[DragonMint T1] All 3 boards died after power failure

[tim-bc]

Hi everyone,

I have one DragonMint T1 in a facility that lost power a couple days ago due to a passing storm. Now, when the T1 boots, it will not mine.

Code:

Oct 13 12:49:22 dragonMint cgminer[1480]: Started cgminer 4.10.0
Oct 13 12:49:22 dragonMint cgminer[1480]: Loaded configuration file /etc/cgminer.conf
Oct 13 12:49:22 dragonMint cgminer[1480]: Error in configuration file, partially loaded.
Oct 13 12:49:22 dragonMint cgminer[1480]: Probing for an alive pool
Oct 13 12:49:22 dragonMint cgminer[1480]: Pool 0 difficulty changed to 8192
Oct 13 12:49:22 dragonMint dm-monitor[1274]: [INFO:2018-10-13 12:49:22.738] b9 pin:957,a10 pin:943
Oct 13 12:49:22 dragonMint dm-monitor[1274]: [INFO:2018-10-13 12:49:22.738] b9 pin:957,a10 pin:943
Oct 13 12:49:22 dragonMint dm-monitor[1274]: [INFO:2018-10-13 12:49:22.738] b9:1,a10:1,val:0x00000003
Oct 13 12:49:22 dragonMint dm-monitor[1274]: [INFO:2018-10-13 12:49:22.738] cow vid vid
Oct 13 12:49:22 dragonMint dm-monitor[1274]: [INFO:2018-10-13 12:49:22.738] b9:1,a10:1,val:0x00000003
Oct 13 12:49:22 dragonMint dm-monitor[1274]: [INFO:2018-10-13 12:49:22.738] cow vid vid
Oct 13 12:49:23 dragonMint cgminer[1480]: Pool 1 difficulty changed to 8192
Oct 13 12:49:23 dragonMint cgminer[1480]: Failed to fread size 8 nmemb 1 from dragonmint_t1.c dragonmint_get_hwver line 1198
Oct 13 12:49:23 dragonMint cgminer[1480]: sys : platform[4] miner_type[1] chain_num[3] chip_num[68]
Oct 13 12:49:24 dragonMint cgminer[1480]: vid type detected: 0
Oct 13 12:49:24 dragonMint cgminer[1480]: fan speed set to 100
Oct 13 12:49:24 dragonMint cgminer[1480]: T1: checking T1 chain
Oct 13 12:49:24 dragonMint cgminer[1480]: chain0: spi speed set to 390K
Oct 13 12:49:24 dragonMint cgminer[1480]: ERROR - failed to recv spi data
Oct 13 12:49:24 dragonMint cgminer[1480]: failed to reset chain 0!
Oct 13 12:49:24 dragonMint cgminer[1480]: chain1: spi speed set to 390K
Oct 13 12:49:24 dragonMint cgminer[1480]: ERROR - failed to recv spi data
Oct 13 12:49:24 dragonMint cgminer[1480]: failed to reset chain 1!
Oct 13 12:49:24 dragonMint cgminer[1480]: chain2: spi speed set to 390K
Oct 13 12:49:25 dragonMint cgminer[1480]: ERROR - failed to recv spi data
Oct 13 12:49:25 dragonMint cgminer[1480]: failed to reset chain 2!

The "ERROR - failed to recv spi data" is present for all 3 boards. I tried both soft reboot and hard reboot, no change. Any ideas?

Update: Got 1 board back after another reboot, but then it dropped back out due to too many resets. So we are back to square one. Here is a copy of the most recent log - https://pastebin.com/M7j8ywNT

[1714832486]

1714832486

[1714832486]

1714832486

[1714832486]

1714832486

[1714832486]

1714832486

[1714832486]

1714832486

[HagssFIN]

Have you tried different firmware versions?

[tim-bc]

It has the latest, the May 15th, which seems to be the only one to show logs. When I tried flashing back to May 5th it froze around 30%. Log - https://pastebin.com/wbq4qsWC. I had this issue before when upgrading, I'll move it to a different network and try flashing some different firmwares.

[sarcheer]

There is a firmware from Innosilicon for the T1 you could try, can still flash back to the Halong firmwares afterwards.

[tim-bc]

There is a firmware from Innosilicon for the T1 you could try, can still flash back to the Halong firmwares afterwards.

Sounds good. I'll try that when I can (in a couple days) and let you know what happens. I guess there is always the Braiins OS as well.

[sarcheer]

Sounds good. I'll try that when I can (in a couple days) and let you know what happens. I guess there is always the Braiins OS as well.

I’d hold off on the Braiins OS for now, its still in alpha and if the install goes wrong (no one has been able to get it working on the T1 yet) there’s no way to undo it since there’s no SD card slot. They actually recommended no one attempt it with the T1 after it was brought up that there is no SD slot.

[philipma1957]

do  you have the older firmwares?

[tim-bc]

do  you have the older firmwares?

Yes, I have five Halong T1 firmwares in total. It is running the latest one right now, so tomorrow I think I'll try downgrading to the second latest and see what happens.

So, I had unplugged it three days ago. When I came back and plugged it in today, all of a sudden all three boards work again.

I don't know what to make of it. It's weird how the T1 hashboards seem to randomly decide whether to work or not.

I had originally left it unplugged for a few hours and nothing happened when I plugged it back in. I guess three days is the optimal time period for a resurrection?