search

madMAx43v3r / chia-gigahorse

221 stars 32 forks source link

CPU plotter crashes on phase 3 with no error #129

Open MrMrRubic opened 1 year ago

MrMrRubic commented 1 year ago

What happened?

When trying to use: -C 2 and up, the plotter crashes without any error or output to the log.

My server has 128gb ram, of which 110g is used for ramdisk. I was watching the memory during the crash, and it never went over 120gb used in total (8gb available)

Version

Gigahorse CPU-plotter v2.0.0-6ddb7b

What platform are you using?

Windows server 2019, with all the latest updates

What ui mode are you using?

CLI

Relevant log output

Multi-threaded pipelined Chia k32 next-gen plotter - 6dd4b7b
Plot Format: mmx-v2.4
Network Port: 8444 [chia]
Final Directory: C:\DiskShelf\Compressed\2TB11\
Stage Directory: P:\GigahorseTemp\
Number of Plots: 1
Crafting plot 1 out of 1 (2023/05/19 08:53:06)
Process ID: 1844
Number of Threads: 16
Number of Buckets P1:    2^9 (512)
Number of Buckets P3+P4: 2^8 (256)
Pool Puzzle Hash:  [Redacted]
Farmer Public Key: [Redacted]
Working Directory:   P:\GigahorseTemp\
Working Directory 2: R:\GigahorseTemp\
Compression Level:   C5 (xbits = 11, final table = 3)
Plot Name: plot-k32-c5-2023-05-19-08-53-59bb2c47418f67723f6d572cd2f65d752dc34a346946c9ab0bd5668c58e26bf2
[P1] Table 1 took 28.5679 sec
[P1] Table 2 took 147.384 sec, found 4294991669 matches
[P1] Table 3 took 190.992 sec, found 4294943539 matches
[P1] Table 4 took 225.463 sec, found 4294826259 matches
[P1] Table 5 took 240.625 sec, found 4294734532 matches
[P1] Table 6 took 247.321 sec, found 4294596326 matches
[P1] Table 7 took 163.757 sec, found 4294124397 matches
Phase 1 took 1244.84 sec
[P2] max_table_size = 4294943539
[P2] Table 7 scan took 27.7048 sec
[P2] Table 7 rewrite took 69.6111 sec, dropped 0 entries (0 %)
[P2] Table 6 scan took 32.2328 sec
[P2] Table 6 rewrite took 57.4147 sec, dropped 581334562 entries (13.5364 %)
[P2] Table 5 scan took 32.0084 sec
[P2] Table 5 rewrite took 55.3441 sec, dropped 761957525 entries (17.7417 %)
[P2] Table 4 scan took 31.3324 sec
[P2] Table 4 rewrite took 54.4059 sec, dropped 828866819 entries (19.2992 %)
Phase 2 took 363.52 sec
[P3-1] Table 3 took 123.389 sec, wrote 3439801636 right entries
madMAx43v3r commented 1 year ago

Hmm yeah, windows doesn't show the error...

If I had to guess, not enough free space on P ?

madMAx43v3r commented 1 year ago

Make sure to delete left-over files in -t and -2 after a crash

MrMrRubic commented 1 year ago

hm, it shouldn't. P: (-t) have 528gb available before a plot and k32 should only need about 220(?), but i'll run it again to double check.

MrMrRubic commented 1 year ago

P: (-t) have 405gb available after the crash, R: (-2) have 69.9gb available. I can't see anything in the event viewer.

madMAx43v3r commented 1 year ago

Try latest version, it should display an error message.

MrMrRubic commented 1 year ago

No error message in the output log.

madMAx43v3r commented 1 year ago

Hmm dang, then it seems like it's being killed by the OS... which probably means not enough RAM.

MrMrRubic commented 1 year ago

I'd assume RAM as well, but watching it live it have about 8gb available, so i can't see how it could run out.

I had a friend test with my keys and it works on his system, so there's got to be someting wrong with my setup. I'm gonna reinstall to Windows 2022 this weekend (would change to some kind of linux, but i have a 12tb storage space i can't easily migrate), so i'll update on whether there's still an issue then.

Block-Captain commented 1 year ago

Better try Linux first