Pages:
Author

Topic: Antminer D3 Blissz firmware (10/12 v1.12 update) - page 29. (Read 125911 times)

newbie
Activity: 56
Merit: 0
Was the beta v2.02 not meant to run for any length of time?  I ran it for ~10 hours and for some reason my Chain 3 is much worse now in terms of HW failures.  I set it on the lowest voltage and before I ran the beta (but ran many different Blissz firmware updates before that) it was fine around 450.  Now it can't get above 419 without multiple HW failures per minute.  Anyone else experience this?

I'd like to add that I've been very happy with what Blissz has been able to do thus far, so thank you!

Yep, that's my problem too.

I have a new highly sensitive chain that WAS my best chain, now is my worst.

I am afraid the aggressive freq/voltage search during autotune screwed my machine/chain up.

No answers? I cannot mine at speeds higher than 19 now, I was at 20.9 before.

It looks like they auto tuned themselves right out of being tuned...

PLEEEAAAH!!!

Pfaaaa!


M,



jr. member
Activity: 136
Merit: 2
Is it possible to set the program Midnight Commander into D3?

Maybe You did not understand how this embedded system of miner works ...
The system file is stored in a non-volatile memory area  ( initramfs.bin.SD ) like ramdisk  that is loaded at each start of the miner ... so it is useless to make changes to the S.O. while it runs, because  at every boot, its load again from the non-volatile memory area ... If you, install any software, on reboot, you re-find it, with the first version. It is necessary to do a backUP like initramfs.bin.SD of memory and then write it in the non-volatile area and then load it at every start-up.
that about cgminer.conf after reboot?
full member
Activity: 240
Merit: 100
Is it possible to set the program Midnight Commander into D3?

Maybe You did not understand how this embedded system of miner works ...
The system file is stored in a non-volatile memory area  ( initramfs.bin.SD ) like ramdisk  that is loaded at each start of the miner ... so it is useless to make changes to the S.O. while it runs, because  at every boot, its load again from the non-volatile memory area ... If you, install any software, on reboot, you re-find it, with the first version. It is necessary to do a backUP like initramfs.bin.SD of memory and then write it in the non-volatile area and then load it at every start-up.
sr. member
Activity: 336
Merit: 258
I took the day today to disassemble one of my D3, check and clean hashing boards, disconnect all cables, and did an hard reset etc.

Hopefully to see if the other 2 hashing boards stopped being hidden, as just one is shown.

All parts were in perfect condition, nothing burned and all looked like new, and still i can just see 1 hashing board.

Cannot understand if this is a internal defect in the cables or some firmware error.

Forgot to say that i also changed frequency and voltages various times, to the lowest and to the highest, to see if the boards appeared, with no luck

I have the machine working now but just at 2/3 of regular speed.

Any clues would be appreciated

If the cables aren't confirmed working yet, disconnect all but the active board, then swap that cable to one of the inactive boards and see if you still have a working board. If not, it's not the cable, it's the hashing boards and you should contact Bitmain for replacements (test both the "dead" ones to be sure). If so, swap the connection on the control board to a different socket. If not working, it's the control board and you should contact Bitmain or search secondary markets for a replacement board. If still working, it's probably the cable, swap the cable with one of the ones you removed and repeat the above steps, then do the same with the last one.

If everything works fine, before reinstalling the other two boards, test your psu vga connectors by swapping them out in a working hash board.

Also do all of this with the Bitmain firmware reinstalled. Testing hardware at stock settings is always recommended.

100% agree on this advice to pinpont the exact problem. Thanks for helping out
member
Activity: 434
Merit: 52
I took the day today to disassemble one of my D3, check and clean hashing boards, disconnect all cables, and did an hard reset etc.

Hopefully to see if the other 2 hashing boards stopped being hidden, as just one is shown.

All parts were in perfect condition, nothing burned and all looked like new, and still i can just see 1 hashing board.

Cannot understand if this is a internal defect in the cables or some firmware error.

Forgot to say that i also changed frequency and voltages various times, to the lowest and to the highest, to see if the boards appeared, with no luck

I have the machine working now but just at 2/3 of regular speed.

Any clues would be appreciated

If the cables aren't confirmed working yet, disconnect all but the active board, then swap that cable to one of the inactive boards and see if you still have a working board. If not, it's not the cable, it's the hashing boards and you should contact Bitmain for replacements (test both the "dead" ones to be sure). If so, swap the connection on the control board to a different socket. If not working, it's the control board and you should contact Bitmain or search secondary markets for a replacement board. If still working, it's probably the cable, swap the cable with one of the ones you removed and repeat the above steps, then do the same with the last one.

If everything works fine, before reinstalling the other two boards, test your psu vga connectors by swapping them out in a working hash board.

Also do all of this with the Bitmain firmware reinstalled. Testing hardware at stock settings is always recommended.
newbie
Activity: 21
Merit: 0
Is it possible to set the program Midnight Commander into D3?
newbie
Activity: 9
Merit: 0
I took the day today to disassemble one of my D3, check and clean hashing boards, disconnect all cables, and did an hard reset etc.

Hopefully to see if the other 2 hashing boards stopped being hidden, as just one is shown.

All parts were in perfect condition, nothing burned and all looked like new, and still i can just see 1 hashing board.

Cannot understand if this is a internal defect in the cables or some firmware error.

Forgot to say that i also changed frequency and voltages various times, to the lowest and to the highest, to see if the boards appeared, with no luck

I have the machine working now but just at 2/3 of regular speed.

Any clues would be appreciated
sr. member
Activity: 336
Merit: 258
Hi admin, i am just curios that, for the 1.5% fees which means that you taking the 1.5% portion of the coined mine? or you use the D3 hashrate to generate 1.5% profit?


It will mine 1,5% of the time (every two hours) on the dev pool.
newbie
Activity: 5
Merit: 0
Hi admin, i am just curios that, for the 1.5% fees which means that you taking the 1.5% portion of the coined mine? or you use the D3 hashrate to generate 1.5% profit?
newbie
Activity: 33
Merit: 0
Thanks !
Waiting for the next releases.


Is there anyway i can read temp from console and put it in a crontab to kill cgminer when it is
over 100 degrees ?
I am scared for fans not to stop for some reason.(software/hardware).
This will save my machine.
Blissz any chance for you to put checkbox on the web so if it is over to kill cgminer. Thats all.
If anyone wants to use it , if none just no checkbox.

Regards,

It will be added in the next version.
member
Activity: 93
Merit: 10
What power cord are you guys using?
A old powercord from ancient coffee robot Saeco Odea Go Smiley It's pretty thick and marked to 2000W.
newbie
Activity: 14
Merit: 0
What power cord are you guys using?
full member
Activity: 308
Merit: 110
There appears to be a bug in v1.12 (non-beta).  Can you please confirm if this is indeed a bug, and if so, is this fixed in an upcoming version?

Notice how the frequency configuration does not match the frequency on the status page.  This is a pretty big deal, imo.  Hopefully I'm missing something.  Thx.

Configuration:
Actual Status:
Kernel Log: https://pastebin.com/embed_js/XviQk8H6

Your boards are connected wrong to the controller board. Use 1 to 3. You're on 2 to 4...

Awesome ! I had the same issue, and now back 1 to 3. Thank you guys. I did wonder for a while but then, forgot about it.  Cheesy
newbie
Activity: 13
Merit: 0
Was the beta v2.02 not meant to run for any length of time?  I ran it for ~10 hours and for some reason my Chain 3 is much worse now in terms of HW failures.  I set it on the lowest voltage and before I ran the beta (but ran many different Blissz firmware updates before that) it was fine around 450.  Now it can't get above 419 without multiple HW failures per minute.  Anyone else experience this?

I'd like to add that I've been very happy with what Blissz has been able to do thus far, so thank you!
sr. member
Activity: 336
Merit: 258
There appears to be a bug in v1.12 (non-beta).  Can you please confirm if this is indeed a bug, and if so, is this fixed in an upcoming version?

Notice how the frequency configuration does not match the frequency on the status page.  This is a pretty big deal, imo.  Hopefully I'm missing something.  Thx.

Configuration:
Actual Status:
Kernel Log: https://pastebin.com/embed_js/XviQk8H6

Your boards are connected wrong to the controller board. Use 1 to 3. You're on 2 to 4...
member
Activity: 166
Merit: 16
hi @blissz,

do you know who is this guy "koffee.1" after 24 hours, with firmware v1.11 the D3 worker changed to that guy ?
my D3 is rented !


I have no clue really...

Hi Blissz,

your firmware is really cool, i love it very much.
But the fact is that , I tried with firmware 1.11 and 1.12, after almost a day running , suddenly my worker is changed to "koffee.1".
It is really strange . My renter at MRR inform it to me, that the D3 does not work for him anymore.

I sadly have to back to stock firmware.

Please investigate the case, find out issue, so that i could come back to your awsome firmware.

Thanks


I'm pretty sure someone is on your network changing it then. Update passwords

Thanks Blissz,

It is my home-network only with kid and grandmother. anyway i changed D3's password.
Is there any possibility that the back-door is in the D3-machine ( not to offend you, but may be from local vendor, from Bitman ) ?




I have to agree with blissz on this, I've had something like it happen before. I had a dmz setup but all the miners were dhcp, power went out and when it came back up one of the miners took the dmz spot (this was years ago with an s1 and I had not changed stock pw) it wasn't long before it was mining for someone else. I'm pretty sure several people out there have bots running just looking to get into miners and then alerts them they can get in. Likely it wasn't anyone driving by getting into wifi, it was from outside your network. That's where I will lay my bets. Maybe do a port scan and see what all is open on your network?


You are correct about the bot, DMZ, and D3 default password. I correct it, and will try the firmware v1.12 again.

Thank you

I just remembered it is all.. hell honestly I think it might have been the same koffee.1 username and it was a bitmain s1 they got into.. so maybe someone has a bot out there loose in the wild just going root root into anything it can get ahold of.
member
Activity: 166
Merit: 16
hi @blissz,

do you know who is this guy "koffee.1" after 24 hours, with firmware v1.11 the D3 worker changed to that guy ?
my D3 is rented !


I have no clue really...

Hi Blissz,

your firmware is really cool, i love it very much.
But the fact is that , I tried with firmware 1.11 and 1.12, after almost a day running , suddenly my worker is changed to "koffee.1".
It is really strange . My renter at MRR inform it to me, that the D3 does not work for him anymore.

I sadly have to back to stock firmware.

Please investigate the case, find out issue, so that i could come back to your awsome firmware.

Thanks


I'm pretty sure someone is on your network changing it then. Update passwords


Scary.. will a Cisco Meraki firewall keep people off network?
Also blissz how we pay you? Any chance you live in Oregon?

Just make sure your miner and maybe router aren't on default passwords.
member
Activity: 166
Merit: 16
v2.02 beta :: 14/12 :: test version

Download:

https://mega.nz/#!DghG2KwC!OJEK4_d47skRB6yrLRE2bfzlX7FWXLXzgoKA_q5dKco


Changes 14/12:
- fixed auto recover from xxxx
- cleaned up kernel logging
- a bit more aggressive tuning
- tunes down faster now when too many HW errors are detected


Changes 13/12:
- Totally rewritten code base, based from latest available bitmain / cgminer sources
- ASIC's now automatically recovers from xxxxx errors without restarting (The auto reboot options are still in there in case it doesn't succeed)
- blazing fast mining status overview in the webinterface Cheesy
- as requested: the dev mining starts later now (after 5 minutes)
- as requested: dev pools status is now visible in the miner status

and last but not least...

- added a first version of the auto tune!  Cool

When enabled this will search for the best possible frequency at the given voltage setting. The first 5 to 10 minutes it will search for a good base to start from. After that it checks for excessive HW errors every 15 minutes and adjusts the frequency accordingly.

Important: This feature is not completely finished and the startup is a bit slow now.. Therefore I need your help. So what do you need to do:

set your advanced settings the same as displayed bellow
let it run for at least 20 minutes
PM your (complete) kernel log
By doing so I can get to a conclusion what the starting frequency of the auto tune algorithm should be.
In case you want to help more: increase the voltage a bit and do the steps above again as it will give some more information

Thank you!



Please test and give feedback / a kernel log how it works for you. I need some feedback on this version guys  Grin



Well, i think it screwed my string 2 up Blissz!

Now no matter what software i run, INCLUDING BM FW Newest, I NOW GET A LOT OF HW on my coolest string.

When I upgraded or downgraded, I left it on keep settings, WAS THAT WRONG? Please inform....

It is still cooler than the other two and not hot, but it gives HW now.

Hmmmmm

GRRRRRRRRRR

Appreciate any help.

M

I went back to 12/10 for now....I Have to run the card on three levels lower frequency than the other two.

It feels like it has an over written temperature warning sensor....

i don't know,

This sucked today.

Lost my 8 bucks!

someone else pointed out that the voltage setting sticks, so before flashing back to stock set everything to default then flash it. That "should" fix that, unless I read that wrong, first sip of coffee in hand..
newbie
Activity: 50
Merit: 0
There appears to be a bug in v1.12 (non-beta).  Can you please confirm if this is indeed a bug, and if so, is this fixed in an upcoming version?

Notice how the frequency configuration does not match the frequency on the status page.  This is a pretty big deal, imo.  Hopefully I'm missing something.  Thx.

Configuration: https://preview.ibb.co/jAQCs6/advanced_configuration.png
Actual Status: https://preview.ibb.co/h6BVem/status.png
Kernel Log: https://pastebin.com/embed_js/XviQk8H6

I saw someone else with a similar issue and it was that the boards were plugged in incorrectly.  Hence the board 2,3,4 numbering... 
Pages:
Jump to: