Temperature missing on UI

[msp2842]

Describe the bug

Temperature monitoring is not displaying in the Pulse dashboard despite sensors being properly configured and accessible. The temperature field shows as blank in the node overview.

Version Information

• Pulse version: v4.34.0
• Platform: LXC container on Proxmox VE
• Proxmox VE version: 8.1.4
• Kernel: 6.8.12-16-pve
• Hardware: Intel CPU (ASUS platform based on sensor output)

Current Behavior

The temperature field in the Pulse dashboard remains blank/empty for the Proxmox node, even though:

• lm-sensors is installed and configured
• sensors command returns valid temperature data
• hwmon devices are present and accessible
• Pulse service is running normally

Steps Taken

1. Installed lm-sensors package
2. Ran sensors-detect --auto
3. Verified coretemp module is loaded: lsmod | grep coretemp shows module is active
4. Confirmed hwmon devices exist:

ls -la /sys/class/hwmon/
total 0
drwxr-xr-x  2 nobody nogroup 0 Nov 29 05:14 .
drwxr-xr-x 85 nobody nogroup 0 Nov 26 07:04 ..
lrwxrwxrwx  1 nobody nogroup 0 Nov 29 05:14 hwmon0 -> ../../devices/virtual/thermal/thermal_zone0/hwmon0
lrwxrwxrwx  1 nobody nogroup 0 Nov 29 05:14 hwmon1 -> ../../devices/pci0000:00/0000:00:06.0/0000:01:00.0/nvme/nvme0/hwmon1
lrwxrwxrwx  1 nobody nogroup 0 Nov 29 05:14 hwmon2 -> ../../devices/platform/asus-nb-wmi/hwmon/hwmon2
lrwxrwxrwx  1 nobody nogroup 0 Nov 29 05:14 hwmon3 -> ../../devices/virtual/thermal/thermal_zone1/hwmon3
lrwxrwxrwx  1 nobody nogroup 0 Nov 29 05:14 hwmon4 -> ../../devices/platform/coretemp.0/hwmon/hwmon4

5. Restarted Pulse service: systemctl restart pulse.service
6. Refreshed browser (hard refresh with cache clear)

Sensors Output

The sensors command shows all temperatures correctly:

coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +48.0°C  (high = +100.0°C, crit = +100.0°C)
Core 0:        +41.0°C  (high = +100.0°C, crit = +100.0°C)
Core 4:        +43.0°C  (high = +100.0°C, crit = +100.0°C)
Core 8:        +44.0°C  (high = +100.0°C, crit = +100.0°C)
Core 12:       +47.0°C  (high = +100.0°C, crit = +100.0°C)
Core 16:       +45.0°C  (high = +100.0°C, crit = +100.0°C)
Core 17:       +45.0°C  (high = +100.0°C, crit = +100.0°C)
Core 18:       +45.0°C  (high = +100.0°C, crit = +100.0°C)
Core 19:       +46.0°C  (high = +100.0°C, crit = +100.0°C)
Core 20:       +45.0°C  (high = +100.0°C, crit = +100.0°C)
Core 21:       +45.0°C  (high = +100.0°C, crit = +100.0°C)
Core 22:       +45.0°C  (high = +100.0°C, crit = +100.0°C)
Core 23:       +45.0°C  (high = +100.0°C, crit = +100.0°C)

asus-isa-0000
Adapter: ISA adapter
cpu_fan:     1800 RPM

acpitz-acpi-0
Adapter: ACPI interface
temp1:        +48.0°C  

iwlwifi_1-virtual-0
Adapter: Virtual device
temp1:            N/A  

nvme-pci-0100
Adapter: PCI adapter
Composite:    +37.9°C  (low  = -273.1°C, high = +81.8°C)
                       (crit = +84.8°C)
Sensor 1:     +37.9°C  (low  = -273.1°C, high = +65261.8°C)
Sensor 2:     +38.9°C  (low  = -273.1°C, high = +65261.8°C)

Expected Behavior

The Pulse dashboard should display CPU and NVMe temperature readings in the temperature field, similar to how CPU, Memory, and Disk usage are displayed.

Screenshots

![Pulse Dashboard - Missing Temperature]
Temperature field is blank in the node overview

![Pulse Settings - Node Configuration]
Node is properly configured and online

Additional Context

This appears similar to issues #536 and #600. The sensors are working correctly at the OS level, but Pulse is not parsing or displaying the temperature data in the UI.

Service Status

systemctl status pulse.service
● pulse.service - Pulse Monitoring Server
     Loaded: loaded (/etc/systemd/system/pulse.service; enabled; preset: enabled)
     Active: active (running) since Sat 2025-11-29 05:19:37 EST

Pulse service logs show no temperature-related errors:

journalctl -u pulse.service -n 100 --no-pager | grep -i "sensor\|temp\|hwmon"
# No output

[rcourtman]

For temperature monitoring in an LXC container, Pulse requires the pulse-sensor-proxy to be installed on the Proxmox host. The proxy reads sensor data and exposes it via a Unix socket that gets bind-mounted into the container.

The fact that journalctl -u pulse.service | grep -i temp shows no output confirms the proxy is not connected.

Setup steps:

1. Go to Settings → Nodes → [Your Node] → Advanced Monitoring and enable "Temperature Monitoring"

2. Run the Setup Script from the Pulse UI, which auto-installs the proxy on the host

3. After installation, verify on the Proxmox host:

   systemctl status pulse-sensor-proxy
   ls -l /run/pulse-sensor-proxy/pulse-sensor-proxy.sock

4. Check that Pulse sees the proxy:

   journalctl -u pulse | grep -i proxy
   # Should show: "Temperature proxy detected - using secure host-side bridge"

See TEMPERATURE_MONITORING.md for full documentation.

[Pouch11]

I executed this command and it solved my issue

curl -fsSL https://github.com/rcourtman/Pulse/releases/latest/download/install-sensor-proxy.sh |
sudo bash -x -s -- --standalone --http-mode --pulse-server http://YOUR-PULSE-IP:7655

[msp2842]

@Pouch11 thank you for the info as that is what actually worked in my situation!

[Prince11592]

my cluster temperatures appear and disappear (they just flicker among the three nodes) any fix for this

[rcourtman]

@Prince11592 That's a different issue from the original report - the flickering suggests inconsistent proxy connectivity across nodes.

To diagnose:

1. Is sensor-proxy installed on all cluster nodes? For a cluster, each node needs the proxy to report its own temperatures.

2. Check proxy health on each node:

   systemctl status pulse-sensor-proxy
   journalctl -u pulse-sensor-proxy -n 20 --no-pager

3. From the Pulse host, can you reliably reach each node's proxy?

   • For socket mode: verify the socket is mounted and accessible
   • For HTTP mode: curl -k https://<node-ip>:8443/api/temperature -H "Authorization: Bearer <token>"

The flickering pattern (rotating between nodes) suggests either network intermittency or the proxy entering cooldown states after failures on specific hosts.

[Prince11592]

Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Validated and normalized subnet configuration allowed_subnets=["10.100.10.153/32","127.0.0.1/32","10.100.10.5/32"]
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Rate limit configuration loaded from config file per_peer_burst=10 per_peer_interval_ms=333
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Audit logger initialized with file backend mode=file path=/var/log/pulse/sensor-proxy/audit.log
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Node allow-list configured allowed_cidr_count=0 allowed_node_count=4
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Starting pulse-sensor-proxy audit_log=/var/log/pulse/sensor-proxy/audit.log config_path=/etc/pulse-sensor-proxy/config.yaml log_level=info socket=/run/pulse-sensor-proxy/pulse-sensor-proxy.sock ssh_key_dir=/var/lib/pulse-sensor-proxy/ssh version=v4.32.0
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Loaded explicit peer UID allow-list entries explicit_uid_allow_count=1
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Loaded subordinate ID ranges for ID-mapped root authentication gid_range_count=1 uid_range_count=1 users=["root"]
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF SSH keypair already exists
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Unix socket ready socket=/run/pulse-sensor-proxy/pulse-sensor-proxy.sock
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Control plane synchronization enabled refresh_interval=60 url=http://10.100.10.153:7655
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Starting HTTPS server addr=:8443 cert=/etc/pulse-sensor-proxy/tls/server.crt
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM INF Metrics server started addr=127.0.0.1:9127
Nov 30 01:02:00 HPE1 pulse-sensor-proxy[2681]: 1:02AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764444720}"
Nov 30 01:03:00 HPE1 pulse-sensor-proxy[2681]: 1:03AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764444780}"
Nov 30 01:04:00 HPE1 pulse-sensor-proxy[2681]: 1:04AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764444840}"
Nov 30 01:05:00 HPE1 pulse-sensor-proxy[2681]: 1:05AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764444900}"
Nov 30 01:06:00 HPE1 pulse-sensor-proxy[2681]: 1:06AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764444960}"
Nov 30 01:07:00 HPE1 pulse-sensor-proxy[2681]: 1:07AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764445020}"
Nov 30 01:08:00 HPE1 pulse-sensor-proxy[2681]: 1:08AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764445080}"
Nov 30 01:09:00 HPE1 pulse-sensor-proxy[2681]: 1:09AM WRN Control plane sync failed error="control plane responded 401: {"error":"Proxy token not recognized","code":"invalid_token","status_code":401,"timestamp":1764445140}"

[Prince11592]

if i use HTTPS proxy this happens and running pulse in lxc container

[rcourtman]

@Prince11592 The screenshot and logs show two separate issues:

1. Sensor proxy token not recognized
The logs show "Proxy token not recognized" - this happens when the sensor-proxy was installed with a token that Pulse doesn't recognize. This can occur if:

• Pulse was reinstalled/reset and the token database was cleared
• The proxy was installed from a different Pulse instance

Fix: Re-run the sensor-proxy installer from the Pulse UI (Settings → Nodes → [Node] → Temperature → Setup Script). This generates a fresh token that Pulse recognizes.

2. Physical disks API timeout
The screenshot shows "context deadline exceeded" for /nodes/.../disks/list. This is unrelated to the temperature flickering you mentioned earlier. The physical disks API can be slow on some hardware, causing timeouts.

If you don't need physical disk monitoring, you can disable it in Settings → Nodes → [Node] → Advanced Monitoring → Physical Disks.

For the temperature flickering across your 3 cluster nodes:
After fixing the token issue above, each node needs sensor-proxy installed. Since Pulse is in an LXC on one of these nodes, you need:

• Socket mode for the local host (where the LXC lives)
• HTTPS mode for the remote nodes

Re-run the setup from each node's Advanced Monitoring section in the UI.

[hughfr4nc15]

I am having some issues with getting this to work too... :|

Executed this on the Proxmox host:

curl -fsSL https://github.com/rcourtman/Pulse/releases/latest/download/install-sensor-proxy.sh | sudo bash -x -s -- --ctid 191 --pulse-server http://192.168.1.91:7655

Install script created this token (I assume):

Service seems to be running:

Temps not showing:

Any ideas?
Thanks in advance.

[rcourtman]

@hughfr4nc15 The sensor-proxy is running - that's good. A few things to check:

1. Verify socket is accessible in the container

From inside the Pulse container (CTID 191):

ls -la /mnt/pulse-proxy/

You should see pulse-sensor-proxy.sock

2. Check Pulse sees the proxy

Look at Pulse logs:

# Inside the Pulse container
journalctl -u pulse --no-pager | grep -i "proxy\|temperature"

Should show something like: "Temperature proxy detected - using secure host-side bridge"

3. Verify temperature monitoring is enabled

In Pulse UI: Settings → Nodes → MAINFRAME → ensure "Temperature Monitoring" is enabled in the Advanced Monitoring section.

4. Check container config has the mount

On the Proxmox host:

pct config 191 | grep -E "lxc.mount|mp[0-9]"

Should show:

lxc.mount.entry: /run/pulse-sensor-proxy mnt/pulse-proxy none bind,create=dir 0 0

If the mount isn't there, rerun the installer or add it manually. If everything looks correct but still no temps, please share the output from commands 1-4 above.

[hughfr4nc15]

Humm... It seems several things are not as intended:

Dec 02 20:33:37 pulse pulse[91]: {"level":"info","component":"pulse-config","guest":"proxy","cpu":0.345654746655224,"memory":13.138961791992188,"disk":52.78231386659286,"thresholds":{"poweredOffSeverity":"warning","cpu":{"trigger":80,"clear":75},"memory":{"trigger":85,"clear":80},"disk":{"trigger":90,"clear":85}},"time":"2025-12-02T20:33:37Z","message":"Checking guest thresholds"}
Dec 02 20:33:47 pulse pulse[91]: {"level":"info","component":"pulse-config","guest":"proxy","cpu":0.0827678666468095,"memory":13.086700439453125,"disk":52.78251156537899,"thresholds":{"poweredOffSeverity":"warning","cpu":{"trigger":80,"clear":75},"memory":{"trigger":85,"clear":80},"disk":{"trigger":90,"clear":85}},"time":"2025-12-02T20:33:47Z","message":"Checking guest thresholds"}
Dec 02 20:33:57 pulse pulse[91]: {"level":"info","component":"pulse-config","guest":"proxy","cpu":1.71939344476121,"memory":13.064956665039062,"disk":52.78261041477206,"thresholds":{"poweredOffSeverity":"warning","cpu":{"trigger":80,"clear":75},"memory":{"trigger":85,"clear":80},"disk":{"trigger":90,"clear":85}},"time":"2025-12-02T20:33:57Z","message":"Checking guest thresholds"}
Dec 02 20:34:07 pulse pulse[91]: {"level":"info","component":"pulse-config","guest":"proxy","cpu":0.23855331094232302,"memory":13.108062744140625,"disk":52.782808113558175,"thresholds":{"poweredOffSeverity":"warning","cpu":{"trigger":80,"clear":75},"memory":{"trigger":85,"clear":80},"disk":{"trigger":90,"clear":85}},"time":"2025-12-02T20:34:07Z","message":"Checking guest thresholds"}
Dec 02 20:34:17 pulse pulse[91]: {"level":"info","component":"pulse-config","guest":"proxy","cpu":0.0827722671593623,"memory":13.108444213867188,"disk":52.782906962951245,"thresholds":{"poweredOffSeverity":"warning","cpu":{"trigger":80,"clear":75},"memory":{"trigger":85,"clear":80},"disk":{"trigger":90,"clear":85}},"time":"2025-12-02T20:34:17Z","message":"Checking guest thresholds"}
Dec 02 20:34:27 pulse pulse[91]: {"level":"info","component":"pulse-config","guest":"proxy","cpu":0.0876373256130041,"memory":13.084793090820312,"disk":52.78300581234431,"thresholds":{"poweredOffSeverity":"warning","cpu":{"trigger":80,"clear":75},"memory":{"trigger":85,"clear":80},"disk":{"trigger":90,"clear":85}},"time":"2025-12-02T20:34:27Z","message":"Checking guest thresholds"}
Dec 02 20:34:31 pulse pulse[91]: {"level":"info","component":"pulse-config","json":"{\"pveInstances\":[{\"Name\":\"MAINFRAME\",\"Host\":\"http://192.168.1.99:8006\",\"GuestURL\":\"https://mainframe.fr4nc15.co\",\"User\":\"\",\"Password\":\"\",\"TokenName\":\"pulse-api@pam!pulse-token\",\"TokenValue\":\"aa6518f3-d7fc-48e3-a0a3-5c2aa60c695c\",\"Fingerprint\":\"\",\"VerifySSL\":false,\"MonitorVMs\":true,\"MonitorContainers\":true,\"MonitorStorage\":true,\"MonitorBackups\":true,\"MonitorPhysicalDisks\":false,\"PhysicalDiskPollingMinutes\":0,\"TemperatureMonitoringEnabled\":true,\"SSHPort\":0,\"TemperatureProxyURL\":\"\",\"TemperatureProxyToken\":\"\",\"TemperatureProxyControlToken\":\"gGRZADLjjeganLxHoYTivZR8dsGxKKcSGW2-SB7mY2M=\",\"IsCluster\":false,\"ClusterName\":\"\",\"ClusterEndpoints\":null}],\"pbsInstances\":[],\"pmgInstances\":[]}","time":"2025-12-02T20:34:31Z","message":"DEBUG: Loaded nodes config JSON"}

3. Yes, it is.

Not sure if is of interest but sensors seems to be working fine:

[rcourtman]

@hughfr4nc15 The issue is clear from your debugging output.

Problem: The container config shows lxc.cgroup2.devices.allow: c 10:200 rwm but it's missing the mount entry. The installer should have added:

lxc.mount.entry: /run/pulse-sensor-proxy mnt/pulse-proxy none bind,create=dir 0 0

This explains why /mnt/pulse-proxy/ is empty - the socket directory from the host isn't being mounted into the container.

To fix:

1. On the Proxmox host, add the mount entry to the container config:

   echo "lxc.mount.entry: /run/pulse-sensor-proxy mnt/pulse-proxy none bind,create=dir 0 0" >> /etc/pve/lxc/191.conf

2. Restart the container:

   pct stop 191 && pct start 191

3. Verify the socket is now visible inside the container:

   pct exec 191 -- ls -la /mnt/pulse-proxy/

   Should show pulse-sensor-proxy.sock

4. Restart Pulse inside the container:

   pct exec 191 -- systemctl restart pulse

The installer should have done this automatically. If you ran it via --ctid 191 it should have detected and modified the config. Could you share what the /etc/pve/lxc/191.conf looks like currently? There might be a reason the mount entry wasn't added (e.g., pmxcfs cluster sync timing issue).

[hughfr4nc15]

Success! 👍 Thanks a lot for all the support. As requested I have attached the conf file before the changes.

191.txt

[rcourtman]

Thanks for confirming and sharing the config file - the install worked but I can see why the mount entry wasn't added initially.

Your container config before the fix shows lxc.cgroup2.devices.allow but no lxc.mount.entry. The installer should have added this automatically. This appears related to issue #738 where the mount entry addition is sometimes skipped.

Could you share how you initially ran the installer? Specifically:

1. Did you use curl ... | bash or download and run the script directly?
2. Did the installer output show "Configuring socket bind mount..." at any point?

This will help track down why the mount entry isn't being added in some cases.