Johannes 'fish' Ziemke
Johannes 'fish' Ziemke
Adding a flag to include/exclude sensors makes sense. Not sure if parallelizing this is worth it, but given we already do it for mount point it's probably fine.
To fix the failing tests, you probably need to update the fixtures to include the values you expect: https://github.com/prometheus/node_exporter/tree/master/collector/fixtures
Maybe a quoting issue? Can you try `--collector.textfile.directory=/var/lib/node_exporter/textfile_collector` instead? Maybe compose adds these a literal. Either way, I don't think this is an problem. We might want to consider logging...
Oh I think the issue is simply that we don't add the --path.rootfs prefix to --collector.textfile.directory.. @SuperQ Should we? I see how this is surprising.. @mrpaulmanley How do you know...
@SuperQ yes, that's why I'm also not sure we should do it. The problem is just that it's not clear to users where `--path.rootfs` is prefixed and where not, as...
I don't think it's feasible to maintain such list given the dynamic nature of the metrics exposed. You can always see the metrics by using curl on the /metrics endpoint
Sorry for not seeing this ealier! > 1. Keep the ioctl calls here, or try and use https://github.com/dennwc/btrfs? Using https://github.com/dennwc/btrfs sounds like a good idea. > 2. Linux 5.14 adds...
What we usually do in these situations is expose a info metric that can be joined with the actual metrics. Something like `device_info{id="2dab4553-c77e-4bd7-92f1-fd916b08093d", device="/dev/sda"} 1`. But yeah, we'd need to...
Related: #295 This seems to be a real problem. Whether we expose all metrics per namespace or just sum them up, we need to handle t his somehow.. @SuperQ wdyt?
@adityaborgaonkar Yes, add the parsing to procfs and after merging it you can bump the procfs dependency in go.mod here