ASROCK H110 BTC Pro+ causes network problems

Author

Message

Topic Search

Topic Options

SPAPPA

Newbie

Joined: 15 Feb 2018
Status: Offline
Points: 2

Post Options

Post Reply

Quote SPAPPA

Report Post

Thanks(0)

Quote

Topic: ASROCK H110 BTC Pro+ causes network problems
Posted: 16 Mar 2018 at 10:49pm

Solved with the last bios update 1.50 Wink

The rig run 48h without crash/network issue Clap

wardog

Moderator Group

Joined: 15 Jul 2015
Status: Offline
Points: 6447

Post Options

Post Reply

Quote wardog

Report Post

Thanks(0)

Quote

Posted: 10 Mar 2018 at 11:17pm

Zandar wrote:

Thanks for the tip, i have wireshark installed and i did a quick test. 4 MB for 30 seconds of network traffic, is that even possible on a home network with ~40 nodes? i will need to set this up to save the logs to a spare 4TB drive i am not using.

it will take me some time to read the guide and learn wireshark. its an awesome tool but packet sniffing is not my speciality.

Save your time. I got a handle on what's happening in my post above.

I feel secure in saying you can uninstall WireShark and not worry using it.

wardog

Moderator Group

Joined: 15 Jul 2015
Status: Offline
Points: 6447

Post Options

Post Reply

Quote wardog

Report Post

Thanks(0)

Quote

Posted: 10 Mar 2018 at 11:14pm

After some digging I must say I'm leaning towards this being possibly driver related or network infrastructure "faulty'.

MPCP - Multiple Point Control Protocol

The below page describes to a tee what is affecting ALL the Users here in this thread
https://github.com/nwholloway/mpcp

.

Everyone here, are the firmware in you network switches/routers/AP's updated?

Maybe the lot of you can post the NIC driver revision here to compare also.

.... wardog

Zandar

Newbie

Joined: 10 Mar 2018
Location: USA
Status: Offline
Points: 2

Post Options

Post Reply

Quote Zandar

Report Post

Thanks(0)

Quote

Posted: 10 Mar 2018 at 8:57pm

Thanks for the tip, i have wireshark installed and i did a quick test. 4 MB for 30 seconds of network traffic, is that even possible on a home network with ~40 nodes? i will need to set this up to save the logs to a spare 4TB drive i am not using.

it will take me some time to read the guide and learn wireshark. its an awesome tool but packet sniffing is not my speciality.

wardog

Moderator Group

Joined: 15 Jul 2015
Status: Offline
Points: 6447

Post Options

Post Reply

Quote wardog

Report Post

Thanks(0)

Quote

Posted: 10 Mar 2018 at 2:57pm

Grab WireShark or it's equivalent and tell us what "data' is flooding the network.

Otherwise you're just guessing. Nicely I say that, 'k?

Zandar

Newbie

Joined: 10 Mar 2018
Location: USA
Status: Offline
Points: 2

Post Options

Post Reply

Quote Zandar

Report Post

Thanks(0)

Quote

Posted: 10 Mar 2018 at 1:04pm

Same problem and situation .

I have 3 of these rigs, 2 are completed builds and one is still in progress. ALL SMOS, DSTM 5.7, Zcash Flypool

2 Molex - each on own rail. Same PSU which powers the Mainboard. 1 SATA own rail same PSU as described in their guide.

Rig1. Online Dec 2017 , 12 X EVGA 1060 SC, 1000 EVGA ; 750 EVGA; 4 GB Micron Ram, Celeron

Rig2. Online January 20 1x 1080ti, 3x 1070TI, 2x EVGA 1060 OC, 6x EVGA 1050TI, 1300 EVGA, 850 EVGA 4GB Micron Ram , Celeron

Rig3. 3x 1080 , 2x 1060 (in progress) , 1000 EVGA, 1000 EVGA, 4GB , Celeron

Cat 6 ethernet run to all, 1 Gigabit Verizon FIOS,

Rig1 - the original is the only one that (seems) to crash. But when it does, it floods the network and the entrire house goes offline.

I have also tried VLANs and that doesnt work. The Thing is Rig 2 has never crashed. Not once. Rig 3 is a build in progress. no helpful data. I wish i could add more data, but i can say there has to be a fix for this. I am thinking about trying HiveOS if only on the main offender.

I cannot help but believe power is an issue here. I have changed all the risers on Rig1. perhaps relevant- i have noticed that if i put one of the rigs on a different coin / pool the crash does not occur (ran without crash for a month)

Roaders

Newbie

Joined: 07 Mar 2018
Status: Offline
Points: 1

Post Options

Post Reply

Quote Roaders

Report Post

Thanks(0)

Quote

Posted: 07 Mar 2018 at 5:21am

Seems very similar to issues that I have had as well.

I am running Claymore Miner on Ubuntu 16 and anything attached to the same switch as the affected machine no longer has network access.

I have detailed the issue in more detail here:

https://askubuntu.com/questions/994466/ubuntu-server-flooding-network-with-mpcp-opcode-pause-packets

I have had 4 of these motherboards and this issue has affected all of them.

SPAPPA

Newbie

Joined: 15 Feb 2018
Status: Offline
Points: 2

Post Options

Post Reply

Quote SPAPPA

Report Post

Thanks(0)

Quote

Posted: 15 Feb 2018 at 11:37pm

Same Problem.

11x GTX1050 ti

Os: SimpleMining(Kill Router)

Os: HiveOs(Kill Router)

I think that the problems is th motherboard Unhappy

digituv

Newbie

Joined: 20 Jan 2018
Status: Offline
Points: 1

Post Options

Post Reply

Quote digituv

Report Post

Thanks(0)

Quote

Posted: 20 Jan 2018 at 9:05am

Have Same Problemes. Have 21 AsRock motherboards. With 12 1060 Cards. on Every Farm.

And All they kill internet network and kill router. And all is work well when disconnect farm from network.

Anybody fix this problem? OS SimpleMining

John Small View Drop Down

Newbie

Joined: 03 Jan 2018
Location: UK
Status: Offline
Points: 6

Post Options

Post Reply

Quote John Small

Report Post

Thanks(0)

Quote

Posted: 09 Jan 2018 at 7:21pm

This report is also at https://www.reddit.com/r/EtherMining/comments/6xkau1/simplemining_os_with_5_or_6_gtx_1070s_makes/

Along with other people who are experiencing the same problems.

Symptom: mining with this board causes network problems with wifi access points

Full report:

My setup, Ubuntu 16.04, ASROCK H110 Pro BTC+, Draytek Vigor 2860ac VDSL router to internet, Draytek AP-902 access point. 8 GPUs, 4 GTX 1070, 4 GTX 1080. 3 other rigs connected to the AP. Mining Zcash with EWBF to flypool.

First try, the H110 rig connected to the AP-902. Kills the AP connection to the Vigor router. Though I can access the AP, and the other rigs from the H110 rig perfectly OK while it's attempting to mine. Also tried the DSTM miner, same result.

Second try. Using a newly purchased Realtek USB wifi on a different channel to the one between the AP-902 and the Vigor 2860. Kills the AP to Vigor as soon at it starts mining. WTF!!!

Third try. Configure a completely separate VLAN and SSID for the H110 to use, configured so each connection is isolated so it should not be able to affect the AP. Kills the AP <-> Vigor connection as soon as it starts mining. WTF!!!

Even more bizarre is that other devices on wifi on the same channel as the AP<->Vigor, e.g an IPad are not affected.

Last try. Turn off mining on the other 3 rigs. The AP stills dies as soon as I start mining on the H110.

Conclusion. The ASROCK H110 is not fit for the purpose advertised.