onlineforums
Explorer
- Joined
- Oct 1, 2017
- Messages
- 56
I'm wondering what tools are at my disposal to see why an offsite replication is so dang slow.
The source dataset has 300 GB of files on average around 3mb per file.
The source internet is roughly 100 GB down / 20 GB upload cable connection
The source router is showing roughly 1 MB outbound per second.
It is using OpenVPN to reach the destination FreeNAS box.
It is using encryption in the replication options.
The replication started about 36 hours ago and has transferred ~ 150 GB of data out of the 300 GB.
The replication GUI shows 80% complete but when I do zfs list -o space on the destination I am showing 150 GB transferred (of the 300 GB). Also the 80% took place in the first 36 hours but I noted that over the past 12 hours the GUI went from 76% to 80% complete. That is 4% per 12 hours and there is 20% more to go which is an additional 60 more hours.
I am hoping that the transfer is complete over the next 24 hours as I have to work on this internet and I can feel that there is network activity causing slow downs in just normal tasks.
My concerns or questions to the group is as follows:
1. The source router says 1 MB outbound per second and if that is true then the whole 300 GB should be done in less than 48 hours and yet after 36 hours it is only 50% complete according to zfs list (80% complete according to source GUI).
2. I know that there is a lot more to a network but is there anyway to see if the 1 MB outbound per second is being throttled by the ISP such as running nperf or something from the FreeNAS source box to destination box?
3. Is the GUI % complete unreliable? Is there a CLI % complete showing what has been done and what is to come?
4. Is there a way to "pause" the replication task to complete (resume) only at night?
5. Is there a way to cancel the already running replication task from GUI or CLI?
Thanks!
The source dataset has 300 GB of files on average around 3mb per file.
The source internet is roughly 100 GB down / 20 GB upload cable connection
The source router is showing roughly 1 MB outbound per second.
It is using OpenVPN to reach the destination FreeNAS box.
It is using encryption in the replication options.
The replication started about 36 hours ago and has transferred ~ 150 GB of data out of the 300 GB.
The replication GUI shows 80% complete but when I do zfs list -o space on the destination I am showing 150 GB transferred (of the 300 GB). Also the 80% took place in the first 36 hours but I noted that over the past 12 hours the GUI went from 76% to 80% complete. That is 4% per 12 hours and there is 20% more to go which is an additional 60 more hours.
I am hoping that the transfer is complete over the next 24 hours as I have to work on this internet and I can feel that there is network activity causing slow downs in just normal tasks.
My concerns or questions to the group is as follows:
1. The source router says 1 MB outbound per second and if that is true then the whole 300 GB should be done in less than 48 hours and yet after 36 hours it is only 50% complete according to zfs list (80% complete according to source GUI).
2. I know that there is a lot more to a network but is there anyway to see if the 1 MB outbound per second is being throttled by the ISP such as running nperf or something from the FreeNAS source box to destination box?
3. Is the GUI % complete unreliable? Is there a CLI % complete showing what has been done and what is to come?
4. Is there a way to "pause" the replication task to complete (resume) only at night?
5. Is there a way to cancel the already running replication task from GUI or CLI?
Thanks!
