ?