Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

rawweb

macrumors 65816
Original poster
Aug 7, 2015
1,126
943
Hello. Hoping to have a better understanding of what is wrong with a small setup I have and maybe some way to fix it.

I have a 2014 MacMini acting as a "file server" via Thunderbolt 2 Bridge with an 8 drive SSD Raid-6 array. I'm getting decent speeds on the host computer.

Screen Shot 2019-08-21 at 9.23.50 AM.png

On a couple of iMac Pro's connected via Thunderbolt 2 adapter, I'm getting much slower speeds and I honestly can't understand why the speeds aren't similar to the host computer.
Screen Shot 2019-08-21 at 9.24.20 AM.png

Is there a better way to setup this network so speeds can increase on the various machines? Right now, everything is connected via Thunderbolt Bridge and native Mojave file sharing.
 

Flint Ironstag

macrumors 65816
Dec 1, 2013
1,334
744
Houston, TX USA
- system preferences
- network
- gear icon
- set service order
- drag Thunderbolt interface to the top of the list
- reboot both Macs and benchmark again.

Looking at those speeds my first guess is that maybe they're talking via wifi.
 

rawweb

macrumors 65816
Original poster
Aug 7, 2015
1,126
943
- system preferences
- network
- gear icon
- set service order
- drag Thunderbolt interface to the top of the list
- reboot both Macs and benchmark again.

Looking at those speeds my first guess is that maybe they're talking via wifi.

Hey Flint, appreciate the response. I went ahead and tried what you suggested step by step and even shut off all other networking interfaces to the devices. Unfortunately, it doesn't appear to be any different in the speed test. Do you have any other suggestions? I would love to solve this once and for all.


Connected to host via SMB:
Screen Shot 2019-08-22 at 3.45.16 PM.png

Connected via AFP:
Screen Shot 2019-08-22 at 3.44.24 PM.png
 
  • Like
Reactions: Flint Ironstag

Flint Ironstag

macrumors 65816
Dec 1, 2013
1,334
744
Houston, TX USA
Have you tried creating a fresh test account on each machine and running the benchmark again? This would tell us whether it's a systemwide or account issue.

[edit] and these are genuine Apple TB3 - TB2 adapters right?
 
Last edited:

rawweb

macrumors 65816
Original poster
Aug 7, 2015
1,126
943
Have you tried creating a fresh test account on each machine and running the benchmark again? This would tell us whether it's a systemwide or account issue.

[edit] and these are genuine Apple TB3 - TB2 adapters right?

Yes genuine Apple TB3/2 adapters. I recall trying a few AD accounts and local admin accounts, but I will try again when I’m back in the office and report back. Thank you kindly for helping me look into this!
 
  • Like
Reactions: Flint Ironstag

Flint Ironstag

macrumors 65816
Dec 1, 2013
1,334
744
Houston, TX USA
The amount of times I've found a solution on a forum... :D

Can you post a diagram of how the machines are connected? I would do it so:

mac mini TB2 port#1 -----> Storage
mac mini TB2 port#2 -----> Adapter to iMac Pro (A) TB3 port#1
iMac Pro (A) TB3 port#2 -----> iMac Pro (B) TB3 port#1

I seem to remember mixing TB2 & TB1 devices in a chain isn't always seamless. Some configurations caused the entire chain to slow down to TB1 speeds. It was an onsite rush job ~2015, and only a few TB of data, so the TB1 wasn't a show stopper, and I wasn't curious enough at the time to investigate further.

- fresh admin account on each machine
- establish baseline performance bench on host
- connect 1 iMac Pro & bench
- use an app like https://www.wireshark.org to analyze the traffic and see if there's anything unusual.

Never hurts to verify that all your cables give rated speed in direct connections.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.