Package: prometheus-node-exporter Version: 1.9.1-1 Severity: normal X-Debbugs-Cc: [email protected]
Dear Maintainer, I recently deployed the prometheus-node-exporter package onto a freshly upgraded Debian 13 trixie installation. On this server, most of the underlying storage is mounted using the ZFS filesystem from the respective Debian packages. I started the exporter (temporarily upgraded to testing) via: `prometheus-node-exporter --web.listen-address=:1234` ...and then accessed the metrics endpoint via http://localhost:1234/metrics. Only the following gauges are reported (xxx was removed): ``` # HELP node_filesystem_avail_bytes Filesystem space available to non-root users in bytes. # TYPE node_filesystem_avail_bytes gauge node_filesystem_avail_bytes{device="/dev/mapper/sda3_crypt",device_error="",fstype="ext4",mountpoint="/"} xxx node_filesystem_avail_bytes{device="/dev/nvme0n1p1",device_error="",fstype="vfat",mountpoint="/boot/efi"} xxx node_filesystem_avail_bytes{device="/dev/nvme0n1p2",device_error="",fstype="ext4",mountpoint="/boot"} xxx node_filesystem_avail_bytes{device="tmpfs",device_error="",fstype="tmpfs",mountpoint="/tmp"} xxx ``` Okay... Then I pulled the upstream package, built it and started it via: ``` git clone https://github.com/prometheus/node_exporter.git cd node_exporter/ git checkout v1.9.1 make build ./node_exporter --web.listen-address=:1235 ``` ...and then accessed the metrics endpoint via http://localhost:1235/metrics. Here the following gauges are reported: ``` # HELP node_filesystem_avail_bytes Filesystem space available to non-root users in bytes. # TYPE node_filesystem_avail_bytes gauge node_filesystem_avail_bytes{device="/dev/mapper/sda3_crypt",device_error="",fstype="ext4",mountpoint="/"} xxx node_filesystem_avail_bytes{device="/dev/nvme0n1p1",device_error="",fstype="vfat",mountpoint="/boot/efi"} xxx node_filesystem_avail_bytes{device="/dev/nvme0n1p2",device_error="",fstype="ext4",mountpoint="/boot"} xxx node_filesystem_avail_bytes{device="some_zfs_dataset/Archive",device_error="",fstype="zfs",mountpoint="/mnt/some_zfs_dataset_volatile/Archive"} xxx node_filesystem_avail_bytes{device="some_zfs_dataset/Data1",device_error="",fstype="zfs",mountpoint="/mnt/some_zfs_dataset_b/data1"} xxx node_filesystem_avail_bytes{device="some_zfs_dataset/Data2",device_error="",fstype="zfs",mountpoint="/mnt/some_zfs_dataset_a/data2"} xxx node_filesystem_avail_bytes{device="some_zfs_dataset/Data3",device_error="",fstype="zfs",mountpoint="/mnt/some_zfs_dataset_b/data3"} xxx <and many, many more> ``` So, for some reason the packaged variant behaves differently in reporting the filesystems. As no systemd service was used for these experiments, its system-isolation should not be the issue. Maybe AppArmor is the problem here? Using the flag "--collector.filesystem.fs-types-exclude" with some non-matching RegEx also did not help. The debug-logs via "--log.level=debug" also did not show anything useful. I have quickly taken a look into the manual-pages and did a quick internet-search, but did not find a specific hint, why ZFS has this problem. Maybe I'm just doing something wrong... Thank you very much in advance, simonmicro -- System Information: Debian Release: 13.1 APT prefers stable-updates APT policy: (500, 'stable-updates'), (500, 'stable-security'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 6.12.41+deb13-amd64 (SMP w/64 CPU threads; PREEMPT) Kernel taint flags: TAINT_PROPRIETARY_MODULE, TAINT_OOT_MODULE, TAINT_UNSIGNED_MODULE Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) (ignored: LC_ALL set to en_US.UTF-8), LANGUAGE=en_US:en Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled Versions of packages prometheus-node-exporter depends on: ii adduser 3.152 ii init-system-helpers 1.69~deb13u1 ii libc6 2.41-12 Versions of packages prometheus-node-exporter recommends: ii dbus 1.16.2-2 ii prometheus-node-exporter-collectors 0.0~git20241119.a2b43e1-1 prometheus-node-exporter suggests no packages. -- Configuration Files: /etc/default/prometheus-node-exporter changed: ARGS="--web.listen-address=:1203 --collector.cgroups" -- no debconf information
