guy.davis

Please paste a screenshot (multiple scrolling down) of the Machinaris docker configuration you are using from the Unraid Admin UI. First example attached, but I'll need to see every part of the configuration, so take multiple screenshots, scrolling down. Also click: "Show more settings..." to expose the hidden settings and screenshot those. Thanks!

After 30 minutes, get into the container by choosing the Docker Console from Unraid for Machinaris and then type: bash plotman version chia_plot --version bladebit --version If any of those give an error instead of a version as shown in my screenshot, please consult the build logs at: cat /tmp/plotman_setup.log cat /tmp/madmax_setup.log cat /tmp/bladebit_setup.log Finally, in the Machinaris WebUI that you are now able to access, browse to the "Settings | Plotting" page and provide the Web Log (webui.log) - available from the Workers page (Chia fullnode row, right side as 'Web Log' button). If "Settings | Plotting" page is returning an 'Internal Server Error', it will be logged in webui.log. Please provide that error. Thanks.

Your install does not seem to have a Chia key. Please stop the Machinaris container. Delete your Machinaris entire appdata folder which you changed to "/mnt/disk3/fastssd/" (defaults to /mnt/user/appdata/machinaris). I would recommend adding `machinaris` folder to "/mnt/disk3/fastssd/" and using "/mnt/disk3/fastssd/machinaris" as your appdata folder, just to be clear about naming. Then restart the container to view the Setup page where you can either Import an Existing Key OR Create a New Key. Once you have successfully Setup your Machinaris install, things should work better for you.

Sorry to hear that. What exactly did you fiind when you consulted the logs? Also, you can see the processes using most CPU by opening the Docker Console for the Machinaris container from the Unraid admin UI's Docker tab, then type: top

Greetings! Always allow at least 15-30 minutes after startup to check on CPU/memory load. Initially, the Chia blockchain attempts to sync from back in 2021 and this syncing process consumes both network and CPU resources. Then, please verify your Unraid configuration, including prerequisites, and then troubleshoot the logs. Hope this helps!

Interesting, when I google for that error, here is the only hit I found for Unraid. No, you definitely can plot using a HDD as temporary plotting space, it's just a fair bit slower. Sorry this is still occurring for you.

Hi, glad you find Machinaris useful. I believe the issue is that Docker doesn't like two volumes mounted where one resides within the other. Do you have a spare SSD you can add to your Unraid box to use as plotting temp space, via "Unassigned Devices" plugin? If so, then your mounts would be like: /plots1 -> /mnt/user/chstorage /plotting -> /mnt/disks/YOURSSD I think this would be faster as well, since Madmax and Bladebit (diskplot) both like SSD as temporary space. Hope this helps.

Hi all! The newest version of Machinaris is now available. Thanks to everyone who provided feedback on version 0.8.6. Added: Re-plotting: Optional background deletion of a few old plots to free space for new plotting. See Farming page, Settings icon, top-right. Table of recent plot archiving (local and remote) on new "Plotting | Transfers" page, including status and transfer speed. "Settings | Alerts" page: new 'Send Test Alert' button to validate Chiadog configs and receive sample alert to mail/discord/etc. Coffee blockchain at version 1.0.1, a slow Silicoin fork. GreenBTC blockchain at version 1.6.3, another slow Silicoin fork. Moon blockchain at version 1.6.0, a recent Chia fork. One blockchain at version 1.6.2, a recent Chia fork. Changed: Fixed broken 'Generate/Import Key' actions for Chia 1.6.1 on Setup page. Thanks @SilverFolfy Missing plots on Farming page when a status.json file was corrupted. Thanks @yurly Fix for duplicated Chiadog alerts of wallet additions. Thanks @GravitasProblem and @doma2345 Improved "Settings | Pools" page with fee amount and delete_unconfirmed_transactions action. Updated: Bladebit to v2.0.1 Cactus to v1.7.0 Chia to v1.6.2 Chinilla to v1.4.0 Chiadog to v0.7.5 Flax to v0.1.11 MMX to (v0.9.3) on testnet9. Supporting the latest AMD GPU driver. Tad to v1.7.0 Notes: Support for new Chia forks DOES NOT imply my endorsement for them. Only run those you are comfortable with.

Hi, you can always go the command line inside the 'machinaris-flora' container and type: flora start farmer -r This is exactly what Machinaris is doing when you unpause the Flora wallet. Watch the Flora blockchain log for errors, wait about 15 minutes and then try: flora wallet show If this doesn't show the wallet running, then you may have a problem with the Flora blockchain, possibly data files, requiring further troubleshooting. If you would prefer to have your Wallets running (default), then change the setting by clicking the gear icon on the top-right of the Wallets page. Hope this helps.

Hi, good question. The old Chia version labelled this as 'P2 singleton address', but currently the output of `chia plotnft show` (shown on the Settings | Pools page) correctly labels this as: Pool contract address (use ONLY for plotting - do not send money to this address): xch12345678ABCD... Initially the Plotman config on the 'Settings | Plotting' page has a placeholder of "REPLACE_WITH_THE_REAL_VALUE". Each time this Settings page is loaded Machinaris looks to see if plotnft has appeared in the blockchain yet and will replace the placeholder with correct "pool contract address" when it can. You still need to uncomment that line in the Plotman config (and comment the `pool_pk` line for solo/OG plotting), then hit Save. If solo/OG plotting is still enabled when you hit Save a warning is displayed to ensure you understand the implications. Most people should be using "pool contract address". Sounds like you got it sorted out though. Let me know if you have further questions or concerns.

Please accept my apologies. There is a defect in Machinaris v0.8.5 that is breaking initial key generation due to changes in the Chia 1.6.1 command structure. I have just pushed a fix to the `machinaris:develop` image, so please try that by switching your Repository line in the Docker configuration for the Machinaris container in the Unraid Admin UI on the Docker tab. You should see a Setup screen, but this time the Generate Key button should not fail. Sorry again.

Perhaps the mount for the mnemonic.txt path (points to the Chia container path) was changed/missing?

Please try `machinaris-flax:develop` image which has the more recent Flax 0.1.11 release and let me know if that helps. The image can be changed on Repository setting line in the Docker configuration screen of Unraid for that container.

Greetings! Sorry for the trouble. Please share the last ~50 lines or so of "/mnt/user/appdata/machinaris/machinaris/logs/webui.log" right after you click the "Generate Key" button. This log should show the issue. If that log file doesn't add new lines on every page load/refresh of the Machinaris, then something more fundamental is wrong such as disk space. Alternatively, there is a way to create a key from the Docker console for Machinaris container, via the Unraid Admin UI, Docker tab: cat /root/.chia/mnemonic.txt # Don't continue if a key mnemonic is already present in this file chia keys show # Don't continue if a key is already shown from above command, or some other error appears chia keys generate chia keys show --show-mnemonic-seed | tail -n 1 > /root/.chia/mnemonic.txt Hope this helps. Feel free to drop into the Discord server for more interactive help if needed.

Greetings! Machinaris v0.8.5 is now available. Changes include: Added - Wallets page - Claim Rewards button offers portable plot reward recovery after one week has elapsed. (EARLY BETA!) - Chart memory usage per container (GiB) as well as total host memory usage (%) for OS and all apps. - Enhanced Forktools to optionally decrease a blockchain's full_node process count, which greatly limits memory usage. - Improve Plotting page to display configured tmp, dst, and archiving directories before starting to plot. Support `site_path_filter` for archive folders under `site_root`. - Bladebit support for new `ramplot` mode (needs 416 GB RAM) Changed - Enhance 'NFT Reward Recovery' tool to support v2 databases. - Fixes for invalid Chiadog harvester alerts. - Fixes for bladebit `diskplot` mode (less than 416 GB RAM, needs 400+GB of tmp disk) Updated - Bladebit to v2.0.0 - Chia to v1.6.1 - Chinilla to v1.3.0 - Littlelambocoin to v1.6.1 - Maize to v1.6.0 - MMX to `testnet8`. - Tad to v1.6.0 Known Issues - Incorrect reward recovery calculations for some blockchains. Please use AllTheBlocks site if this affects you. Links: - https://github.com/Chia-Network/bladebit/releases/tag/v2.0.0 - https://github.com/Chia-Network/chia-blockchain/releases/tag/1.6.1 Big thanks to all who tested and provided feedback!

Thanks for your question. Please see the Machinaris wiki, where site_path_filter (of 'plots') allows one to exclude a drive (like '/mnt/plotting') from being an archive target like others at the same level such as /mnt/plots1, /mnt/plots2, /mnt/plots3 etc. This applies to both local (same machine) and remote (another machine) plot archiving.

Hi, good question. In my experience, Plotman treats `dst` folders as a secondary staging location, so it will attempt to move completed plot files there in a round-robin fashion, without taking drive space into account. Please try Plotman's local archiving feature which is meant to distribute plots to their final destination, taking into account free space amongst the drives locally mounted. Update: In the next release of Machinaris, I've improved the display of Plotman's configured directories/targets as the original yaml file is not easy to troubleshoot. Here's an example from the current `machinaris:develop` image:

Alright, perhaps you have some connectivity error with ATB. Please remove the cache file on your disk: /mnt/user/appdata/machinaris/machinaris/cache/cold_wallet_cache.json

Hi, so the cold wallet balances are pulled from AllTheBlocks. So, to check for yourself, enter the exact same cold wallet address you put into the Machinaris WebUI Wallets page, into the Search box in the top-right of the ATB site to view the wallet balance, at least as known to ATB. Example: https://alltheblocks.net/chia/address/xch123456789 Hope this helps.

Sorry to hear that. Please troubleshoot the `webui.log` to see the cause of the error. Hope this helps.

Greetings! Machinaris v0.8.4 is now available. Changes include: Scaling-Down: Optional mode where wallets are synced daily, not run 24/7. Saves ~35% memory so smaller farmers can farm more blockchains concurrently on the same machine. See Wallets page, top-right Settings. Scaling-Up: Improved plot tracking efficiency for multi-PB sized farms. Thanks @grobalt! Warnings for duplicated, invalid, or key-less plots. See Farming page. Update: Flax to v0.1.10, Cactus to v1.5.2, Chia to v1.6.0 Security: Disable Setup page's mnemonic import field autocomplete from caching value in your local browser. Thanks @Baatezu! Fixes: Better handling of farmed block logging for certain blockchains like Apple & BPX, Alerts from Chia 1.5.1 for added coins missing due to blockchain logging changes. Improved Smartctl response processing. Thanks @gnomuz!

Hi! Please try this from within the Machinaris container: chia wallet send --help This will show the arguments (wallet_id #, amount and target address) you'll need. If moving a lot of XCH, I'd recommend trying one small send transaction first, verifying it works, then sending another with the remainder later.

Good suggestion, this notification feature was actually added to Machinaris v0.8.0, released 3 months ago. However, here are the legacy blockchains still stuck using a v1 format: 'bpx', 'ecostake', 'flora', 'gold', 'hddcoin', 'mint', 'nchain', 'petroleum', 'profit', 'silicoin', 'stor'. For a particular blockchain, visit the Warnings pane of the individual fullnode, by clicking into the worker from the table on main Workers page. If a v1 database is present and a v2 database is *probably* supported, a warning will be shown there. Hope this helps.

Thanks for the report. Happy to look into the issues you are facing with over 110,000 plots in your Chia farm running Machinaris. Please ensure you follow the guidelines for Scaling. Don't hesitate to drop into the Discord for additonal help.

This is also in the ':test' branch as well. Probably safer to test there as lots going on in `:develop`.