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


Recommended Posts

So do we need to follow a procedure to move to the v2 db?    (never mind i found it)

 

Anyone wondering, click on Machinaris in the docker tab and open Console.....

In the Console window type......

chia db upgrade

 

Then wait a few hours.

Once the window shows as complete (i haven't got that far yet but this is what I'm told)

you can delete the v1 db file from mnt/cache/appdata/machinaris/mainnet/db/blockchain_v1_mainnet.sqlite

 

that'll save you a few Gb's on your cache drive

Edited by mdrodge
  • Thanks 1
Link to comment
3 hours ago, demdream said:

Thx for the upgrade. 

But, how to come back to english langage ? The FR translation is very bad and i'm not enaugh good in english to help you :( . 

 

Hi, the language selection is determined by the "Language-Accept" header sent by your web browser.  This is a configurable setting in each browser, generally you can set the Language priority-order you want.  If you only want English for Machinaris, then use Firefox (for example as a secondary browser) with English as the top selection.

That said, I would really appreciate any help you could offer to improve the translation.  It's really easy, you would only be editing 2 or 3 message files for French.  Each message would have English source on the line above, and the French translation on the line below.    For example:

image.png.342c39a70137ff470cebaf2ffa1649e7.png

 

Sorry again for the French translation.  It's a very weak second language of mine and I had to translate one language when I was adding/testing internationalization support to the Machinaris WebUI.  Please help if you can!

Link to comment
33 minutes ago, guy.davis said:

 

Hi, only Chia v1.3 supports the new v2 blockchain database format so far.  I expect some forks to follow suit in the coming weeks and months however.

ok great, thanks. This just finished for me, v1 db was 74GB, v2 db is 39GB. are we safe to delete v1 db at this point? I can see the sqlite-shm and sqlite-wal are for v2 after a container restart.

Link to comment
11 hours ago, sworcery said:

ok great, thanks. This just finished for me, v1 db was 74GB, v2 db is 39GB. are we safe to delete v1 db at this point? I can see the sqlite-shm and sqlite-wal are for v2 after a container restart.

 

Yes, go ahead and remove the old v1 files.  Regarding other forks, I spoke too soon.  Seems Flax has already released a new version too.  So you can run "flax db upgrade" within the Machinaris-Flax container as well.

  • Like 1
Link to comment
7 hours ago, guy.davis said:

 

Hi, not quite sure what is being measured here.  However, here are the log locations if you want to truncate them.  Most are auto-rotated.  Hope this helps.

 

The image shows the chia and chive fork are both writing to the docker image. The other forks to not.

 

The log location provided are outside of the docker image. Any ideas what could be writing to it?

Edited by rukiftw
Link to comment
7 hours ago, rukiftw said:

The image shows the chia and chive fork are both writing to the docker image. The other forks to not.

 

The log location provided are outside of the docker image. Any ideas what could be writing to it?

 

Sure, you are likely building the Madmax, Bladebit plotters, among other components at container launch time.  You can disable those on the Chia & Chives fullnodes if you are not going to be plotting at all.  Also, you can check the temp build logs, short files, found at /var/log in the container.

 

 

Edited by guy.davis
Link to comment

I am having a bit of wallet trouble. Running chia wallet show returns "No keys loaded. Run 'chia keys generate' or import a key".

I am running 0.7.0. Chia keys show does show keys are loaded (a Master, Farmer and Pool publick keys are displayed along with a First wallet address). I am earning payouts on XCHpool as well according to the xchpool explorer tool. Help?

Link to comment
On 3/27/2022 at 6:54 AM, MortenSchmidt said:

I am having a bit of wallet trouble. Running chia wallet show returns "No keys loaded. Run 'chia keys generate' or import a key".

I am running 0.7.0. Chia keys show does show keys are loaded (a Master, Farmer and Pool publick keys are displayed along with a First wallet address). I am earning payouts on XCHpool as well according to the xchpool explorer tool. Help?


Hello!  Sorry for the delayed response.  I understand that you successfully performed the Machinaris setup to either generate or import a key.  Since Machinaris is simply a WebUI over the Chia CLI binaries, please try the following from the Unraid Console shell for the Machinaris container:

 

chia keys show
chia farm summary
chia wallet show

 

If you are getting errors to any of the commands above, rather than useful info & details, then you may need to perform a manual key add.  This should automatically get run every time the container is started (as logged in Unraid's Docker Log for the Machinaris container).  However, it doesn't hurt to run it manually to test:

 

chia keys add -f /root/.chia/mnemonic.txt


Hope this helps!

Link to comment
49 minutes ago, guy.davis said:


Hello!  Sorry for the delayed response. 

No worries. My problem was a corruption in the wallet files. The problem first occurred when the SSD had run out of space (while doing a db upgrade attempt). Deleting the following and starting up helped:

blockchain_wallet_v2_mainnet_xxxxxxxxxx.sqlite-shm

blockchain_wallet_v2_mainnet_xxxxxxxxxx.sqlite-wal

I left the main wallet file (.sqlite) and after that the wallet quickly synced up and "chia wallet show" returns the expected output.

Now.. to figure out how much space is needed to do the db upgrade, I believe I had around 84GB free before starting the process and yet it just failed on me again.

Link to comment

Hi all, Chia Network devs have released a patch version 1.3.2.  This has been available within the `:develop` image Machinaris just minutes after the Chia release yesterday.  The `:test` image of Machinaris got this new version a few hours later.  So, if you want to immediately run the new Chia version, please use either the `:develop` or `:test` images today.  I expect the next full release of Machinaris v0.7.1, including Chia 1.3.2, will be available shortly.

  • Like 1
Link to comment
On 3/29/2022 at 11:11 PM, MortenSchmidt said:

Now.. to figure out how much space is needed to do the db upgrade, I believe I had around 84GB free before starting the process and yet it just failed on me again.

In case anyone is wondering.. I got a longwinded errormessage concluding with "sqlite3.OperationalError: database or disk is full " and it turned out to be the docker image running out of space. 2GB free was apparently not enough. After increasing the max docker image size to 50GB it was able to complete the db upgrade.

Edit: Oh, and after upgrading you have to stop, delete (or keep as backup) the V1 database and start, and don't put this off as it does not automatically switch over to the v2 database and will need to sync up the v2 database from the time the upgrade process started (mine took more than 24 hours and I was more than 8000 blocks behind when I switched over to the V2 database)

Edited by MortenSchmidt
Link to comment

Good day! Machinaris v0.7.1 is now available. Changes include:

  •  Drive monitoring using Smartctl (WebUI status currently, alerting to come soon)
  •  Dutch translations (nl_NL) provided by @bdeprez. Thanks!
  •  Updates to various fork blockchains and tools including BTCGreen, Flax, HDDCoin, Madmax, MMX and Shibgreen
  •  Chia - v1.3.1 patch release
  •  Chia - v1.3.2 patch release

  drives_grid.thumb.png.e3415d65bff4d05cc37b95b83fbfd31f.png


Notes:
 - Drive monitoring not supported on Windows hosts unfortunately due to Docker device pass-thru.
 - Migration to new Flax database format requires manual step: `flax db upgrade`

  • Like 1
Link to comment
7 hours ago, karldonteljames said:

Have I mis-configured something?

The "mainnet" folder within machinaris appdata folder has grown to a huge 27gb. is this expected behaviour? I wasn't sure if I could move this onto one of the disks containing the plots?

 

Hi, as per the install guide, that's about half of an expected Chia blockchain (v2) database: 

image.thumb.png.b4679ee2384c93776eca3b4eb43d80ec.png

 

You definitely can hold the Machinaris appdata directory on another SSD.  I am using an Unassigned Device SSD currently:

image.thumb.png.c6dd9d9c75999e54c6609d63d77b77f8.png

 

Hope this helps!

Link to comment
On 4/2/2022 at 6:46 PM, guy.davis said:

Good day! Machinaris v0.7.1 is now available. Changes include:

  •  Drive monitoring using Smartctl (WebUI status currently, alerting to come soon)
  •  Dutch translations (nl_NL) provided by @bdeprez. Thanks!
  •  Updates to various fork blockchains and tools including BTCGreen, Flax, HDDCoin, Madmax, MMX and Shibgreen
  •  Chia - v1.3.1 patch release
  •  Chia - v1.3.2 patch release

  drives_grid.thumb.png.e3415d65bff4d05cc37b95b83fbfd31f.png


Notes:
 - Drive monitoring not supported on Windows hosts unfortunately due to Docker device pass-thru.
 - Migration to new Flax database format requires manual step: `flax db upgrade`

Do we need to upgrade all the alt coin DBs as well to v2? I've done the main Chia one, and am currently doing the Flax one (as per instructions above), but what about the rest? They all seem to have v1 dbs in their mainnet/db folders...

Link to comment
5 hours ago, DoeBoye said:

Do we need to upgrade all the alt coin DBs as well to v2? I've done the main Chia one, and am currently doing the Flax one (as per instructions above), but what about the rest? They all seem to have v1 dbs in their mainnet/db folders...

 

Not supported yet as far as I know. Only Chia & Flax so far.  Each fork blockchain will need to update their codebase with the latest changes from Chia.  Not all will or may take a long time.  Good indication of the activity/support for a blockchain fork actually.

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

 

Not supported yet as far as I know. Only Chia & Flax so far.  Each fork blockchain will need to update their codebase with the latest changes from Chia.  Not all will or may take a long time.  Good indication of the activity/support for a blockchain fork actually.

 

The forks that I'm aware have updated to >= 1.3 (chia version numbers)

BTCgreen to 1.3.1

Cactus to 1.3.3

Flax to 1.3.3

SHIBgreen to 1.3.1

  • Upvote 1
Link to comment
  • 2 weeks later...

So I'm having a weird plotting issue running on unRAID. I had, using previous versions, plotted about 35tb or so without issue, but as I upgraded a couple drives in my array I decided to use the old ones to add to my chia farm. I am running the plotter the exact same way (settings below) using Madmax, a 110gb ram disk, an ssd R0, and the same drive pool I was using before (I moved the old plots to the new disks and am repopulating the originals now), and the plotter just randomly stops. Sometimes on the second plot, sometimes on the tenth, it just stops plotting. The docker keeps running and farming, the plotter just sits idle with nothing new in the queue, and abandoning temp files in multiple locations (including a full ram disk that I usually have to empty via cli). Did something change in the latest update? My settings are:

 

unRIAD v6.9.2

Machinaris v0.7.1

Plotter: Madmax

k: 32

threads: 16

buckets: 512

buckets3: 128

rmulti2: 1

 

running on a 3950x (hence the threads) with 128gb ram. The plotter runs at night so the only other active docker or service is pihole.

 

Edit to add: the main screen lists plotting as idle, the plotting screen says it is active but nothing is being plotted, the queue is empty and there is no disc activity. Emptying the ramdisk and pausing then resuming plotting starts it up again.

 

 

 

Edited by greyday
Link to comment
On 4/19/2022 at 3:31 PM, greyday said:

So I'm having a weird plotting issue running on unRAID. I had, using previous versions, plotted about 35tb or so without issue, but as I upgraded a couple drives in my array I decided to use the old ones to add to my chia farm. I am running the plotter the exact same way (settings below) using Madmax, a 110gb ram disk, an ssd R0, and the same drive pool I was using before (I moved the old plots to the new disks and am repopulating the originals now), and the plotter just randomly stops. Sometimes on the second plot, sometimes on the tenth, it just stops plotting. The docker keeps running and farming, the plotter just sits idle with nothing new in the queue, and abandoning temp files in multiple locations (including a full ram disk that I usually have to empty via cli). Did something change in the latest update? My settings are:

 

unRIAD v6.9.2

Machinaris v0.7.1

Plotter: Madmax

k: 32

threads: 16

buckets: 512

buckets3: 128

rmulti2: 1

 

running on a 3950x (hence the threads) with 128gb ram. The plotter runs at night so the only other active docker or service is pihole.

 

Edit to add: the main screen lists plotting as idle, the plotting screen says it is active but nothing is being plotted, the queue is empty and there is no disc activity. Emptying the ramdisk and pausing then resuming plotting starts it up again.

 

 

 

 

Hi, plotters like Madmax have been stable/unchanged for quite a while now.  Please take a close look at the logs in the "/mnt/usr/appdata/machinaris/plotman" folder to see if anything related to the full ramdisk appears.  Perhaps Plotman is attempting to move tmp/final files and can not?

 

 

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

 

Hi, plotters like Madmax have been stable/unchanged for quite a while now.  Please take a close look at the logs in the "/mnt/usr/appdata/machinaris/plotman" folder to see if anything related to the full ramdisk appears.  Perhaps Plotman is attempting to move tmp/final files and can not?

 

 

I'll take a look, but somehow when I restarted the docker for a fifth time (after four consecutive crashes) it just...started working right again. I changed absolutely nothing, but it's been humming along for a few days now, so I'm just going to let it do its thing until the disks are full and then I'll look into it more. Thanks for responding!

  • Like 1
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.