Slow upload

core3d.tech

@dc42 Got it, it was more to check that my router wasn't crapping out. I have some class 10 cards coming my way and I'll will post any updates on any progress.

clcollins4

Hey, just chiming in to say I'm having the same issue all of a sudden (I've had my board for around half a year now and I believe this issue started a day or two ago).

Uploads are fine for an instant-- it says it's hovering around 3 - 5Mb/s-- but then immediately crashes down to around 60 Kib/s.

Uploading a 3 MB file took long enough for me to google other peoples' concerns in this forum, find no solutions, and then register an account and start typing up this reply before the file finished uploading.

I've already tried reformatting my SD card with 64k clusters but no success; same behavior as before.

Running M122 command immediately after upload gives the following data for my SD card:

"SD card 0 detected, interface speed: 20.0MBytes/sec
SD card longest block write time: 184.6ms"

Any thoughts/suggestions?

dc42

@clcollins4 said in Slow upload:

SD card longest block write time: 184.6ms"

Replace the SD card.

sigxcpu

Just uploaded a big file (15.6MB) and got this:

SD card longest block write time: 660.7ms, max retries 0

M39 reports:

SD card in slot 0: capacity 15.93Gb, free space 15.89Gb, speed 20.00MBytes/sec, cluster size 64kb

The upload hovered between 500 and 700 KB/sec. I am not sure that value is so relevant.

I've seen short stalls un upload, but then it resumed at the mentioned speed.

Maybe some cards have more stalls, so the total speed is reduced much more?

I need to mention that I've replaced original unbranded Duet card (4GB) with this one (SandDisk UHS something) after reading this topic and the old one was a little better. This one has a pause when it starts the upload then continues fine.

dc42

@sigxcpu said in Slow upload:

SD card longest block write time: 660.7ms, max retries 0

I haven't seen a block write time as long as that before. Does M122 show any SD card retries?

As SD cards get older, they reach the point at which some of the storage on the card has been written several times, and they start to do wear levelling. This involves copying blocks of data around and/or writing data multiple times. The result is that some write operations start to take much longer.

core3d.tech

I just received my brand new SandDisk 32Gb UHS-I class 10, so here's my latest. It's not good news

Upload speed out of box: ~110KiB/s (not as slow as before but hardly impressive)

M122
=== Diagnostics ===
RepRapFirmware for Duet 2 WiFi/Ethernet version 2.0(RTOS) running on Duet WiFi 1.02 or later
Board ID: 08DGM-9T6BU-FG3S0-7J9FJ-3SD6P-KS57F
Used output buffers: 1 of 20 (13 max)
=== RTOS ===
Static ram: 28380
Dynamic ram: 95872 of which 16 recycled
Exception stack ram used: 332
Never used ram: 6472
Task NETWORK ready, free stack 396
Task HEAT blocked, free stack 1256
Task MAIN running, free stack 3568
=== Platform ===
Last reset 00:03:37 ago, cause: power up
Last software reset at 2018-11-30 14:12, reason: User, spinning module GCodes, available RAM 6472 bytes (slot 0)
Software reset code 0x0003 HFSR 0x00000000, CFSR 0x00000000, ICSR 0x0441f000, BFAR 0xe000ed38, SP 0xffffffff
Error status: 0
Free file entries: 10
SD card 0 detected, interface speed: 20.0MBytes/sec
SD card longest block write time: 34.7ms
MCU temperature: min 23.9, current 31.5, max 31.6
Supply voltage: min 22.2, current 24.2, max 24.8, under voltage events: 0, over voltage events: 0
Driver 0: standstill, SG min/max not available
Driver 1: standstill, SG min/max not available
Driver 2: standstill, SG min/max not available
Driver 3: standstill, SG min/max not available
Driver 4: standstill, SG min/max not available
Date/time: 2018-11-30 15:04:37
Slowest loop: 4.36ms; fastest: 0.07ms
=== Move ===
Hiccups: 0, StepErrors: 0, LaErrors: 0, FreeDm: 240, MinFreeDm 240, MaxWait: 0ms, Underruns: 0, 0
Scheduled moves: 0, completed moves: 0
Bed compensation in use: none
Bed probe heights: 0.000 0.000 0.000 0.000 0.000
=== Heat ===
Bed heaters = 0 -1 -1 -1, chamberHeaters = -1 -1
=== GCodes ===
Segments left: 0
Stack records: 1 allocated, 0 in use
Movement lock held by null
http is idle in state(s) 0
telnet is idle in state(s) 0
file is idle in state(s) 0
serial is idle in state(s) 0
aux is idle in state(s) 0
daemon is idle in state(s) 0
queue is idle in state(s) 0
autopause is idle in state(s) 0
Code queue is empty.
=== Network ===
Slowest loop: 109.64ms; fastest: 0.01ms
Responder states: HTTP(0) HTTP(0) HTTP(0) HTTP(0) FTP(0) Telnet(0) Telnet(0)
HTTP sessions: 1 of 8

WiFi -
Network state is running
WiFi module is connected to access point
Failed messages: pending 0, notready 0, noresp 0
WiFi firmware version 1.21
WiFi MAC address 84:f3:eb:41:95:b3
WiFi Vcc 3.38, reset reason Turned on by main processor
WiFi flash size 4194304, free heap 15192
WiFi IP address 192.168.1.177
WiFi signal strength -27dBm, reconnections 0, sleep mode modem
Socket states: 0 0 0 0 0 0 0 0
=== Expansion ===

Next I reformatted at with FAT32 64Kb (not sure what it was before) Uploaded the same file
Speed down to below 100KiB/S

M122
=== Diagnostics ===
RepRapFirmware for Duet 2 WiFi/Ethernet version 2.0(RTOS) running on Duet WiFi 1.02 or later
Board ID: 08DGM-9T6BU-FG3S0-7J9FJ-3SD6P-KS57F
Used output buffers: 3 of 20 (8 max)
=== RTOS ===
Static ram: 28380
Dynamic ram: 95872 of which 16 recycled
Exception stack ram used: 332
Never used ram: 6472
Task NETWORK ready, free stack 396
Task HEAT blocked, free stack 1256
Task MAIN running, free stack 3568
=== Platform ===
Last reset 00:02:48 ago, cause: power up
Last software reset at 2018-11-30 14:12, reason: User, spinning module GCodes, available RAM 6472 bytes (slot 0)
Software reset code 0x0003 HFSR 0x00000000, CFSR 0x00000000, ICSR 0x0441f000, BFAR 0xe000ed38, SP 0xffffffff
Error status: 0
Free file entries: 10
SD card 0 detected, interface speed: 20.0MBytes/sec
SD card longest block write time: 36.6ms
MCU temperature: min 27.7, current 32.9, max 32.9
Supply voltage: min 24.0, current 24.2, max 24.3, under voltage events: 0, over voltage events: 0
Driver 0: standstill, SG min/max not available
Driver 1: standstill, SG min/max not available
Driver 2: standstill, SG min/max not available
Driver 3: standstill, SG min/max not available
Driver 4: standstill, SG min/max not available
Date/time: 2018-11-30 15:10:19
Slowest loop: 5.26ms; fastest: 0.07ms
=== Move ===
Hiccups: 0, StepErrors: 0, LaErrors: 0, FreeDm: 240, MinFreeDm 240, MaxWait: 0ms, Underruns: 0, 0
Scheduled moves: 0, completed moves: 0
Bed compensation in use: none
Bed probe heights: 0.000 0.000 0.000 0.000 0.000
=== Heat ===
Bed heaters = 0 -1 -1 -1, chamberHeaters = -1 -1
=== GCodes ===
Segments left: 0
Stack records: 1 allocated, 0 in use
Movement lock held by null
http is idle in state(s) 0
telnet is idle in state(s) 0
file is idle in state(s) 0
serial is idle in state(s) 0
aux is idle in state(s) 0
daemon is idle in state(s) 0
queue is idle in state(s) 0
autopause is idle in state(s) 0
Code queue is empty.
=== Network ===
Slowest loop: 111.20ms; fastest: 0.01ms
Responder states: HTTP(0) HTTP(0) HTTP(0) HTTP(0) FTP(0) Telnet(0) Telnet(0)
HTTP sessions: 1 of 8

WiFi -
Network state is running
WiFi module is connected to access point
Failed messages: pending 0, notready 0, noresp 0
WiFi firmware version 1.21
WiFi MAC address 84:f3:eb:41:95:b3
WiFi Vcc 3.38, reset reason Turned on by main processor
WiFi flash size 4194304, free heap 15168
WiFi IP address 192.168.1.177
WiFi signal strength -28dBm, reconnections 0, sleep mode modem
Failed to request ESP stats
Socket states: 0 0 0 0 0 0 0 0
=== Expansion ===

I'm not really sure this is a card issue

M39
SD card in slot 0: capacity 31.91Gb, free space 31.89Gb, speed 20.00MBytes/sec, cluster size 64kb

Phaedrux

@core3d-tech Are you using a router that supports both 2.4Ghz and 5Ghz networks? If you can, try only enabling the 2.4Ghz network and test again.

Does the router have any traffic control functions enabled?

Can you disable IPv6 on the router and sending PC?

core3d.tech

@phaedrux turned off 5Ghz, turned off Qos (only traffic control I'm aware of), turned off ipv6 both on asus as well as network sharing adapter settings. Upload speed still around ~100KiB/s. Did not seem to do the trick

stewwy

I've been having slow uploads for some time.

The cure for it in my case was to rotate the printer 90 degrees.

2.4 mhz is quite sensitive to antenna orientation and position

core3d.tech

Some development here. I just removed my current ASUS router (2 feet away) from the duet and added my NETGEAR router (on that is on the other side of the house, 1 wall, 1 ceiling, 50 feet away).

Upload tripled in speed running over 350KiB/s (still not 600 but much better). Clearly there's something it doesn't like about the ASUS setup.

Asus Router: RT-AC68U
Netgear: R6700 Nighthawk

M122
=== Diagnostics ===
RepRapFirmware for Duet 2 WiFi/Ethernet version 2.0(RTOS) running on Duet WiFi 1.02 or later
Board ID: 08DGM-9T6BU-FG3S0-7J9FJ-3SD6P-KS57F
Used output buffers: 3 of 20 (9 max)
=== RTOS ===
Static ram: 28380
Dynamic ram: 95872 of which 16 recycled
Exception stack ram used: 332
Never used ram: 6472
Task NETWORK ready, free stack 396
Task HEAT blocked, free stack 1256
Task MAIN running, free stack 3584
=== Platform ===
Last reset 00:08:59 ago, cause: watchdog
Last software reset at 2018-11-30 14:12, reason: User, spinning module GCodes, available RAM 6472 bytes (slot 0)
Software reset code 0x0003 HFSR 0x00000000, CFSR 0x00000000, ICSR 0x0441f000, BFAR 0xe000ed38, SP 0xffffffff
Error status: 0
Free file entries: 10
SD card 0 detected, interface speed: 20.0MBytes/sec
SD card longest block write time: 37.6ms
MCU temperature: min 23.7, current 31.0, max 31.2
Supply voltage: min 0.0, current 24.2, max 25.0, under voltage events: 0, over voltage events: 0
Driver 0: standstill, SG min/max not available
Driver 1: standstill, SG min/max not available
Driver 2: standstill, SG min/max not available
Driver 3: standstill, SG min/max not available
Driver 4: standstill, SG min/max not available
Date/time: 2018-12-01 10:27:43
Slowest loop: 38.37ms; fastest: 0.06ms
=== Move ===
Hiccups: 0, StepErrors: 0, LaErrors: 0, FreeDm: 240, MinFreeDm 240, MaxWait: 0ms, Underruns: 0, 0
Scheduled moves: 0, completed moves: 0
Bed compensation in use: none
Bed probe heights: 0.000 0.000 0.000 0.000 0.000
=== Heat ===
Bed heaters = 0 -1 -1 -1, chamberHeaters = -1 -1
=== GCodes ===
Segments left: 0
Stack records: 1 allocated, 0 in use
Movement lock held by null
http is idle in state(s) 0
telnet is idle in state(s) 0
file is idle in state(s) 0
serial is idle in state(s) 0
aux is idle in state(s) 0
daemon is idle in state(s) 0
queue is idle in state(s) 0
autopause is idle in state(s) 0
Code queue is empty.
=== Network ===
Slowest loop: 110.44ms; fastest: 0.01ms
Responder states: HTTP(0) HTTP(0) HTTP(0) HTTP(0) FTP(0) Telnet(0) Telnet(0)
HTTP sessions: 1 of 8

WiFi -
Network state is running
WiFi module is connected to access point
Failed messages: pending 0, notready 0, noresp 0
WiFi firmware version 1.21
WiFi MAC address 84:f3:eb:41:95:b3
WiFi Vcc 3.38, reset reason Turned on by main processor
WiFi flash size 4194304, free heap 15008
WiFi IP address 192.168.1.177
WiFi signal strength -69dBm, reconnections 0, sleep mode modem
Socket states: 0 0 0 0 0 0 0 0
=== Expansion ===

stewwy

The sub-optimal aerials on most routers ( so they look nice and neat ) tend to have blindspots in surprising places, multiple aerials can interact and make this worse when close by

dc42

Interesting, I am using an Asus DSL-AC68U router and I get 750 to 800kbyes/sec upload speed for a 10Mbyte file. WiFi signal strength is -51dBm.

core3d.tech

@dc42 yes especially considering the Netgear that runs faster and is on the other side of the house runs through the ethernet port of my ASUS. Not sure what that means but it seems that it's not a soft traffic throttling setting on the ASUS. I will try the antenna thing mentioned earlier, tomorrow. Bit skeptical about that one

dc42

It may also be worthwhile trying a different WiFi channel.

core3d.tech

@dc42 just switched back to Asus and tried different channels. before it it was 3, I tried, 2 which had no change, tried channel 4 which got better but still sub 200KiB/s. Tried 7 and again baqd news. Channels do seem to make a difference but still nothing that say, yep fixed it.

Also tried pointing the antennas in different directions (to no avail).

core3d.tech

last update on this. My internet was just out and as part of the usual exercise I turned off modem and router and now that everything is back up speed is in the 600s. I have turned off the router before but this time seemed to have done the magic

gnydick

Just curious, what is the expected network throughput on the Maestro? I get 660KB/sec (I think it's B and not b) so roughly 5mbit. The SD card I have is super fast when plugged into my computer, so just wondering if the network interface on board is the bottleneck or the SD card drive on the board?

dc42

@gnydick said in Slow upload:

Just curious, what is the expected network throughput on the Maestro? I get 660KB/sec (I think it's B and not b) so roughly 5mbit. The SD card I have is super fast when plugged into my computer, so just wondering if the network interface on board is the bottleneck or the SD card drive on the board?

If you are running firmware 2.02 then you can send M122 P104 to measure the SD card write speed. The optional S parameter is the file size in Mbytes, default 10. I've just measured it on my Crane Quad and it reports 0.99Mbytes/sec for a 10Mb file and 1.23mb/sec for a 40Mb file. The firmware doesn't yet read network data and write to the SD card concurrently, so the upload speed will be slower. I just measured it at 510kbytes/sec for a 14Mbyte file.

You may get a better upload speed if you use a better SD card.

There are two reasons why writing to the SD card may appear to be faster on your PC:

Your PC has much more RAM, so it can write much larger chunks of data, which helps maintain write speed;
When you copy a file, at the point at which the write appears to have finished, there may be data still being written from the cache to the card. If you copy large files to the SD card on a PC and then ask to eject the card as soon as the copy appears to be finished, it may be a little while before the PC tells you it is safe to remove the card.

resam

My results:
SD write speed for 10.0Mbyte file was 1.11MBytes/sec
and
SD write speed for 40.0Mbyte file was 1.14MBytes/sec

With typical uploads speeds also in the 500-600kB/s range.
WiFi signal strength -69dBm
SD card longest block write time: 72.5ms, max retries 0

gnydick

@dc42 I totally know why the PC is faster, I was just letting you know it is a fast card, so it isn't the bottleneck