Ubuntu 20.04 crash without error

[diegotanel]

Hello.

I have a problem with the Mad Max plotter. I've been testing it for 6 days and it always fails. Crashes without registration error and in any phase. Also I tested with many combinations of parameters. I have the lastest commit from github.

I have 3 PCs for plots. Specs: Ubuntu 20.04 32 Gb mem Intel I5 10400 (6c/12Th) Tmp drive: 2x1 TB NVMe WD 750 black (RAID 0) Dst drive: HDD 10 TB Seagate

Logs

PC 1: [P4] Finished writing C2 table Phase 4 took 63.1196 sec, final plot size is 108820004972 bytes Total plot creation time was 2761.72 sec (46.0287 min) Started copy to /media/dtanel/hdd1plot/plot-k32-2021-06-15-17-42-ce5581aa1b3d7cefbff21e297d515f6f3f66529abd288df0abf2d4>Crafting plot 5 out of 6 Process ID: 27037 Number of Threads: 10 Number of Buckets: 2^7 (128) Pool Public Key: 928a308d319b9144c66258e504afc5513e0af5da995b4a805fe1a2dca092e37d162d4402e95ac9cf06701d1dfc2654a6 Farmer Public Key: 8f81d7a136514c4e2445573387623c3bf13c7610d963e39815ecb0bda98c5ffc95e627ab3e642f155d0f85c4ca12e1f8 Working Directory: /mnt/ssdtemp/ Working Directory 2: /mnt/ssdtemp/ Plot Name: plot-k32-2021-06-15-18-28-9cfc1805abf1ef8eb3377fdc389c03912b0647f4e40e56645bd79f1ed2a7d552 [P1] Table 1 took 23.6587 sec [P1] Table 2 took 198.889 sec, found 4295039052 matches

PC 2: Number of Plots: 2 Crafting plot 1 out of 2 Process ID: 3252846 Number of Threads: 12 Number of Buckets: 2^7 (128) Pool Public Key: 928a308d319b9144c66258e504afc5513e0af5da995b4a805fe1a2dca092e37d162d4402e95ac9cf06701d1dfc2654a6 Farmer Public Key: 8f81d7a136514c4e2445573387623c3bf13c7610d963e39815ecb0bda98c5ffc95e627ab3e642f155d0f85c4ca12e1f8 Working Directory: /mnt/ssdtemp/ Working Directory 2: /mnt/ssdtemp/ Plot Name: plot-k32-2021-06-15-18-00-7544723d5d6dbebafff23b862445a2768805cb1d6c8b33cc7f9f8f4db3497f07 [P1] Table 1 took 20.3757 sec [P1] Table 2 took 187.388 sec, found 4295043619 matches [P1] Table 3 took 216.747 sec, found 4295123903 matches [P1] Table 4 took 265.005 sec, found 4295096255 matches [P1] Table 5 took 260.14 sec, found 4295125307 matches [P1] Table 6 took 249.603 sec, found 4295090098 matches [P1] Table 7 took 195.129 sec, found 4294939112 matches Phase 1 took 1394.51 sec [P2] max_table_size = 4295125307 [P2] Table 7 scan took 15.5494 sec [P2] Table 7 rewrite took 32.9628 sec, dropped 0 entries (0 %) [P2] Table 6 scan took 39.874 sec [P2] Table 6 rewrite took 67.0465 sec, dropped 581374701 entries (13.5358 %) [P2] Table 5 scan took 38.4818 sec [P2] Table 5 rewrite took 63.9578 sec, dropped 762078716 entries (17.7429 %) [P2] Table 4 scan took 38.0322 sec [P2] Table 4 rewrite took 63.7838 sec, dropped 828922210 entries (19.2993 %) [P2] Table 3 scan took 37.7558 sec [P2] Table 3 rewrite took 62.6074 sec, dropped 855189544 entries (19.9107 %) [P2] Table 2 scan took 37.6382 sec [P2] Table 2 rewrite took 63.1996 sec, dropped 865585616 entries (20.1531 %) Phase 2 took 568.038 sec Wrote plot header with 268 bytes [P3-1] Table 2 took 51.7436 sec, wrote 3429458003 right entries [P3-2] Table 2 took 44.3415 sec, wrote 3429458003 left entries, 3429458003 final [P3-1] Table 3 took 67.7006 sec, wrote 3439934359 right entries [P3-2] Table 3 took 46.3265 sec, wrote 3439934359 left entries, 3439934359 final

PC 3: Crafting plot 5 out of 6 Process ID: 61559 Number of Threads: 10 Number of Buckets: 2^8 (256) Pool Public Key: 928a308d319b9144c66258e504afc5513e0af5da995b4a805fe1a2dca092e37d162d4402e95ac9cf06701d1dfc2654a6 Farmer Public Key: 8f81d7a136514c4e2445573387623c3bf13c7610d963e39815ecb0bda98c5ffc95e627ab3e642f155d0f85c4ca12e1f8 Working Directory: /mnt/ssdtemp/ Working Directory 2: /mnt/ssdtemp/ Plot Name: plot-k32-2021-06-15-18-31-3c67e66c7b417e01e54c1cdbacc9e1ff4805e5171c68839ad2fb73949841f204 [P1] Table 1 took 24.6473 sec

Can you help me? Thank you

[69chargr]

I've been having the same problem in Ubuntu. I had to back off my CPU overclock and I also switched my nvme RAID 0 to ext4, I don't think it liked XFS. 13 plots into a 90 plot run and so far no freezes.

[diegotanel]

@69chargr , thank you for your response. My RAID 0 is ext4. Maybe the crash is related to the CPU configuration? I have all the stock values. I don't OC my CPU. What test can I take? Or how can the error be logged? Thank you

[69chargr]

That's beyond my ability. For all I know this could crash after my next reboot but for now it's working. I know my setup despises 256 bucket size. Crashes 100% of the time. 128 is much more stable. I'm doing P1 on the raid 0 array and ramdisk for the rest. Never crashes in P2 thru P4 so I know it's something with the NVMe array. Both drives pass all standard tests individually.

[rsulym]

On ubuntu 20.04 madmax stops working between P1 Table 3 and P1 Table 4. It "kills" the process. Any solution for the problem?

same here

[diegotanel]

Does anyone have any solutions?

[madMAx43v3r]

if it gets killed, you ran out of RAM

[diegotanel]

I tried with 2 TH and 128 buckets Crafting plot 2 out of 6 Process ID: 34337 Number of Threads: 2 Number of Buckets: 2^7 (128) [P2] Table 4 scan took 71.1961 sec [P2] Table 4 rewrite took 167.015 sec, dropped 828948930 entries (19.3 %) [P2] Table 3 scan took 71.5442 sec [P2] Table 3 rewrite took 166.475 sec, dropped 855128835 entries (19.9096 %) [P2] Table 2 scan took 71.7522 sec dtanel@chia-rig1:~/chia-plotter$

[vmallet]

Same here, on an older machine with 2x Xeon 5410 (old!), Ubuntu 20.04.

However it gets killed because of an invalid opcode (as found from dmesg): "[39678.126371] traps: chia_plot[3512] trap invalid opcode ip:55e0702a51a8 sp:7ffc3a423380 error:0 in chia_plot[55e070267000+17a000]"

(edit: 16GB ram on the machine; only running one chia_plot w/ -r 4 -u 256)

[diegotanel]

if it gets killed, you ran out of RAM

Hello @madMAx43v3r . First, thank you for your plotter development!

I´m monitoring the PC and have a lot of free memory when plotting.

[diegotanel]

Hello @madMAx43v3r

I found the problem and have no relation to the memories. When I ran chia_plotter mad max, I added the ampersand sign to do the background process. Like this: ./build/chiaplot -p $$$ -f $$$ -n -1 -r 12 -t /mnt/ssdtemp/ -d /media/dtanel/hdd1plot/ >> $(date '+%Y-%m-%d%H%M%S').log & And with that, the process dies at random. I don't know why, but that's the problem. I appreciate for your help. Thank you.

[rsulym]

i have same issue here, but i'm running via screen command, e.g. screen -d -m -S chiaX bash -c 'plot here| tee ' some processes went fine, some got sporadically killed after Phase 2 @diegotanel did you managed your issue?

[diegotanel]

Hello @rsulym. I´m using Screen o tmux, and I had no problems again.

[rsulym]

Hi @diegotanel could you pls give your full string? i have to try, maybe some argument did the trick.

[diegotanel]

You must have the uuid package installed. This is my script:

cat chia_plotter_startup.sh

!/bin/bash

export POOLKEY="$$$"; \ export FARMERKEY="$$$"; \ ~/chia-plotter/build/chiaplot \ --poolkey=$POOLKEY \ --farmerkey=$FARMERKEY \ --tmpdir=/mnt/ssdtemp/ \ --finaldir=/media/dtanel/hdd3plot/ \ --count=60 \ --threads=10 \ --buckets=7 \ 2>&1 | tee ~/chia-plotter/log/$(date '+%Y-%m-%d%H%M%S')-$(uuid).log;

[rsulym]

@diegotanel thanks, but how exactly do you run your script in background?

[diegotanel]

1. tmux
2. ./chia_plotter_startup.sh