---
url: 'http://sey.gr/node/51'
title: 'The gauges almost certainly show N/A because every disk query filters'
author:
  name: caseyg
  url: 'http://sey.gr/user/1'
date: '2026-11-16T23:37:59+00:00'
updated: '2026-11-16T23:38:24+00:00'
type: article
published: true
---
    Submitted by caseyg on Mon, 11/16/2026 - 15:37 

 The gauges almost certainly show N/A because every disk query filters on &#96;mountpoint="/rootfs"&#96;, and the node-exporter you run now never reports that mountpoint. I couldn't confirm this against the live Prometheus because it needs the basic-auth login. The cause comes from reading the config and the node\_exporter source.

\*\*Why:\*\* commit &#96;a49dca6&#96; (July 30) switched node-exporter from &#96;stefanprodan/swarmprom-node-exporter:v0.16.0&#96; to &#96;prom/node-exporter:v1-busybox&#96;. The old version reported the host disk as &#96;/rootfs&#96;. Version 1.x reports the host's real mountpoints (&#96;/&#96;, &#96;/boot&#96; and so on) and only adjusts paths if &#96;--path.rootfs&#96; is set. Your service in &#96;swarmprom/swarmprom.yml&#96; still mounts &#96;/:/rootfs:ro&#96; but doesn't pass that flag. So every query in &#96;swarmprom/templates/swarmprom-disk-dash.json.j2&#96; and the per-node gauges in &#96;\_node\_disk\_gauges.json.j2&#96; comes back empty. That's 12 filters across the two files, covering the summary tiles, gauges, graphs and table. Only the Disk I/O panels still work.

\*\*Suggested fix:\*\*  
1\. In the node-exporter command in &#96;swarmprom/swarmprom.yml&#96;, add &#96;--path.rootfs=/rootfs&#96;. Also rename the deprecated flag to &#96;--collector.filesystem.mount-points-exclude&#96;, which 2.0 will require, and add &#96;rootfs&#96; to its exclude pattern:  
&#96;&#96;&#96;diff  
-exec /bin/node\_exporter --path.procfs=/host/proc --path.sysfs=/host/sys --collector.textfile.directory=/tmp/node-meta --collector.filesystem.ignored-mount-points=''^/(sys\|proc\|dev\|host\|etc)($$\|/)'' --no-collector.ipvs  
+exec /bin/node\_exporter --path.rootfs=/rootfs --path.procfs=/host/proc --path.sysfs=/host/sys --collector.textfile.directory=/tmp/node-meta --collector.filesystem.mount-points-exclude=''^/(sys\|proc\|dev\|host\|etc\|rootfs)($$\|/)'' --no-collector.ipvs  
&#96;&#96;&#96;  
2\. In both dashboard templates, change &#96;mountpoint=\\"/rootfs\\"&#96; to &#96;mountpoint=\\"/\\"&#96;:  
&#96;&#96;&#96;  
sed -i 's\|mountpoint=\\\\"/rootfs\\\\"\|mountpoint=\\\\"/\\\\"\|g' swarmprom/templates/swarmprom-disk-dash.json.j2 swarmprom/templates/\_node\_disk\_gauges.json.j2  
&#96;&#96;&#96;  
 Optionally, switch &#96;node\_filesystem\_free\_bytes&#96; to &#96;node\_filesystem\_avail\_bytes&#96; so the gauges match what &#96;df&#96; shows.  
3\. Redeploy with the swarmprom play so &#96;docker stack deploy&#96; updates node-exporter.

\*\*A second possible problem:\*\* as committed, the node-exporter command writes to &#96;/etc/node-exporter&#96; before starting the exporter. If the busybox image has no such directory, that write fails and node\_exporter never starts. Your uncommitted &#96;swarmprom.yml&#96; already fixes this with &#96;/tmp/node-meta&#96;, and it also adds the &#96;node\_name&#96; label that the per-node gauges need. Make sure that version is the one deployed.

\*\*To confirm with your Prometheus login:\*\*  
\- &#96;count by (mountpoint) (node\_filesystem\_size\_bytes)&#96; should show no &#96;/rootfs&#96; today.  
\- After the fix, &#96;count by (node\_name) (node\_filesystem\_size\_bytes{mountpoint="/"})&#96; should list every node.  
\- If a single gauge still shows N/A, check &#96;up{job="node-exporter"}&#96; for a down target, for example port 9100 blocked by the firewall.

I haven't changed anything in the repo, and I can make these edits on a branch and open an MR if you want. Separately, &#96;swarmprom/defaults/main.yml&#96; has the Grafana and Prometheus admin password committed in plain text.



- [Log in](/user/login?destination=/node/51%23comment-form) to post comments