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

IdentityCrisis

Suspended
Sep 9, 2018
684
345
Does the link aggregation actually make any difference if you don't use a computer with two ethernet? Link aggregation may increase fault tolerance for a single device but does it speed the transferring without two ethernet in a client? If you have multiple device connecting to the nas at the same time, those may use different ip's (nas/2 ethernet) though.

I haven't seen a difference using two ethernet ports with link aggregation. Maybe if I had a ton of devices pulling from it at once, but it didn't make a speed difference.
 

Howard2k

macrumors 603
Mar 10, 2016
5,699
5,646
Does the link aggregation actually make any difference if you don't use a computer with two ethernet? Link aggregation may increase fault tolerance for a single device but does it speed the transferring without two ethernet in a client? If you have multiple device connecting to the nas at the same time, those may use different ip's (nas/2 ethernet) though.


The Synology Link Aggregation is surprisingly decent. The documentation is not great, but it looks like it supports 802.3ad so with the right hardware on the other ends it's possible to see the benefit if you have only one computer transferring data, although the assumption is that the computer is able to oversubscribe the NAS links.

https://www.synology.com/en-global/...p/DSM/AdminCenter/connection_network_linkaggr

So a computer with a 10Gb/s Ethernet interface should see more than 1Gb/s of throughput if the NAS has 2 x 1Gb/s links and if the 802.3ad protocol is used and is configured appropriately.

But in real life, at home, it's more about fault tolerance.
 

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
have you checked what version of smb macOS is using? (terminal - "smbutil statshare -a")


I've tried different kind of protocols. MacOS Mojave uses SMB 2.1. as default. AFP has been the same as SMB, while NFS has been the fastes so far (with a smal margin).

you can also try to turn off client signing - some has said that it helped to improve the speed. You can see if the signing is on in terminal (the same command as above) (https://support.apple.com/en-bh/HT205926)

Thanks for the tip, I've actually tried that as well. It gave some better speed, but we are still just talking about a smal increase.

I have the same problem with my nas and mbp. Win laptop gets double the speed while mbp is slower. Both are connected on the same way. I haven't figured it out yet but browsing the content (seeing thumbnails) from the nas with mbp is painfully slow if the directory has lots of stuff inside, e.g. photos... But that is another issue than just slow transferring speed and apple has made recently an article about improving the browsing speed. unfortunately I tried it and it didnt work.

I don't have that. The speeds are the same on Windows and on MacOS. My directories are pretty clean as well. Not many files in each folder.

Good stuff, very thorough. You do have another network using the same channel as yours in the 5GHz space (Get-72337C?) and the area up by channel 150 is free. So I would move my network there.

I don't think that's the issue though, unless that Get-72337C network is being hammered, which would seem unlikely. But if we're grasping at straws...

The good news is we've ruled out the NAS, since you're seeing poor performance with some other configurations. So that's progress - test 3.1, for example.

And aside from 4.1, it kinda looks like it's a wifi issue, though it's inconclusive.

These are the challenges that I see:
* There's another network using the same channel.
* Your RSSI shown above is -57dBm.
* Your noise shown above is -88dBm.

None of these things individually should be killing your wifi. And in totality, I don't think there's enough there to cripple it either, to be honest. but wifi can be odd. Especially when the signal degrades and you get retransmissions.

So I would change the channel to 153, or thereabouts, and ensure that all your other wifi devices are off, and try test 3.1 and 4.1 again.

But again, I'm not convinced it will fix anything. At least we can rule out that other network.

Thanks. Yeah, I agree on what you're saying. It should not cripple my WiFi that much. Though, it seems like that Get-net has moved. Still no change in speed. I can only change between channel 36, 40, 44 and 48.

upload_2018-12-8_22-43-39.png

ANYHOW, I've had a really busy day today. I went to the store and bought two new routers. Another one of the same I own, just in case mine was defect somehow, and Netgear Nighthawk X4S AC2600. Unfortunately, this gave me NOTHING valuable :((. Both routers gave me the same speed. Around 20-30MB/s read. I even bought brand new ethernet cables.

Now.... NOW I'm really lost.

I think you are covering the issue and being pretty thorough. Not much more you can rule out. You have tested on multiple machines, ethernet, wireless, etc.

I really hope you figure this out because I would like to know what happened.

I use a Synology NAS with port aggregation. I have two ethernet cables plugged into my Asus AC5300 which also supports port aggregation. I have not tested speeds but it is quick enough to run flex and transcode 4k. I have noticed my wireless signal and speeds fluctuate depending on where I am in the house. 2.4ghz vs 5ghz really makes a difference.

Have not read the whole thread, but have you tried reseting your NAS and setting up fresh?

Yeah, thanks. I'm getting desperate to get this work. 4K is not possible for me, I even have a hard time with 1080p. Yes, I actually did that before I posted here. I thought first that it could be because my NAS was too full, so I upgraded it with another 4TB. At the same time I factory reset the whole thing.
 
Last edited:
  • Like
Reactions: Howard2k

OLDGUYWITHAHIFI

Suspended
Nov 14, 2018
235
354
I've tried different kind of protocols. MacOS Mojave uses SMB 2.1. as default. AFP has been the same as SMB, while NFS has been the fastes so far (with a smal margin).



Thanks for the tip, I've actually tried that as well. It gave some better speed, but we are still just talking about a smal increase.



I don't have that. The speeds are the same on Windows and on MacOS. My directories are pretty clean as well. Not many files in each folder.



Thanks. Yeah, I agree on what you're saying. It should not cripple my WiFi that much. Though, it seems like that Get-net has moved. Still no change in speed. I can only change between channel 36, 40, 44 and 48.

View attachment 809496

ANYHOW, I've had a really busy day today. I went to the store and bought two new routers. Another one of the same I own, just in case mine was defect somehow, and Netgear Nighthawk X4S AC2600. Unfortunately, this gave me NOTHING valuable :((. Both routers gave me the same speed. Around 20-30MB/s read. I even bought brand new ethernet cables.

Now.... NOW I'm really lost.



Yeah, thanks. I'm getting desperate to get this work. 4K is not possible for me, I even have a hard time with 1080p. Yes, I actually did that before I posted here. I thought first that it could be because my NAS was too full, so I upgraded it with another 4TB. At the same time I factory reset the whole thing.

Are you sure you are on a 5ghz network? You stated I can only change between channel 36, 40, 44 and 48. I am no expert, but I thought 2.4 runs in the lower range of numbers, and 5 Ghz runs a lot higher like around 150 or 160?

If you are actually on a 2.4 band your numbers would be a lot lower.
 

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
Are you sure you are on a 5ghz network? You stated I can only change between channel 36, 40, 44 and 48. I am no expert, but I thought 2.4 runs in the lower range of numbers, and 5 Ghz runs a lot higher like around 150 or 160?

If you are actually on a 2.4 band your numbers would be a lot lower.

Yes. Given what I can see from my WiFI Scanner app. 2.4Ghz goes from 0-14, and 5Ghz from 34 to ~65 and ~100 to ~160+.

Anyhow, these are the settings on my router.
upload_2018-12-9_0-45-1.png
 

OLDGUYWITHAHIFI

Suspended
Nov 14, 2018
235
354
Yes. Given what I can see from my WiFI Scanner app. 2.4Ghz goes from 0-14, and 5Ghz from 34 to ~65 and ~100 to ~160+.

Anyhow, these are the settings on my router.
View attachment 809526

What sucks about these sort of problems is you could easily kill a whole day or two just screwing with this stuff. Hope you get it sorted. You seem to know a lot more about troubleshooting than I do so I'll bow out. Best of luck.
 
  • Like
Reactions: Howard2k

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
What sucks about these sort of problems is you could easily kill a whole day or two just screwing with this stuff. Hope you get it sorted. You seem to know a lot more about troubleshooting than I do so I'll bow out. Best of luck.

Well, I'm on day 5 or 6 now, so yeeeaah...
Hehe, thank you! I need it.
 
  • Like
Reactions: OLDGUYWITHAHIFI

Howard2k

macrumors 603
Mar 10, 2016
5,699
5,646
Yes. Given what I can see from my WiFI Scanner app. 2.4Ghz goes from 0-14, and 5Ghz from 34 to ~65 and ~100 to ~160+.

Anyhow, these are the settings on my router.
View attachment 809526


Hey what other options do you have in the A, N, and AC field (BAND item)?

With the Netgear, you see the same speeds using wifi to ethernet and ethernet to Ethernet?
Do you have any software firewalls enabled on any of the computers? Any network "security" apps?
 

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
Hey what other options do you have in the A, N, and AC field (BAND item)?

With the Netgear, you see the same speeds using wifi to ethernet and ethernet to Ethernet?
Do you have any software firewalls enabled on any of the computers? Any network "security" apps?

These are my choices.
upload_2018-12-9_15-33-3.png

No, nothing. Everything is clean and default from factory.
 

Howard2k

macrumors 603
Mar 10, 2016
5,699
5,646
These are my choices.
View attachment 809642

No, nothing. Everything is clean and default from factory.


Thanks. ok, you're got the right choice for sure.

I’m out of ideas pretty much. I think it’s possible that you might have more than one issue?

NAS
You're not getting the speeds I'd expect for a large file transfer between the Mini and MacBook, but your Mini is seeing great speeds from the NAS directly. So that rules out that NAS as any sort of issue. Or at least as an exclusive issue.

Wifi Hardware
But you've tried a different wireless router of a different brand, so we can rule out a fault in your specific Lynx router, as well as some sort of compatibility issue with Apple's wifi chipset and the Lynx. Unless there is an issue with the Pro's wireless chipset talking to both the vendor's wireless routers, which would seem incredibly unlikely. But not impossible.

Protocols
You've also tried different protocols. SMB1 will be slow, but even if this was an SMB1 issue, you should still see faster speeds. SMB1 will do more than 10MB/s over a local connection, I had it running at 40MB/s over wifi to my NAS and I'm sure it'll go faster than that too.


Going back to these tests:
https://forums.macrumors.com/threads/nas-serving-low-speed.2158882/#post-26890222

With Mac Mini connected directly to router, rest with WiFi:
3.1) from Mac Mini - Ethernet - Router - WiFi - MacBook Pro: ~24MB/s
4.1) from MacBook Pro - WiFi - Router - Ethernet - Mac Mini: ~50MB/s
5.1) from MacBook Pro - Ethernet - Router - WiFi - Windows PC: ~10MB/s
6.1) from Mac Mini - Ethernet - Router - WiFi - Windows PC: ~10MB/s


4.1 is an anomaly. And you said that it was the second time it was faster than the first IIRC. So I'm not sure how to interpret that one to be honest.

So assuming that it's a wireless issue, despite my comments about the wireless hardware above, I'd consider DFS and channel size.

For 802.11ac in many countries the specification requests or requires something called DFS - Dynamic Frequency Selection. This applies particularly in the higher numbered channels. This MIGHT also be why you cannot select the higher channels. Some airports use radar in the 5GHz range and so if your wireless router sees radar (or thinks that it does) it's supposed to jump to another channel.

For the channel size - you're using 80MHz right now, which is a wide channel. I live in apartment building and use 80MHz and so I'm a good candidate for having this issue that I'm about to describe, but I'm not having this issue. The challenge with 80MHz on 802.11ac is that you're taking a really wide channels, which is great for jamming data through, but it's also opening you up to exposure to radio interference over a wider range (because the channel is wider). Faster, but more open to interference. More interference will also cripple throughput because there's a kind of black hole where the signal degrades and you end up doing a mass of retransmissions, which then compounds the issue. Again, I'm not seeing this issue myself, and my 5GHz space is more crowded than yours (a little, not a lot).

Your signal strength here:
https://forums.macrumors.com/threads/nas-serving-low-speed.2158882/#post-26889051
shows RSSI -57dBm and noise of -88dBm. Ideally you want the gap between those numbers to be larger. The RSSI should be as closed to 0 as possible, and the noise should be as close to 100 as possible. Right now I have -49 RSSI and -92 noise, and a 702Mb/s connection. I'm not terribly close to my router.


So if I was you, I would take the Netgear router, your Mac Mini, and your MBPro, and forget the rest for now. I would shutdown my current router and other devices, and I would focus on trying to get a working traffic flow between just these devices.

On the router, I would decrease the channels size from 80MHz to 40MHz, but also during the test place the MBPro as close to the router as possible. Decreasing that channel size will decrease your bandwidth too (which seems counter intuitive because we want more speed, not less speed) but by moving closer to the router you'll negotiate a higher speed link to compensate. With 40MHz you should top out at either 540Mb/s or 600Mb/s when your MacBook is about a foot from the router. With the NetGear, hopefully you can try this with a channel in the ~34 range as well as a channel in the ~154 range.

Mini <-> Ethernet <-> Router <-> wifi <-> MBPro

Do the opt-click wifi thing and verify that you're seeing a 40MHz connection in the 500Mb/s or higher range, and then re-do 3.1 and 4.1. And possibly 5.1 and 6.1 too. The problem here is that 500Mb/s is 62.5MB/s, but you previous had test results of 24MB/s and 50MB/s. I'd hope to see more than 24MB/s for sure. I wouldn't be too alarmed if you don't see 50MB/s as you won't get anywhere close to 100% efficiency.


Then I would try the same in the 2.4Ghz space, pick what appears to be a relatively open channel using the app from there:
https://forums.macrumors.com/threads/nas-serving-low-speed.2158882/#post-26890222

And compare. Use the largest channels size for 2.4GHz.

Again, nothing else connected to the wifi. I would also turn off wifi security during the test too. Just don't forget to turn it back on.


I guess it's possible that your MBPro wifi is questionable too? I don't think we've seen a test with the MBPro to any other device that has shown high wifi throughout have we?
 

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
Thanks. ok, you're got the right choice for sure.

I’m out of ideas pretty much. I think it’s possible that you might have more than one issue?

Yeah, so am I :(. Of course, there is always a possibility for that, but I'm not really sure where to start. I feel like I've checked everything.

NAS
You're not getting the speeds I'd expect for a large file transfer between the Mini and MacBook, but your Mini is seeing great speeds from the NAS directly. So that rules out that NAS as any sort of issue. Or at least as an exclusive issue.
Well, the Mini doesn't see great speeds through WiFi. The speed is the same on all units. Ethernet is of course fast.

Wifi Hardware
But you've tried a different wireless router of a different brand, so we can rule out a fault in your specific Lynx router, as well as some sort of compatibility issue with Apple's wifi chipset and the Lynx. Unless there is an issue with the Pro's wireless chipset talking to both the vendor's wireless routers, which would seem incredibly unlikely. But not impossible.

Yeah, and given that I've tried on three different units, with all different hardware it's unlikely there is a problem with my HW.

Protocols
You've also tried different protocols. SMB1 will be slow, but even if this was an SMB1 issue, you should still see faster speeds. SMB1 will do more than 10MB/s over a local connection, I had it running at 40MB/s over wifi to my NAS and I'm sure it'll go faster than that too.

Agreed. Everything will give me faster speeds than what I have right now.

Going back to these tests:
https://forums.macrumors.com/threads/nas-serving-low-speed.2158882/#post-26890222

4.1 is an anomaly. And you said that it was the second time it was faster than the first IIRC. So I'm not sure how to interpret that one to be honest.

So assuming that it's a wireless issue, despite my comments about the wireless hardware above, I'd consider DFS and channel size.

For the channel size - you're using 80MHz right now, which is a wide channel. I live in apartment building and use 80MHz and so I'm a good candidate for having this issue that I'm about to describe, but I'm not having this issue. The challenge with 80MHz on 802.11ac is that you're taking a really wide channels, which is great for jamming data through, but it's also opening you up to exposure to radio interference over a wider range (because the channel is wider). Faster, but more open to interference. More interference will also cripple throughput because there's a kind of black hole where the signal degrades and you end up doing a mass of retransmissions, which then compounds the issue. Again, I'm not seeing this issue myself, and my 5GHz space is more crowded than yours (a little, not a lot).
I'll try play around with those settings, tho I don't belive it would help. Given that I don't have much interference.

Your signal strength here:
https://forums.macrumors.com/threads/nas-serving-low-speed.2158882/#post-26889051
shows RSSI -57dBm and noise of -88dBm. Ideally you want the gap between those numbers to be larger. The RSSI should be as closed to 0 as possible, and the noise should be as close to 100 as possible. Right now I have -49 RSSI and -92 noise, and a 702Mb/s connection. I'm not terribly close to my router.
I moved the computer next to the router and checked again. I get RSSI: -42 dBm, noise -88 dBm and TX 975Mbps, but still only 15-20MB/s.

So if I was you, I would take the Netgear router, your Mac Mini, and your MBPro, and forget the rest for now. I would shutdown my current router and other devices, and I would focus on trying to get a working traffic flow between just these devices.

On the router, I would decrease the channels size from 80MHz to 40MHz, but also during the test place the MBPro as close to the router as possible. Decreasing that channel size will decrease your bandwidth too (which seems counter intuitive because we want more speed, not less speed) but by moving closer to the router you'll negotiate a higher speed link to compensate. With 40MHz you should top out at either 540Mb/s or 600Mb/s when your MacBook is about a foot from the router. With the NetGear, hopefully you can try this with a channel in the ~34 range as well as a channel in the ~154 range.

Mini <-> Ethernet <-> Router <-> wifi <-> MBPro

Do the opt-click wifi thing and verify that you're seeing a 40MHz connection in the 500Mb/s or higher range, and then re-do 3.1 and 4.1. And possibly 5.1 and 6.1 too. The problem here is that 500Mb/s is 62.5MB/s, but you previous had test results of 24MB/s and 50MB/s. I'd hope to see more than 24MB/s for sure. I wouldn't be too alarmed if you don't see 50MB/s as you won't get anywhere close to 100% efficiency.


Then I would try the same in the 2.4Ghz space, pick what appears to be a relatively open channel using the app from there:
https://forums.macrumors.com/threads/nas-serving-low-speed.2158882/#post-26890222

And compare. Use the largest channels size for 2.4GHz.

Again, nothing else connected to the wifi. I would also turn off wifi security during the test too. Just don't forget to turn it back on.

Sure, I could try to test this. It's a bit more extensive, but maybe I can rule something out.

Unfortunately I can't choose channels above 48 on 5Ghz and 13 on 2.4Ghz. So 2.4Ghz is probably a bust, given the amount of disturbance.

What security are you referring to? I'm not running any kind of security software on my devices, as far as I know.

I guess it's possible that your MBPro wifi is questionable too? I don't think we've seen a test with the MBPro to any other device that has shown high wifi throughout have we?

Not quite sure what you mean here. The only test (4.1.) that showed some high throughput was the MacBook Pro?
 

Howard2k

macrumors 603
Mar 10, 2016
5,699
5,646
Unfortunately I can't choose channels above 48 on 5Ghz and 13 on 2.4Ghz. So 2.4Ghz is probably a bust, given the amount of disturbance.

Is that on the Jensen unit and the Netgear? Neither allows you to pick a higher channel in 5GHz? Did you take the Netgear back?

For 2.4GHz, any channel is ok. Whichever has the least apparent interference. 1, 6, 11. You might have access to 14 too.

What security are you referring to? I'm not running any kind of security software on my devices, as far as I know.


WPA2 or similar.

Not quite sure what you mean here. The only test (4.1.) that showed some high throughput was the MacBook Pro?

Quite correct, yeah I must have overlooked that.
 

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
Is that on the Jensen unit and the Netgear? Neither allows you to pick a higher channel in 5GHz? Did you take the Netgear back?


For 2.4GHz, any channel is ok. Whichever has the least apparent interference. 1, 6, 11. You might have access to 14 too.

I was planing to take em back, since I couldn't find any differences between the units. But I do have em both until January 20th. I don't think so, but I have to check. Unfortunately I don't have time right now.


WPA2 or similar.

Aha. Didn't even know that WPA2 mattered.
 

Howard2k

macrumors 603
Mar 10, 2016
5,699
5,646
I was planing to take em back, since I couldn't find any differences between the units. But I do have em both until January 20th. I don't think so, but I have to check. Unfortunately I don't have time right now.




Aha. Didn't even know that WPA2 mattered.

WPA2 shouldn't matter in terms of performance. But then, you shouldn't be having these problems too. :)
 

DCWolfie

macrumors member
Original poster
Aug 5, 2010
44
7
WPA2 shouldn't matter in terms of performance. But then, you shouldn't be having these problems too. :)

Hehe, yeah. That's true.

I don't get time to test this before I go on holiday, but I'll continue to try what you mentioned above in early January.
 

mrex

macrumors 68040
Jul 16, 2014
3,458
1,527
europe
back to the slow smb.

i bought usb-c type gigabit ethernet and using smb is still as slow as before and as slow as over wifi. The adapter is recognized as a gigabit connection, eventually - had some issue that macos recognized it as a 100mb connection at first.

win laptop is still double faster even when using wifi while my macbook is slow as a snail for example creating thumbnails from nas share via wifi/ethernet. transfering speed is half of the speed of win laptop, even when using the ethernet adapter.

im ready to give up trying to get smb shares working well with macos.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.