Pages:
Author

Topic: HashFast Sierra's Owners Thread (Read 23805 times)

legendary
Activity: 2856
Merit: 1520
Bitcoin Legal Tender Countries: 2 of 206
July 25, 2015, 03:54:11 PM
Hey Mr. Teal / et. al.,

I happen to be in possession of a number of Sierra machines of recent purchase.  They arrived, I spun up their PSUs by using the power button on the PSU and nothing happened.  So ... I decided to press one of the buttons on each of the three boards in the case and viola the CoolT fans started spinning adn I felt certain there was 12VDC on the board.  It seemd to me thsi should have caused ther USB device to spin up as well.  I felt good about this part.  So, I followed the "compiling cgminer for HashFast stuff" guide at http://hashfast.com/wp-content/uploads/2014/05/Compiling-CG-Miner-to-show-multiple-die-temps.pdf and thought I was about to have me some mining success.  This was not the case.  It seems cgminer cannot see the sierra boards (each is USB connected and not chained).  I have tried to connect the Sierra to a USB3 hub and a USB2 hub and I get the same outcome ... nothing shows up in dmesg (oh yea ... I'm running Ubuntu 14.04LTS on said MiniITX / shuttle machine) *AND* thus nothing gets seen by cgminer.  I am using a MininItx form-factor PC.  The wierd part of all of this is tha my BFL SC devices adn Jalos get seen and my BE USB devices get seen.  Could it be that I'm getting borked by said MiniItx machine?  If I followed the HashFast guide set-by-step shouldn't I see  predictable outcome / the outcome HashFast described?  Any amount of your wisdom is HUGELY appreciated.

H@shKraker
If you type lsusb, do you see the devices?
Also, you might want to try with a vanilla version of cgminer first just to ensure it's not an issue with the build of cgminer.

Mr Teal,

Many thanks for your reply.  I surely do wish these HashFast units were as bullet proof as you PepperMining units.  Back on subject now.  When I type sudo lsusb I do not see any HashFast devices even though I can hear the Sierra fans spinning like turbine engines.  Also, I've used both the HashFast cgminer git pull *AND* the latest standard cgminer git pull.  I have a standard ATX Intel-I7 machine running some Habaneros right now sooooooo .... I brought one of the Sierras to that same mining location and am gonna connect it up with the Habaneros and see if  I can eliminate the mini-itx form factor / USB hardware as the culprit.

H@shKraker

can someone provide this pdf file above in the quote please?
newbie
Activity: 12
Merit: 0
January 18, 2015, 10:09:18 PM
Hi, I am trying to help a friend who has a Seirra he bought but was never able to get working.  He bought one back in April with someone else and that guy flaked on him and he has no tech expertise and was basically stuck with it until he met me now.  I dont think a Seirra is even close to profitable now but I told him I would try to help him get it working anyways.

can anyone provide me with the latest firmware for the sierra?  I assume his has the original firmware since he has had it sitting around for almost a year.  I am able to get it working and hashing at 1.2TH/s but it constantly fails after 1 - 20 minutes with lots of USB errors.  Sometimes it will start and stop in less than a minute with USB errors and sometimes it will run for almost 20 minutes before the errors happen and cgminer crashes.   I am hoping new firmware will help.  Hashfast seems defunct or close to it, their setup.hashfast.com page doesnt work for me (i just get their website), I cant register for their site and they wont respond to support requests.

Thanks much.


Can you post the logs.
newbie
Activity: 12
Merit: 0
January 18, 2015, 09:42:15 PM
At present I am running Hashfast Sierra Batch 2. on Ubuntu.

A couple of months ago I had an issue with 3 rd top the coolers . I replaced and it fixed the issue. 2 Days ago I have the similar issue with over heating . I replaced the fan , but after installing, I get the following issue.

Here is the error on console.
HFB 0 NOTICE: Hash Core Error: Still emitting hashes: die 0 core 0, count: 1300
HFB 0 NOTICE: Hash Core Error: Still emitting hashes: die 0 core 0, count: 1400
HFB 0 NOTICE: Hash Core Error: Still emitting hashes: die 0 core 0, count: 1500
HFB 0 NOTICE: Hash Core Error: Still emitting hashes: die 0 core 0, count: 1600

If I take out the third card and run the other 2 cards it Runs fine.

Any Ideas would be appreciated .

newbie
Activity: 35
Merit: 0
January 14, 2015, 11:05:53 AM
Hi, I am trying to help a friend who has a Seirra he bought but was never able to get working.  He bought one back in April with someone else and that guy flaked on him and he has no tech expertise and was basically stuck with it until he met me now.  I dont think a Seirra is even close to profitable now but I told him I would try to help him get it working anyways.

can anyone provide me with the latest firmware for the sierra?  I assume his has the original firmware since he has had it sitting around for almost a year.  I am able to get it working and hashing at 1.2TH/s but it constantly fails after 1 - 20 minutes with lots of USB errors.  Sometimes it will start and stop in less than a minute with USB errors and sometimes it will run for almost 20 minutes before the errors happen and cgminer crashes.   I am hoping new firmware will help.  Hashfast seems defunct or close to it, their setup.hashfast.com page doesnt work for me (i just get their website), I cant register for their site and they wont respond to support requests.

Thanks much.

legendary
Activity: 1274
Merit: 1004
December 01, 2014, 02:23:30 AM
I have developed a way to determine easily which one it is as its hard to tell.

Care to elaborate?

How much does a replacement cooling block cost? I wonder if it's even worth it at todays difficulty
It's an H80i (with a standard fan connector instead of the SATA on the H80i), but most any water cooler will work.
legendary
Activity: 1112
Merit: 1000
December 01, 2014, 01:09:02 AM
I have developed a way to determine easily which one it is as its hard to tell.

Care to elaborate?

[edit]

How does one find out which of the three boards is causing the problem?

[/edit]

How much does a replacement cooling block cost? I wonder if it's even worth it at todays difficulty
sr. member
Activity: 446
Merit: 250
November 30, 2014, 07:11:22 PM
LARRY didn't want to run for more then 2 secs... so i lowered the clocks settings down to 300 (--hfa-hash-clock 300) and LARRY decided to hash for more the 10mins. After which I kept upping the MHz by 50 every 10 mins, all the way to 604 (--hfa-hash-clock 604).

Hey,

I have 2 hfa devices that have been running in a colo for months without any problems.

Running cgminer 4.8 these days now

One system decided to throw in the towel last week with the same error, overheating.

If one starts up cgminer the first time it states:

Code:
HFS hfsb: Hit overheat temp 96.1, throttling!
HFS hfsb: Thermal overload tripped! Shutting down device

(the following times you launch cgminer it just mentions

Code:
HFS hfsb: Thermal overload tripped! Shutting down device

and no additional info. Only if you power it down for a while it cools down enough)

I only managed to keep it running if I set the freq to 125, where the temp is hovering around 98 degrees and it complains all the time

Code:
HFS hfsb: Hit overheat temp 95.3, throttling!

or similar temperatures above 95 degrees

I guess I have a cooling issue (bad heatsink, all three fans are running at --hfa-fan 100 and do pass a load of air, or a bad temperature sensor).

How does one isolate the problem to one of the three boards?

Do I have to hook up each board separately like in the BabyJet?

Or could this error be the symptom of having a bad power supply (I guess that is the next suspect on my list)?

I've had this happen in excess of 21 times on different Sierras. Its the water block. The pump is failing or the fluid has leaked out.

I have purchased a backup supply of these water coolers. They are the exact ones hashfast used. If you need one I can sell you one. Just PM me. btw, its almost always just one of the coolers, there are 3 in the sierra. I have developed a way to determine easily which one it is as its hard to tell.
legendary
Activity: 1112
Merit: 1000
November 30, 2014, 06:51:12 PM
LARRY didn't want to run for more then 2 secs... so i lowered the clocks settings down to 300 (--hfa-hash-clock 300) and LARRY decided to hash for more the 10mins. After which I kept upping the MHz by 50 every 10 mins, all the way to 604 (--hfa-hash-clock 604).

Hey,

I have 2 hfa devices that have been running in a colo for months without any problems.

Running cgminer 4.8 these days now

One system decided to throw in the towel last week with the same error, overheating.

If one starts up cgminer the first time it states:

Code:
HFS hfsb: Hit overheat temp 96.1, throttling!
HFS hfsb: Thermal overload tripped! Shutting down device

(the following times you launch cgminer it just mentions

Code:
HFS hfsb: Thermal overload tripped! Shutting down device

and no additional info. Only if you power it down for a while it cools down enough)

I only managed to keep it running if I set the freq to 125, where the temp is hovering around 98 degrees and it complains all the time

Code:
HFS hfsb: Hit overheat temp 95.3, throttling!

or similar temperatures above 95 degrees

I guess I have a cooling issue (bad heatsink, all three fans are running at --hfa-fan 100 and do pass a load of air, or a bad temperature sensor).

How does one isolate the problem to one of the three boards?

Do I have to hook up each board separately like in the BabyJet?

Or could this error be the symptom of having a bad power supply (I guess that is the next suspect on my list)?
sr. member
Activity: 403
Merit: 250
November 22, 2014, 11:25:57 PM
AhHAAA .... looking at dmesg I noticed TONS of error messages saying the USB controller was constantly going off line, restarting, experiencing I.O errors, going offline ... lather, rinse repeat ....  Shocked  Me thinks it's time to scavenge the CPU, SSD, DVD-R and build a new full ATX machine as opposed to a mini-itx.

H@shKraker
legendary
Activity: 1274
Merit: 1004
November 22, 2014, 05:48:35 PM
I wonder .... could it be that Ubuntu 14.04LTS isn't playing nice?

H@shKraker
Possibly. I would try plugging them into another device to see if they show up there.
sr. member
Activity: 403
Merit: 250
November 22, 2014, 03:19:49 PM
I wonder .... could it be that Ubuntu 14.04LTS isn't playing nice?

H@shKraker
sr. member
Activity: 403
Merit: 250
November 22, 2014, 02:43:15 PM
Mr. Teal,

I plugged a Sierra into the same USB hub that my Habanero machines are plugged into and cgminer successfully saw the Habanero units but DID NOT see the Sierra.  This tells me that the ATC side of the USB interface is not the problem.  It's as if, even when the Sierra fans are spinning like a turbine and the board SHOULD have power that the Sierra just isn't there.  The odd thing is the person from whom I purchased these said they were racked up and actively mining right up to the point where they were shut down, boxed up and sent to me.  Oh yea, I do not see the Sierra boards present themselves in dmesg and I don't see them in sudo lsusb.  What gives?

H@shKraker
sr. member
Activity: 403
Merit: 250
November 22, 2014, 10:26:54 AM
Hey Mr. Teal / et. al.,

I happen to be in possession of a number of Sierra machines of recent purchase.  They arrived, I spun up their PSUs by using the power button on the PSU and nothing happened.  So ... I decided to press one of the buttons on each of the three boards in the case and viola the CoolT fans started spinning adn I felt certain there was 12VDC on the board.  It seemd to me thsi should have caused ther USB device to spin up as well.  I felt good about this part.  So, I followed the "compiling cgminer for HashFast stuff" guide at http://hashfast.com/wp-content/uploads/2014/05/Compiling-CG-Miner-to-show-multiple-die-temps.pdf and thought I was about to have me some mining success.  This was not the case.  It seems cgminer cannot see the sierra boards (each is USB connected and not chained).  I have tried to connect the Sierra to a USB3 hub and a USB2 hub and I get the same outcome ... nothing shows up in dmesg (oh yea ... I'm running Ubuntu 14.04LTS on said MiniITX / shuttle machine) *AND* thus nothing gets seen by cgminer.  I am using a MininItx form-factor PC.  The wierd part of all of this is tha my BFL SC devices adn Jalos get seen and my BE USB devices get seen.  Could it be that I'm getting borked by said MiniItx machine?  If I followed the HashFast guide set-by-step shouldn't I see  predictable outcome / the outcome HashFast described?  Any amount of your wisdom is HUGELY appreciated.

H@shKraker
If you type lsusb, do you see the devices?
Also, you might want to try with a vanilla version of cgminer first just to ensure it's not an issue with the build of cgminer.

Mr Teal,

Many thanks for your reply.  I surely do wish these HashFast units were as bullet proof as you PepperMining units.  Back on subject now.  When I type sudo lsusb I do not see any HashFast devices even though I can hear the Sierra fans spinning like turbine engines.  Also, I've used both the HashFast cgminer git pull *AND* the latest standard cgminer git pull.  I have a standard ATX Intel-I7 machine running some Habaneros right now sooooooo .... I brought one of the Sierras to that same mining location and am gonna connect it up with the Habaneros and see if  I can eliminate the mini-itx form factor / USB hardware as the culprit.

H@shKraker
newbie
Activity: 23
Merit: 0
November 22, 2014, 07:27:24 AM
Hello,
I have problem after playing around with one of cards:

[2014-11-20 01:36:59] Started cgminer 4.7.0
[2014-11-20 01:37:00] HFA: Found device with name 4f577ab79ee6z
[2014-11-20 01:37:00] Probing for an alive pool
[2014-11-20 01:37:00] Pool 0 difficulty changed to 32
[2014-11-20 01:37:01] Network diff set to 40.3G
[2014-11-20 01:37:03] ERR: Asked to memcpy 0 bytes from usbutils.c _usb_read():3197
[2014-11-20 01:37:03] HFA : OP_USB_INIT failed! Operation status 17 (ASIC core power fault)
[2014-11-20 01:37:03] HFA : OP_USB_INIT: Quadrant 3 (of 4) regulator failure
[2014-11-20 01:37:06] API running in UNRESTRICTED read access mode on port 4028 (9)
After that it goes in loop and trying to start.

What is Quadrant and regulator?
Each of the four dies is a quadrant. The regulator is what feeds low voltage power to the chip. Quadrant 3 would be the one diagonally across from the side with the USB header.
Sounds like your board has a fault on it.

http://www.swissapp.mobi/wp-content/uploads/2014/05/mqdefault5.jpg

In picture quadrant 3, it is high diagonally on left, closest to USB header? Will check regulators, it can be faulty capacitor for that quadrant maybe too?
It is possible to get diagram for Hashfast boards?  Wink
I had Antminer s1 with broken capacitor and regulator, i fixed regulator and worked ok without capacitor... Smiley

If you figure it out or find a way to bypass that die or something, please share. I have the same issue with several cards.

Set the voltage for that die to 0 and it will bypass it.

Can you explain little more how to do that in your opinion?
legendary
Activity: 1274
Merit: 1004
November 21, 2014, 11:26:20 PM
Hello,
I have problem after playing around with one of cards:

[2014-11-20 01:36:59] Started cgminer 4.7.0
[2014-11-20 01:37:00] HFA: Found device with name 4f577ab79ee6z
[2014-11-20 01:37:00] Probing for an alive pool
[2014-11-20 01:37:00] Pool 0 difficulty changed to 32
[2014-11-20 01:37:01] Network diff set to 40.3G
[2014-11-20 01:37:03] ERR: Asked to memcpy 0 bytes from usbutils.c _usb_read():3197
[2014-11-20 01:37:03] HFA : OP_USB_INIT failed! Operation status 17 (ASIC core power fault)
[2014-11-20 01:37:03] HFA : OP_USB_INIT: Quadrant 3 (of 4) regulator failure
[2014-11-20 01:37:06] API running in UNRESTRICTED read access mode on port 4028 (9)
After that it goes in loop and trying to start.

What is Quadrant and regulator?
Each of the four dies is a quadrant. The regulator is what feeds low voltage power to the chip. Quadrant 3 would be the one diagonally across from the side with the USB header.
Sounds like your board has a fault on it.



In picture quadrant 3, it is high diagonally on left, closest to USB header? Will check regulators, it can be faulty capacitor for that quadrant maybe too?
It is possible to get diagram for Hashfast boards?  Wink
I had Antminer s1 with broken capacitor and regulator, i fixed regulator and worked ok without capacitor... Smiley

If you figure it out or find a way to bypass that die or something, please share. I have the same issue with several cards.

Set the voltage for that die to 0 and it will bypass it.
sr. member
Activity: 446
Merit: 250
November 21, 2014, 11:24:02 PM
Hello,
I have problem after playing around with one of cards:

[2014-11-20 01:36:59] Started cgminer 4.7.0
[2014-11-20 01:37:00] HFA: Found device with name 4f577ab79ee6z
[2014-11-20 01:37:00] Probing for an alive pool
[2014-11-20 01:37:00] Pool 0 difficulty changed to 32
[2014-11-20 01:37:01] Network diff set to 40.3G
[2014-11-20 01:37:03] ERR: Asked to memcpy 0 bytes from usbutils.c _usb_read():3197
[2014-11-20 01:37:03] HFA : OP_USB_INIT failed! Operation status 17 (ASIC core power fault)
[2014-11-20 01:37:03] HFA : OP_USB_INIT: Quadrant 3 (of 4) regulator failure
[2014-11-20 01:37:06] API running in UNRESTRICTED read access mode on port 4028 (9)
After that it goes in loop and trying to start.

What is Quadrant and regulator?
Each of the four dies is a quadrant. The regulator is what feeds low voltage power to the chip. Quadrant 3 would be the one diagonally across from the side with the USB header.
Sounds like your board has a fault on it.



In picture quadrant 3, it is high diagonally on left, closest to USB header? Will check regulators, it can be faulty capacitor for that quadrant maybe too?
It is possible to get diagram for Hashfast boards?  Wink
I had Antminer s1 with broken capacitor and regulator, i fixed regulator and worked ok without capacitor... Smiley

If you figure it out or find a way to bypass that die or something, please share. I have the same issue with several cards.

legendary
Activity: 1274
Merit: 1004
November 21, 2014, 10:21:31 PM
Hey Mr. Teal / et. al.,

I happen to be in possession of a number of Sierra machines of recent purchase.  They arrived, I spun up their PSUs by using the power button on the PSU and nothing happened.  So ... I decided to press one of the buttons on each of the three boards in the case and viola the CoolT fans started spinning adn I felt certain there was 12VDC on the board.  It seemd to me thsi should have caused ther USB device to spin up as well.  I felt good about this part.  So, I followed the "compiling cgminer for HashFast stuff" guide at http://hashfast.com/wp-content/uploads/2014/05/Compiling-CG-Miner-to-show-multiple-die-temps.pdf and thought I was about to have me some mining success.  This was not the case.  It seems cgminer cannot see the sierra boards (each is USB connected and not chained).  I have tried to connect the Sierra to a USB3 hub and a USB2 hub and I get the same outcome ... nothing shows up in dmesg (oh yea ... I'm running Ubuntu 14.04LTS on said MiniITX / shuttle machine) *AND* thus nothing gets seen by cgminer.  I am using a MininItx form-factor PC.  The wierd part of all of this is tha my BFL SC devices adn Jalos get seen and my BE USB devices get seen.  Could it be that I'm getting borked by said MiniItx machine?  If I followed the HashFast guide set-by-step shouldn't I see  predictable outcome / the outcome HashFast described?  Any amount of your wisdom is HUGELY appreciated.

H@shKraker
If you type lsusb, do you see the devices?
Also, you might want to try with a vanilla version of cgminer first just to ensure it's not an issue with the build of cgminer.
legendary
Activity: 1274
Merit: 1004
November 21, 2014, 10:18:19 PM
In picture quadrant 3, it is high diagonally on left, closest to USB header? Will check regulators, it can be faulty capacitor for that quadrant maybe too?
It is possible to get diagram for Hashfast boards?  Wink
I had Antminer s1 with broken capacitor and regulator, i fixed regulator and worked ok without capacitor... Smiley
Bottom right. As to what the problem could be, hard to say. I don't think there's public schematics but I could be wrong.
sr. member
Activity: 403
Merit: 250
November 21, 2014, 09:26:51 PM
Hey Mr. Teal / et. al.,

I happen to be in possession of a number of Sierra machines of recent purchase.  They arrived, I spun up their PSUs by using the power button on the PSU and nothing happened.  So ... I decided to press one of the buttons on each of the three boards in the case and viola the CoolT fans started spinning adn I felt certain there was 12VDC on the board.  It seemd to me thsi should have caused ther USB device to spin up as well.  I felt good about this part.  So, I followed the "compiling cgminer for HashFast stuff" guide at http://hashfast.com/wp-content/uploads/2014/05/Compiling-CG-Miner-to-show-multiple-die-temps.pdf and thought I was about to have me some mining success.  This was not the case.  It seems cgminer cannot see the sierra boards (each is USB connected and not chained).  I have tried to connect the Sierra to a USB3 hub and a USB2 hub and I get the same outcome ... nothing shows up in dmesg (oh yea ... I'm running Ubuntu 14.04LTS on said MiniITX / shuttle machine) *AND* thus nothing gets seen by cgminer.  I am using a MininItx form-factor PC.  The wierd part of all of this is tha my BFL SC devices adn Jalos get seen and my BE USB devices get seen.  Could it be that I'm getting borked by said MiniItx machine?  If I followed the HashFast guide set-by-step shouldn't I see  predictable outcome / the outcome HashFast described?  Any amount of your wisdom is HUGELY appreciated.

H@shKraker
newbie
Activity: 23
Merit: 0
November 21, 2014, 05:28:39 PM
Hello,
I have problem after playing around with one of cards:

[2014-11-20 01:36:59] Started cgminer 4.7.0
[2014-11-20 01:37:00] HFA: Found device with name 4f577ab79ee6z
[2014-11-20 01:37:00] Probing for an alive pool
[2014-11-20 01:37:00] Pool 0 difficulty changed to 32
[2014-11-20 01:37:01] Network diff set to 40.3G
[2014-11-20 01:37:03] ERR: Asked to memcpy 0 bytes from usbutils.c _usb_read():3197
[2014-11-20 01:37:03] HFA : OP_USB_INIT failed! Operation status 17 (ASIC core power fault)
[2014-11-20 01:37:03] HFA : OP_USB_INIT: Quadrant 3 (of 4) regulator failure
[2014-11-20 01:37:06] API running in UNRESTRICTED read access mode on port 4028 (9)
After that it goes in loop and trying to start.

What is Quadrant and regulator?
Each of the four dies is a quadrant. The regulator is what feeds low voltage power to the chip. Quadrant 3 would be the one diagonally across from the side with the USB header.
Sounds like your board has a fault on it.

http://www.swissapp.mobi/wp-content/uploads/2014/05/mqdefault5.jpg

In picture quadrant 3, it is high diagonally on left, closest to USB header? Will check regulators, it can be faulty capacitor for that quadrant maybe too?
It is possible to get diagram for Hashfast boards?  Wink
I had Antminer s1 with broken capacitor and regulator, i fixed regulator and worked ok without capacitor... Smiley
Pages:
Jump to: