[Support] Machinaris - Chia cryptocurrency farming + Plotman plotting + Unraid WebUI


Recommended Posts

Good day! The next version of Machinaris is now available: v0.5.2.

Thanks to all those who tested and provided feedback during the release cycle. All contributions are appreciated!

Changes: 
* Machinaris - Docker images now available for Apple M1 and Raspberry Pi OS architectures.
* Chiadog - Update to new v0.7.0 to support for parsing partials and solo blocks.
* Chia - Update to patch release of 1.2.3. See their changelog for details.
* Flax - Update to version 0.1.1. See their changelog for details.
* TrueNAS - Support for Machinaris deployment via helm chart. Issue #78 - Big thanks to @kmoore134 for this!
* Machinaris - Daily Farming Summary now available on Farming page for both Chia and Flax. Add new plot type column.
* Machinaris - Pools - Show each Pool's status including link to your pool provider. List pool point events on Summary page.
* Machinaris - Workers - Use hostname for Worker display name, even when using IP addresses behind the scenes. Also show versions. Automated harvester setup.
* Machinaris - Connections page has link to test your router port forward for farming.
* Machinaris - New public website with launch Wizard for generating first Docker run/compose of Machinaris: http://www.machinaris.app/

  • Thanks 2
Link to comment
13 hours ago, rav007 said:

Getting this error when I select the pools tab with the latest (just updated) test branch

Screenshot_20210814-152639.jpg

 

I encountered the same error when I tried to purge an old worker. I got the following suggestion from Guy in the Discord channel, which resolved it:

1) Stop Machinaris container in Unraid Admin UI

2) Delete only this file: /mnt/user/appdata/machinaris/machinaris/dbs/machinaris.db

3) Start Machinaris container in Unraid Admin UI

4) Wait about ~3-5 minutes for everything to start and get re-populated. Then try clicking around in Machinaris WebUI.

  • Like 1
  • Thanks 1
Link to comment
On 8/13/2021 at 11:46 AM, guy.davis said:

Hi, not sure I understand. Please provide some more details.  Also take a look at setting up a single Machinaris Controller (fullnode) and multiple Machinaris Workers (harvesters,plotters or both):  https://github.com/guydavis/machinaris/wiki/Workers

I only have 1 full node doing all the farming, harvesting, etc.  

 

The issue is a formatting one, it seems.  When I browse the Machinaris URL to administer the application, only 1 of my computers actually sees the correct formatting on the site.  All my other systems I use outputs a poorly formatted website for the same URL. I've tried multiple browsers (phone, desktop, tablet, laptop) and I see the same problem.  The only one that formats correctly is the laptop I used to remotely set up the application initially on Unraid.

 

It's really strange.

 

EDIT: I figured it out.  It was my pihole DNS filter blocking out one of the URLs your app reaches out to:

 

https://cdn.jsdelivr.net

 

 

Edited by Einsteinjr
  • Like 1
Link to comment
10 hours ago, makkish said:

 

I encountered the same error when I tried to purge an old worker. I got the following suggestion from Guy in the Discord channel, which resolved it:

1) Stop Machinaris container in Unraid Admin UI

2) Delete only this file: /mnt/user/appdata/machinaris/machinaris/dbs/machinaris.db

3) Start Machinaris container in Unraid Admin UI

4) Wait about ~3-5 minutes for everything to start and get re-populated. Then try clicking around in Machinaris WebUI.

This did the trick! Thank you very much !

  • Like 1
Link to comment

I've been running the Madmax plotter on a dual E5-2695v2 machine with 256GB RAM.  A plot finishes in about 30 minutes currently with Madmax using the unraid /dev/shm ramdisk (128GB).  I also have an unassigned SSD which gets used as the other plotting drive, but all plots get stored on my slow HDD array when completed, which takes about 15 minutes.

 

I've noticed that currently, my plotting pauses while copying the completed plot to HDD.  From what I can tell, Madmax is supposed to move on immediately, but it looks like Machinaris/Plotman jobs run with '-n 1', which means it does 1 plot per job, and the job isn't "complete" until the copy to HDD is done.  What option am I missing in the configuration to have the plotter start the next plot while the previous one copies?  I can't figure out how to pass a new -n/--count argument from the plotman.yaml configuration.

Edited by trgKai
Link to comment
18 hours ago, trgKai said:

I've been running the Madmax plotter on a dual E5-2695v2 machine with 256GB RAM.  A plot finishes in about 30 minutes currently with Madmax using the unraid /dev/shm ramdisk (128GB).  I also have an unassigned SSD which gets used as the other plotting drive, but all plots get stored on my slow HDD array when completed, which takes about 15 minutes.

 

I've noticed that currently, my plotting pauses while copying the completed plot to HDD.  From what I can tell, Madmax is supposed to move on immediately, but it looks like Machinaris/Plotman jobs run with '-n 1', which means it does 1 plot per job, and the job isn't "complete" until the copy to HDD is done.  What option am I missing in the configuration to have the plotter start the next plot while the previous one copies?  I can't figure out how to pass a new -n/--count argument from the plotman.yaml configuration.

 

Glad to see you got this sorted out on our Discord server.  For others on this thred, the suggestion was to use the following Plotman settings:

 

        tmpdir_stagger_phase_major: 5
        tmpdir_stagger_phase_minor: 0
        # Optional: default is 1
        tmpdir_stagger_phase_limit: 1

        # Don't run more than this many jobs at a time on a single temp dir.
        # Increase for staggered plotting by chia, leave at 1 for madmax sequential plotting
        tmpdir_max_jobs: 2

        # Don't run more than this many jobs at a time in total.
        # Increase for staggered plotting by chia, leave at 1 for madmax sequential plotting
        global_max_jobs: 2

 

Thanks again to user doma for this great improvement to Plotman.

  • Like 1
Link to comment

Hi there - looking for a little guidance on getting my Flax wallet Synced. On the Network tab I can see that both Chia/Flax have Full Node Synced.  On the Wallet tab, my Chia wallet shows as Synced, but my Flax remains Not Synced (both are set to the same mnemonic key).  Am I missing a piece in the config somewhere to get my Flax wallet synced up?

 

Thanks!

Link to comment
2 hours ago, Khalbrogo said:

Hi there - looking for a little guidance on getting my Flax wallet Synced. On the Network tab I can see that both Chia/Flax have Full Node Synced.  On the Wallet tab, my Chia wallet shows as Synced, but my Flax remains Not Synced (both are set to the same mnemonic key).  Am I missing a piece in the config somewhere to get my Flax wallet synced up?

 

Hi!  I recommend retrying the Flax wallet sync by:

  1. Stopping Machinaris container
  2. Deleting this folder from your Unraid filesystem:  /mnt/user/appdata/machinaris/flax/mainnet/wallet/db/
  3. Start up the Machinaris container again

This will get Flax wallet syncing from zero again and hopefully it will work better.  Issue could be due to db file corruption.  Hope this helps!

Link to comment
6 hours ago, guy.davis said:

 

Hi!  I recommend retrying the Flax wallet sync by:

  1. Stopping Machinaris container
  2. Deleting this folder from your Unraid filesystem:  /mnt/user/appdata/machinaris/flax/mainnet/wallet/db/
  3. Start up the Machinaris container again

This will get Flax wallet syncing from zero again and hopefully it will work better.  Issue could be due to db file corruption.  Hope this helps!

That did the trick, thanks for your help!!

  • Like 1
Link to comment

 

On 8/18/2021 at 11:12 AM, guy.davis said:

 

Glad to see you got this sorted out on our Discord server.  For others on this thred, the suggestion was to use the following Plotman settings:

 

        tmpdir_stagger_phase_major: 5
        tmpdir_stagger_phase_minor: 0
        # Optional: default is 1
        tmpdir_stagger_phase_limit: 1

        # Don't run more than this many jobs at a time on a single temp dir.
        # Increase for staggered plotting by chia, leave at 1 for madmax sequential plotting
        tmpdir_max_jobs: 2

        # Don't run more than this many jobs at a time in total.
        # Increase for staggered plotting by chia, leave at 1 for madmax sequential plotting
        global_max_jobs: 2

 

Thanks again to user doma for this great improvement to Plotman.

 

I have been running into this same issue but ultimately I am still bottlenecked by the transfer process. I plot in approx 45 minutes but the transfer process takes a bit longer (around 20MB/s). When I complete a 2nd plot, I now have Unraid transferring two plots on the same 20MB/s which slows everything down again.

 

The ideal solution would be for Unraid to select a 2nd drive for the 2nd plot. Currently it always seems to pick the same drive for both file transfers. Any guidance on how to get Unraid to automatically pick another drive?

Edited by rik3k
Link to comment

Any suggestions on getting the harvester+plotter working? 

Error: Connection Refused

 

certs are are in both the farmer_ca and \mainnet\config\ssl\ca folders

IPs are used instead of hostnames.

 

Here are the docker command:

draft1: generated on the full node, using the new worker button.

docker run -d -t --name machinaris-plotter -h chiaharvest -v /mnt/user/appdata/machinaris/:/root/.chia -v "/mnt/user/chia/:/plots" -v "/mnt/ssdnvme/:/plotting" -e TZ=America/Los_Angeles -e farmer_address=10.0.0.2 -e farmer_port=8447 -e flax_farmer_port=6888 -e farmer_pk=* -e pool_pk=* -e pool_contract_address=* -e mode=harvester,plotter -e controller_host=10.0.0.2 -e controller_api_port=8927 -e plots_dir=/plots -e blockchains=chia,flax -p 8926:8926 -p 8927:8927 ghcr.io/guydavis/machinaris

draft2: (https://github.com/guydavis/machinaris/wiki/Workers)

docker run -d -t --name "machinaris" -h "chiaharvester" -v "/mnt/user/appdata/machinaris/:/root/.chia" -v "/mnt/user/chia/:/plots" -v "/mnt/ssdnvme/:/plotting" -e TZ="America/Los_Angeles" -e farmer_address="10.0.0.2" -e farmer_pk="*" -e pool_pk="*" -e pool_contract_address="*" -e mode="harvester+plotter" -e controller_host="10.0.0.2" -e plots_dir=/plots -e blockchains=chia -p 8926:8926 -p 8927:8927 ghcr.io/guydavis/machinaris

 

there is no firewall present to be blocking communications.

 

image.thumb.png.38ca804ee72c232d6bbd6714837f989e.png

Link to comment
On 8/24/2021 at 7:02 AM, rik3k said:

 

 

I have been running into this same issue but ultimately I am still bottlenecked by the transfer process. I plot in approx 45 minutes but the transfer process takes a bit longer (around 20MB/s). When I complete a 2nd plot, I now have Unraid transferring two plots on the same 20MB/s which slows everything down again.

 

The ideal solution would be for Unraid to select a 2nd drive for the 2nd plot. Currently it always seems to pick the same drive for both file transfers. Any guidance on how to get Unraid to automatically pick another drive?

I use two drives in a raid 1 for my transfer cache. They write/read at 160mb/s. Then mover runs at 11pm. The array with dual parity reads at 150mb/s, but write is only 35mb/s. Reconstruct mode pushes it to 55mb/s. My dockers are on an  r1  satassd cache drives, and the plotter is on an nvme drive with high endurance. using this setup, a plot is created every 36 minutes, start to finish.

Link to comment
5 hours ago, rukiftw said:
docker run -d -t --name machinaris-plotter -h chiaharvest -v /mnt/user/appdata/machinaris/:/root/.chia -v "/mnt/user/chia/:/plots" -v "/mnt/ssdnvme/:/plotting" -e TZ=America/Los_Angeles -e farmer_address=10.0.0.2 -e farmer_port=8447 -e flax_farmer_port=6888 -e farmer_pk=* -e pool_pk=* -e pool_contract_address=* -e mode=harvester,plotter -e controller_host=10.0.0.2 -e controller_api_port=8927 -e plots_dir=/plots -e blockchains=chia,flax -p 8926:8926 -p 8927:8927 ghcr.io/guydavis/machinaris

 

Hi, please add a Worker IP Address to the Wizard that opens from New Worker button.  Right now, the Machinaris fullnode/controller is trying to resolve a hostname on your LAN of "chiaharvest" which may not be resolving to a local IP.  By setting an IP for the worker, then the controller will use that IP without any DNS lookup.  Hope this helps!

Link to comment
On 8/26/2021 at 7:07 PM, guy.davis said:

 

Hi, please add a Worker IP Address to the Wizard that opens from New Worker button.  Right now, the Machinaris fullnode/controller is trying to resolve a hostname on your LAN of "chiaharvest" which may not be resolving to a local IP.  By setting an IP for the worker, then the controller will use that IP without any DNS lookup.  Hope this helps!

 

is sware i already tried that, but its working now. Your help is highly apprecated.

 

for the next person, you add the following to the template from the ca app, under extra parameters:

 

-t -e TZ=America/Los_Angeles -e worker_address=10.0.0.3 -e farmer_address=10.0.0.2 -e farmer_port=8447 -e flax_farmer_port=6888 -e farmer_pk=* -e pool_contract_address=* -e controller_host=10.0.0.2 -e controller_api_port=8927

 

then fill in the rest of the template as normal.

Edited by rukiftw
Link to comment

Good day! The next version of Machinaris is now available: v0.5.3. Thanks to all those who tested and provided feedback during the release cycle. All contributions are appreciated!

  • Worker page for each worker shows Warnings for certain configuration issues.
  • Farming page now shows current disk usage for each drive (free and used)
  • Plotting page now shows recent disk usage for each drive (free and used)
  • Summary page now charts recent blockchain challenges and partial proofs
  • Fixes from feedback received by users on the Machinaris Discord. Thanks all!
  • Chia - Update to version 1.2.4. See their changelog for details.
Link to comment

Hi there, had my machinaris running without issues for 3 weeks now. This morning:super-nas-diagnostics-20210828-1448.zip


 

Internal Server Error

The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.


 

I followed everythin here, unfortunately no luck..


 

https://github.com/guydavis/machinaris/issues/75


 

Does anybody have an idea? I understand this is a NGINX error, but I can't seem to see what's causing this. Diags attached, thanks for the help in advance!

 

Link to comment
16 hours ago, theunraidhomeuser said:

Internal Server Error

The server encountered an internal error and was unable to complete your request. Either the server is overloaded or there is an error in the application.

 

Hi, it's possible you're encountering the Chia SSL issue that has resulted in new v1.2.5 from them.  Please try to switch to the test image of upcoming Machinaris v0.5.4 which now has Chia 1.2.5 and other fixes.  Set your Repository to "ghcr.io/guydavis/machinaris:test" and restart the container.   Any further such errors, please attach:

  • /mnt/user/appdata/machinaris/logs/webui.log and also apisrv.log

Thanks!

Edited by guy.davis
Link to comment

Hello, I'm missing something in the machinaris wiki. At https://github.com/Chia-Network/chia-blockchain/wiki/Farming-on-many-machines it says: Sektioon Harvester + Plotter: "Don't forget to copy your fullnode's ca certificates over as outlined above in the harvester section. "
Unfortunately, there is no description of where I should copy the CA from the full node and then integrate it in the new worker.
Can you please help me here?

Link to comment
7 hours ago, DaPedda said:

Hello, I'm missing something in the machinaris wiki. At https://github.com/Chia-Network/chia-blockchain/wiki/Farming-on-many-machines it says: Sektioon Harvester + Plotter: "Don't forget to copy your fullnode's ca certificates over as outlined above in the harvester section. "
Unfortunately, there is no description of where I should copy the CA from the full node and then integrate it in the new worker.
Can you please help me here?

 

Hi, this manual copy step for regular Chia is unnecessary for Machinaris workers.  To start a Machinaris worker on your local network:

1) Visit the Machinaris WebUI of your main system (controller & fullnode)

2) Click on the Workers page. Then click 'New Worker' on that page.

3) Complete the Wizard form that appears with details about your new worker system.

4) Take either the Docker run or the Docker compose and execute on the new computer.

 

The Worker will automatically connect to the Controller.  Look for a new row to appear in the Workers table.  If the workers is harvesting, check the Farming page and select the Worker by name, then View Chia Logs to ensure that harvesting has started successfully.  Hope this helps!

Link to comment
15 hours ago, AsTuRkInG said:

Hi, I'm seeing that my harvester node doesnt initialize the flax chain.

I can't see in my log where should be the certs for flax (with chia is already done and working like a charm)

 

Hi, please ensure that you click 'New Worker' button on your Workers page before launch the Machinaris Harvester.  Both Chia and Flax certificates will be automatically installed on the Machinaris harvester from your Machinaris fullnode/controller.

Link to comment

Version 0.5.4 now available with fixes for recent Machinaris and Chia issues:

 

- Chia - Update to version 1.2.5.  See their changelog for details.

- Machinaris - Fix for Chia 1.2.4 SSL issue seen on plotter-only systems.

- Machinaris - Displayname for Workers tabs of Plotting and Farming pages.

- Machinaris - Fix for partials chart on Summary page.

- Machinaris - Fix for plotting log view.

  • Thanks 1
Link to comment

I just saw that all settings in plotman.yaml are gone. A backup from this morning is ignored. I am not amused 😞

Next surprise: no matter what I enter under Settings> Plotting: it doesn't work. It is reset to "factory settings" after i revisite die Settings Tabs

P.S.: manuell Copy from plotman.yaml works, plotter do his Job. But when i revisited the Settings Page for Plotting, all Settings are gone. Wrong Path?

Edited by DaPedda
Link to comment
15 hours ago, DaPedda said:

I just saw that all settings in plotman.yaml are gone. A backup from this morning is ignored. I am not amused 😞

Next surprise: no matter what I enter under Settings> Plotting: it doesn't work. It is reset to "factory settings" after i revisite die Settings Tabs

P.S.: manuell Copy from plotman.yaml works, plotter do his Job. But when i revisited the Settings Page for Plotting, all Settings are gone. Wrong Path?

 

Good day.  The /mnt/user/appdata/machinaris/plotman folder holds plotman.yaml.  That same folder holds a dated backup copy, taken every single time you clicked the Save button on the Settings | Plotting page.  Are you saying that all those files are missing or that entire parent folder structures are missing?  

Regarding a blank settings area shown on Settings | Plotting, please share your /mnt/user/appdata/machinaris/logs/webui.log file here so I can see the source of the problem.

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.