diff --git a/docs/getting-started/timelords.md b/docs/getting-started/timelords.md index 0dfe8ef01e..20c85735ea 100644 --- a/docs/getting-started/timelords.md +++ b/docs/getting-started/timelords.md @@ -10,7 +10,7 @@ import TabItem from '@theme/TabItem'; **DO NOT** overclock ASICs, overclocking diminishes the life of the ASIC! ::: -Timelord architecture information can be found [here](/timelord-architecture) +Timelord architecture information can be found [here](/timelord-architecture). The hw_vdf_client parameter information can be found [here](/asic-cli). --- @@ -18,7 +18,7 @@ The hw_vdf_client parameter information can be found [here](/asic-cli). ## Timelord Requirements and Dependencies :::info -Due to restrictions on how [MSVC](https://en.wikipedia.org/wiki/Microsoft_Visual_C%2B%2B) handles 128 bit numbers and how Python relies upon MSVC, it is not possible to build and run Timelords of all types on Windows. +Due to restrictions on how [MSVC](https://en.wikipedia.org/wiki/Microsoft_Visual_C%2B%2B) handles 128 bit numbers and how Python relies upon MSVC, it is not possible to build and run Timelords of all types on Windows. Running a timelord on a farming machine will reduce the efficiency of the farmer and the timelord, for this reason it is recommended to have a dedicated machine for running timelords. ::: @@ -113,7 +113,7 @@ chia start node timelord & :::warning -**DO NOT** overclock ASICs, overclocking diminishes the life of the ASIC! +**DO NOT** overclock ASICs, overclocking diminishes the life of the ASIC! Detailed information about the hw_vdf_client parameters can be found [here](/asic-cli). ::: @@ -147,7 +147,7 @@ chia start node timelord-only ### Installing a Timelord from Source :::info -On MacOS x86_64 and all Linux distributions, building a Timelord is as easy as running `chia start timelord &` in the virtual environment. You can also run `./vdf_bench square_asm 400000` once you've built Timelord to give you a sense of your optimal and unloaded ips. Each run of `vdf_bench` can be surprisingly variable and, in production, the actual ips you will obtain will usually be about 20% lower due to load of creating proofs. The default configuration for Timelords is good enough to just let you start it up. Set your log level to INFO and then grep for "Estimated IPS:" to get a sense of what actual ips your Timelord is achieving. +On MacOS x86_64 and all Linux distributions, building a Timelord is as easy as running `chia start timelord &` in the virtual environment. You can also run `./vdf_bench square_asm 400000` once you've built Timelord to give you a sense of your optimal and unloaded ips. Each run of `vdf_bench` can be surprisingly variable and, in production, the actual ips you will obtain will usually be about 20% lower due to load of creating proofs. The default configuration for Timelords is good enough to just let you start it up. Set your log level to INFO and then grep for "Estimated IPS:" to get a sense of what actual ips your Timelord is achieving. Detailed information about the hw_vdf_client parameters can be found [here](/asic-cli). ::: @@ -208,7 +208,7 @@ chia start node timelord-only ## ASIC Timelord Systemd Setup -Below is an example of a systemd service file to run the ASIC hw vdf processes. +Below is an example of a systemd service file to run the ASIC hw vdf processes. NOTE - make sure to replace `USERNAME` with your system's username. ```bash @@ -290,7 +290,7 @@ This means that the ASIC cluster will always have an advantage but there are tim While one can overclock the ASIC we very strongly recommend against doing such. Overclocking the ASICs will lead to diminishing longevity of the machine and only provides a minor increase in performance making it inefficient to overclock an ASIC. -## What Voltage Should I Use for an ASIC Timelord? +### What Voltage Should I Use for an ASIC Timelord? It is highly recommend to use the default `0.88` voltage. diff --git a/docs/troubleshooting/timelords.md b/docs/troubleshooting/timelords.md index aec5267e46..1a78521cec 100644 --- a/docs/troubleshooting/timelords.md +++ b/docs/troubleshooting/timelords.md @@ -13,24 +13,24 @@ Make sure to set your logs to INFO for the below commands to work. To do so run: `chia configure --log-level INFO` ::: -The below commands are for use with linux OS. -The `tail` command only pulls new information as it is added to the log files allowing one to see new information as it is added. +The below commands are for use with linux OS. +The `tail` command only pulls new information as it is added to the log files allowing one to see new information as it is added. The `cat` command pulls all current data from the log files providing a snapshot of where it was and is currently. The matched strings are missing the first letter (E or S) intentionally. This is due to the two instances of each log, for example the log where others create the proof of time has the string "Not skipping" while the log where your machine creates the PoT has the string "Skipping". -Continuous: +Continuous: `tail -F ~/.chia/mainnet/log/debug.log | grep "stimated"` : running output of the current estimated IPS as seen by the network (note this will be lower than the IPS reported by the ASIC software). `tail -F ~/.chia/mainnet/log/debug.log | grep "kipping"` : running output of peak heights being added to the node. If a height is skipped than very likely your timelord created the PoT, if the height is not skipped then a different timelord created the PoT. -Static: +Static: `cat ~/.chia/mainnet/log/debug.log | grep "stimated"` : static output of the current estimated IPS as seen by the network (note this will be lower than the IPS reported by the ASIC software). `cat ~/.chia/mainnet/log/debug.log | grep "kipping"` : statis output of peak heights added to the node. If a height is skipped than very likely your timelord created the PoT, if the height is not skipped then a different timelord created the PoT. If you have set up the timelord services to run through systemD you can use the below command(s) for pulling those records. -If running as a service: +If running as a service: `journalctl -f -u chia-hw-vdf | grep "MHz"` : output of the ASIC timelord reported IPS (this will be higher than the estimated IPS as seen by the network). If not using systemD this information is available in the the terminal output where the `hw_vdf_client` command was run. ## Common Timelord Error Logs @@ -45,12 +45,12 @@ timelord chia.timelord.timelord : ERROR Error while handling message: Trace File "chia/timelord/timelord.py", line 221, in _stop_chain ``` -This error occurs when the ASIC software has been shutdown unexpectedly and will start spamming your log file. +This error occurs when the ASIC software has been shutdown unexpectedly and will start spamming your log file. The current resolution is to restart the chia processes with something like `chia start full_node timelord-only -r`. ### ASIC logs -`Warning: too much work for VDF aux threads` - This generally means that the frequency is set too high, it is recommended to use the auto-frequency flag for the software to self adjust and resolve this error `--auto-freq 60` updates the frequency every 60 seconds. +`Warning: too much work for VDF aux threads` - This generally means that the frequency is set too high, it is recommended to use the auto-frequency flag for the software to self adjust and resolve this error `--auto-freq 60` updates the frequency every 60 seconds. `Bad data size after stop: -1` - This is a known logging error related to how the chia client and ASIC disconnect after a stop. There is no concern with having this error in your logs. ---