[BUG]The job task grows indefinitely until the cache hard disk runs out of space

[peterwillcn]

The job task grows indefinitely until the cache ubuntu hard disk runs out of tmp space

[altendky]

I'm sorry, I'm not clear what is being reported here. What files are taking up space? Which drive are you referring to? The usual terminology here is something like tmp, tmp2, dst, and then just wherever everything else like chia and plotman and the OS etc go.

[BasilHorowt]

@peterwillcn any chance you are on OSX? There was a fixed issue related to that platform but I don't believe it's been merged into main yet #142

[zrjaa1]

+1. Running plotman in WSL2 with 1 tmp drive and 1 dst drive without archiving:

directories:
        log: /home/jay/chia-blockchain/plotman-logs
        tmp:
                - /mnt/p
        tmp_overrides:
                "/mnt/p":
                        tmpdir_max_jobs: 8

        # tmp2: /mnt/tmp/a

        dst:
                - /mnt/d
                - 
        # archive:
                # rsyncd_module: plots
                # rsyncd_path: /plots
                # rsyncd_bwlimit: 80000  # Bandwidth limit in KB/s
                # rsyncd_host: myfarmer
                # rsyncd_user: chia

Here is my configuration for scheduling:

        tmpdir_stagger_phase_major: 2
        tmpdir_stagger_phase_minor: 1
        # Optional: default is 1
        tmpdir_stagger_phase_limit: 2

        # Don't run more than this many jobs at a time on a single temp dir.
        tmpdir_max_jobs: 8

        # Don't run more than this many jobs at a time in total.
        global_max_jobs: 10

        # Don't run any jobs (across all temp dirs) more often than this.
        global_stagger_m: 10

        # How often the daemon wakes to consider starting a new plot job
        polling_time_s: 20

My understanding is that maximum of 8 jobs will be created with this configuration, and plotman status should be able to display the status of all the 8 jobs. However, there are 10 plotting logs created and the plotman status only shows 2 jobs：

(venv) root@JAY-DESKTOP:/home/jay/chia-blockchain/plotman-logs# ls
2021-04-23-16:44:25.log  2021-04-23-17:23:34.log  2021-04-23-18:22:51.log  2021-04-23-20:01:25.log
2021-04-23-16:54:26.log  2021-04-23-17:38:59.log  2021-04-23-18:54:41.log
2021-04-23-17:09:51.log  2021-04-23-17:53:23.log  2021-04-23-19:28:13.log
(venv) root@JAY-DESKTOP:/home/jay/chia-blockchain/plotman-logs# plotman status
 plot id    k      tmp      dst   wall   phase    tmp    pid   stat    mem   user    sys   io
3bef6f46   32   /mnt/p   /mnt/d   0:20     1:2    68G   8034    SLP   5.4G   0:19   0:03   0s
b8a0387a   32   /mnt/p   /mnt/d   0:54     1:3   111G   7943    SLP   5.4G   0:47   0:04   0s

Did I miss anything or mis-understand the configuration? Please help, thanks!

[asukahan]

I'm sorry, I'm not clear what is being reported here. What files are taking up space? Which drive are you referring to? The usual terminology here is something like tmp, tmp2, dst, and then just wherever everything else like chia and plotman and the OS etc go.

I think he means "jobs kept adding until temp disk was full".

I've got the same issue in macOS, logs keeps growing every 20 secs and chia process added simultaneously with logs, but no process is in list above with interactive mode. same situation with plotman plot .

[altendky]

@asukahan that does seem likely but I figured it would be good to know exactly.

what version of plotman are you using? Could you share the output of ps aux | grep 'plots create'? Also, how did you install chia? If you are using the dmg then this may be fixed on the development branch by #140.

[asukahan]

@asukahan that does seem likely but I figured it would be good to know exactly.

what version of plotman are you using? Could you share the output of ps aux | grep 'plots create'? Also, how did you install chia? If you are using the dmg then this may be fixed on the development branch by #140.

thanks a lot m8.
the 'development' branch works like charm!

[altendky]

@peterwillcn given the various discussion and the fix in #140, is this issue still open?

[BasilHorowt]

@peterwillcn ping