All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Anyone hosted with MicrotronixDC? Slow upload issues?
Hey guys,
I have a colocation with MicrotronixDC and everything else is working great, but I'm running into a frustrating issue with slow uploads. Maybe someone here has experienced something similar?
My setup:
Dell PowerEdge R630 server
Broadcom Inc. and subsidiaries NetXtreme BCM5720 Gigabit Ethernet PCIe network card
Proxmox 8 and Proxmox 9 (tried both)
The issue:
No matter what I do, my upload speeds are really slow. I've tested with different operating systems, not just Proxmox, and the problem persists. Downloads are fine, but uploads are consistently bottlenecked.
Has anyone else hosted at MicrotronixDC with similar hardware and experienced upload speed issues? Any ideas what could be causing this? Could it be something with the network card drivers, the datacenter's network configuration, or am I missing something obvious?
Comments
Someone's colo'ed microwave oven caused electromagnetic interference, so that your upload line has been fried.
@jfreak53 will soon prohibit microwave ovens from being colo'ed.
This card is E waste.
We recommend Intel XXV710 or Mellanox ConnectX-5, paired with dedicated 10Gbps port.
Thank @yoursunny for the response. I have tested everything download is also normal, and browsing is also good, major problem comes at upload 10MB file take 20 mins for the upload.
Right now, I do not have 10GBPS Port connected to hardware. It just supports 1GPS over RJ45.
I tried so many other linux os and tried many ways to upload but its very slow in all type of OS.
RJ45 is always slow.
You need 10Gbps Ethernet adapter and SFP+ port.
Will this work.
Dell 5N7Y5 Intel X710-DA2 Dual Port 10Gb SPF PCIe Network Adapter Card
Ask them to swap the network cable.
They swapped the network cables also I think
Don't listen to @yoursunny, he's only trolling you.
We have a server with them and have constantly good speeds both ways, even from Europe where I'm sitting.
From where are you trying to upload files?
What does nws.sh and yabs.sh say?
This is the result url https://result.nws.sh/r/1765456771_Q3J357_GLOBAL.txt
iperf3 Network Speed Tests (IPv4):
This is the other result.
Well that is a bit low..
On our server:
I managed to get around 150Mbit from my homeconnection to the server also, no complaints from here.
I'd say to just continue working with them, @jfreak53 has been very nice for us at least.
@daffy what is the NIC card type ?
Please stop discrediting us.
What version firmware is the network card using?
Might be worth looking into a firmware update since its a Dell server the firmware should be easily available too.
ethtool -i eno1
driver: tg3
version: 6.12.0-55.12.1.el10_0.x86_64
firmware-version: FFV22.00.6 bc 5720-v1.39
expansion-rom-version:
bus-info: 0000:01:00.0
supports-statistics: yes
supports-test: yes
supports-eeprom-access: yes
supports-register-dump: yes
supports-priv-flags: no
sudo ethtool eno1 to see if you are in full duplex.
To check for errors:
ip -s link show dev eno1
sudo ethtool -S eno1 | egrep 'err|drop|coll|fifo|timeout'
ethtool -S eno1 | egrep 'err|drop|coll|fifo|timeout'
rx_fcs_errors: 0
rx_align_errors: 0
rx_frame_too_long_errors: 0
rx_in_length_errors: 0
rx_out_length_errors: 0
tx_collisions: 0
tx_mac_errors: 0
tx_single_collisions: 0
tx_mult_collisions: 0
tx_deferred: 0
tx_excessive_collisions: 0
tx_late_collisions: 0
tx_collide_2times: 0
tx_collide_3times: 0
tx_collide_4times: 0
tx_collide_5times: 0
tx_collide_6times: 0
tx_collide_7times: 0
tx_collide_8times: 0
tx_collide_9times: 0
tx_collide_10times: 0
tx_collide_11times: 0
tx_collide_12times: 0
tx_collide_13times: 0
tx_collide_14times: 0
tx_collide_15times: 0
tx_carrier_sense_errors: 0
tx_errors: 0
rx_errors: 0
ip -s link show dev eno1
2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
link/ether 14:18:77:4a:5b:ab brd ff:ff:ff:ff:ff:ff
RX: bytes packets errors dropped missed mcast
3350156 51541 0 0 0 4404
TX: bytes packets errors dropped carrier collsns
0 0 0 0 0 0
altname enp1s0f0
altname enx1418774a5bab
ethtool eno1
Settings for eno1:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Supported pause frame use: No
Supports auto-negotiation: Yes
Supported FEC modes: Not reported
Advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Advertised pause frame use: No
Advertised auto-negotiation: Yes
Advertised FEC modes: Not reported
Link partner advertised link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Half 1000baseT/Full
Link partner advertised pause frame use: No
Link partner advertised auto-negotiation: Yes
Link partner advertised FEC modes: Not reported
Speed: 1000Mb/s
Duplex: Full
Auto-negotiation: on
Port: Twisted Pair
PHYAD: 1
Transceiver: internal
MDI-X: off
Supports Wake-on: g
Wake-on: d
Current message level: 0x000000ff (255)
drv probe link timer ifdown ifup rx_err tx_err
Link detected: yes
Is this a rented server?
Thanks @daffy glad everything is working for you!
The issue is software or hardware related, meaning firmware or NIC card. We've swapped your cable and port on the switch. The switch you are on also has another 30 devices on it, without issues, so we know its not the switch, and its also only using about 3% of its BW capacity, so that's not it.
Usually, when its DC or upline, its slow in both directions up/down (same when its cable or port), unless its routing, but even then usually routing is percentage-wise slow comparatively.
What you've experienced from the beginning is port max or close to it download and slow upload, even with iperf. This usually indicates software issue on the server or NIC card issues. Once we rule out our cabling, port, and switch, in this instance, it only leaves the customer's equipment.
Our first test when a customer has these issues is to test from the rack you are in to cloudflare's iperf server, per the knowledgebase article I sent you. Instructions for you to test, we do the same from one of our servers on that switch. In this case we have done this, which means it's not internal equipment or routing, as well not DC routing issue to cloudflare iperf servers. We maxed out a 10G port on that switch testing to cloudflare's iperf server.
By reason of deduction, the only piece of the puzzle left is customer-facing equipment or software. If you have tried multiple OS' that leaves you with firmware, NIC card, or hardware bottleneck.
However, to test further you must follow the knowledgebase article I sent you, we have yet to be sent back the results from all the tests in that article. That article is there to make it easier for us to track down issues and not spend days back and forth asking for more info. I will say though, I think the issue is your card, so I feel the tests would result in the same conclusion and you would be back at the card being the issue.
Nah, I took the microwave out last week, it was causing interference when I was trying to watch the Michigan/OSU game, can't have that! Now, it could be caused by the nuclear power generator sitting in Rack 5 in that row
hadn't thought of that, maybe we'll move it. 
X710 should be okay.