-
Notifications
You must be signed in to change notification settings - Fork 62
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
No way to handle sensor paths that change? #17
Comments
This problem has existed for some time now with certain distributions. In theory you can work around this by fixing the load order of the relevant kernel modules, however that can be non-trivial depending on your init system. Solving this properly is surprisingly difficult because there may be multiple instances of a single hwmon driver, all of which have the same name. It might be solvable by allowing hwmon matching either by name, or by PCI ID, so that the user would have to figure out whether an hwmon's name is unique or not... |
I've begun implementing a solution for this problem along with the new YAML config format. It supports specifying only a base path for any given hwmon driver, along with a number of indices that specify which |
This feature is now merged into the current master branch. The YAML config is still missing some documentation, but at least there's an example config that shows how to use sensor/fan paths with indices, which makes the config robust against changes in driver load order. |
Closing since the new feature is merged. Continuing with the new related issues that came up. |
Sorry to comment on a closed issue, but I am having the same issue as described by @lynxite and the fix you mentioned seems to address a different problem.
It is not the temp*_input digit that changes, it is the hwmon* digit. Is there any way to handle when the hwmon base path changes between hwmon2 and hwmon3? |
That's exactly what this feature does. You specify the
And specify indices 1 and 3 to use |
Ok I have it working now. I had read the example config but it wasn't clear to me that in
actually refers to the path |
Thanks a lot for implementing the feature, I was stuck with the same situation. Could you maybe add documentation directly to the config file? How it operates, how the paths in the config file correspond to the paths in the file system and how to set it up. It was hard to understand that it will automatically look for temp[indices]_input in all sub directories of a given base path as this behavior is quite different compared to the other config files. |
This comment has been minimized.
This comment has been minimized.
Created issue #87 to keep track of missing YAML config documentation, because this is a technical issue that should be fixed in the current master and in the 1.1 release. |
This is not exact @bssb . Like vmatare said:
from the man page
for |
@thobianchi Note that the config manpage has been updated several times now, so people that read one of the older versions may well have been confused by some unclear or incomplete descriptions that used to be in there. Also keep in mind that getting feedback from people who misunderstood something is instrumental to improving the quality of documentation ;-) |
After figuring out that thinkfan.conf isn't exactly the same as thinkfan.yaml, yet both can be used interchangeably (every documentary on the web other than the official really is outdated), I was still running into the problem that the sensors on my Laptop couldn't be found upon boot.
Thinkfan and the config work flawlessly after startup. I solved it by delaying the systemd unit with After=multi-user.target |
The path to my CPU's sensors sometimes changes on restart. Specifically, the digit after "hwmon" varies. This results in the expected error from thinkfan:
Does thinkfan support any sort of pattern matching in the config file to accommodate this?
The text was updated successfully, but these errors were encountered: