Monitoring hardware in KDE (4.9 up) [Solved]

[linuxfluesterer]

Hallo...
I'm searching for Plasmoids or Widgets in KDE 4.9.0 or up to monitor temperatures of especially graphics adapters, hdd and other hardware, which is shipped with a built-in sensor.
In Gnome (Mate maybe also?) there are streamlets.
I've googled for that, but many suggestions are outdated or concerning to plasma-karamba, which failed, when I, e.g. tried to install "plasmapkg --type superkaramba --install 107448-nvidia.skz".
Because my konsole freezed, I fear a mess also...
Any suggestions? Thank you...

-Linuxfluesterer (I love KDE ... )

[Fitzcarraldo]

I'm using the Hardware Temperature widget by Petri Damstén, which comes as standard with KDE. In my case it shows the temperatures of the hardware I am using: the four cores of the Core i7 CPU plus two other temperatures, both of which are apparently the laptop's motherboard temperature. It does not see the HDD temperature, though, so I am also using GKrellM, which is not a widget per se but may as well be. See Re: Notebook running hot [Solved] for how to install the HDD temperature monitoring daemon.



I haven't tried them myself, but the following Plasma widgets, which can be installed using the GUI Plasma Desktop Shell Add-On Installer, would be worth installing to see if they work:

- Radeon Power Management
- Simple CPU Sensor
- Simple NVidia Info v2
- Simple NVidia Info
- CPU Temp
- SH Teplomer
- Very simple CPU temperature
- pyATImonitor
- miminon

[linuxfluesterer]

Hallo Fitzcarraldo.
Thanks a lot for your reply. You really do a great job here.
In last days I tried your suggestions. That's the moments results:
I get displayed informations about cpu temperature and NVidia temperature.
For tests I have run the built in plasma plasmoid 'Hardware Monitor', which (only) shows the CPU temperatures.
Then there is the 'Simple NVidia Info' which works, but 2nd version, I can't install.
1. Then there is a widget with many informations, but I forgot the name and I can't change any settings (see attachm.)
2. And then there is your suggested GKrellM, which I installed also.
In both last programs, there is no harddrive temperature (0°) in widget window,
though when I start '/usr/sbin/hddtemp /dev/sda' I get a console result with (e.g.)
/dev/sda: WDC WD5000BEVT-22A0RT0: 37°C
In my eyes it seems, that hdd temperature is recognized, but can not be read by the plasmoids or widget or GKrellM.
'sensors_detect' does not detect and show me any hdd...

-Linuxfluesterer (I love KDE ...)

[Fitzcarraldo]

Strange, as GKrellM is reading the HDD temperature on my laptop.

(BTW, can you see the image of GKrellM that I posted above? In the hotel I'm in at the moment I can only see the image of the Plasmoids.)

It's so long since I set up GKrellM that I can't recall what I did precisely, but it has been able to read the HDD temperature on my previous main laptop and on my present main laptop. I kept some notes on it when I first set it up on my previous main laptop, so I'll post them below in case they are of any help...

Monitoring HDD temperature

Installation

Merge the package hddtemp to monitor the temperature of SMART drives:

emerge -1v hddtemp

Configuration

Edit the file /etc/conf.d/hddtemp so that the variable HDDTEMP_DRIVES lists the drives to check. For example, here is my hddtemp file:

# Copyright 1999-2006 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
# $Header: /var/cvsroot/gentoo-x86/app-admin/hddtemp/files/hddtemp-conf.d,v 1.3 2006/07/15 01:11:28 spock Exp $

# the hddtemp executable
HDDTEMP_EXEC=/usr/sbin/hddtemp

# various options to pass to the daemon
HDDTEMP_OPTS="-l 127.0.0.1"

# a list of drives to check
#HDDTEMP_DRIVES="/dev/hda /dev/hdb"
HDDTEMP_DRIVES="/dev/sda"

Also look in /usr/share/hddtemp/hddtemp.db and, if my HDD is not listed, add the line:

"HTS541616J9SA00" 194 C "Hitachi Travelstar 5K160GB (5400RPM)"

Starting hddtemp daemon

Also need to start the hddtemp daemon:

# /etc/init.d/hddtemp start

or get the hddtemp daemon to run automatically at boot up by:

# rc-update add hddtemp default

GKrellM

It is not necessary to merge the GKrellm plugin gkrellm-hddtemp any more. To get GKrellM to display the HDD temperature, just merge hddtemp, configure it and start the daemon as described above.

Using from the command line

To use hddtemp from the command line:

# hddtemp --help

# hddtemp /dev/sda
/dev/sda: Hitachi HTS541616J9SA00: 40°C

I used Portage to install app-admin/hddtemp, but it's also in Entropy so can be installed using Equo or Rigo.

[linuxfluesterer]

Hallo Fitzcarraldo...
You did it! I did it! Thank you very much....

though when I start '/usr/sbin/hddtemp /dev/sda' I get a console result with (e.g.)
/dev/sda: WDC WD5000BEVT-22A0RT0: 37°C

I already installed hddtemp (with Rigo), but my hdd was not in database and my try to add my WDC failed
After your last post here, I tried again and searched in Google for my WDC and constructed a new line in /usr/share/hddtem.db. It is:
"WDC WD5000BEVT-22A0RT0" 194 C "Western Digital Scorpio Blue 500GB SATA2 5400RPM 8MB"
I did:

Code: Select all

/etc/init.d/hddtemp restart

then I stopped GKrellM and started it again.
Then I could see my WDC in 'temperature configuration'. I clicked it, and then I saw it...

(BTW, can you see the image of GKrellM that I posted above? In the hotel I'm in at the moment I can only see the image of the Plasmoids.)

I'm sorry, I see only your plasmoid in full size, but from your GKrellM there is only a very small symbol (5x5 mm).

-Linuxfluesterer (I love KDE ...)

[Fitzcarraldo]

Good to know it's working.

I see only your plasmoid in full size, but from your GKrellM there is only a very small symbol (5x5 mm).

I have fixed the link. Can you see it now?

[linuxfluesterer]

Hallo Fitzcarraldo...
Yes, I see it now .. And this is mine now. Is it correct, that I must also start

Code: Select all

hddtemp

every time I boot (so Autostart it)? Because only after start that program (not the daemon!) then I see the hdd temp.



But in Oxygen System Monitor, there is still no display of (any) temperature (I hope, you can recognize on the photo).
I don't know, why there is a temperature in GKrellM and not also in Oxygen System Monitor (wether cpu nor hdd).
Any idea?

Good evening,

-Linuxfluesterer (I love KDE ...)

[Fitzcarraldo]

You shouldn't need to run hddtemp yourself. Is the initscript definitely running?:

Code: Select all

# rc-update show -v | grep hddtemp
              hddtemp |      default                                 
# rc-status default | grep hddtemp
 hddtemp                                                           [  started  ]
#

[linuxfluesterer]

Hallo Fitzcarraldo.
As you can see, it is definitely running:

Code: Select all

# rc-update show -v | grep hddtemp
              hddtemp |      default
# rc-status default | grep hddtemp
 hddtemp                                                           [  started  ]
#


But in GKrell-config Window I can not see any hdd sensor settings, so there is no hdd temperature displayed.


After manual start of hddtemp I see hdd sensor in GKrell config window and can mark it.
Then I need to stop and restart GKrellM, and I will have a hdd temperature in display.
Maybe the sequence while booting is wrong?
But ... another question. Where is the cpu temperature in Oxygen System Monitor?
As you can see in post before, in GKrellM it is present.

Good night,
-Linuxfluesterer (I love KDE ...)

[Fitzcarraldo]

I don't know why you are having to issue the command hddtemp manually if the daemon is already running. However, you could implement a work-around to run the command automatically by using a local initscript. Create a file /etc/local.d/01hddtemp.start containing the following lines:

Code: Select all

#!/bin/bash
# Run the hddtemp command as on this machine the hddtemp daemon alone does not work
hddtemp

and make it executable:

Code: Select all

# chmod +x /etc/local.d/01hddtemp.start

Regarding Oxygen System Monitor 0.7:

If "HDD Temp" is not working, you need to active hddtemp deamon becuase this script use this command to show hard disk temperature : nc localhost 7634 | cut -d'|' -f4

Therefore, did you install the net-analyzer/netcat package? The script cannot use the nc (netcat) command if netcat is not installed. I see that net-analyzer/netcat is not in Entropy, so the alternatives would be:

1. Install net-analyzer/netcat via Portage.

2. The package net-analyzer/netcat6 is in Entropy. The latter would be the command nc6. Therefore you could install net-analyzer/netcat6 and edit the Plasmoid's oxygen.theme file inside the archive file 86664-oxygen.skz file and change "nc" to "nc6" in the script.

3. The package net-analyzer/gnu-netcat is in Entropy. I don't know what the resulting gnu-netcat command would be if you were to install that package. Perhaps also "nc". So you could install that and try it first as-is. If the gnu-netcat is not "nc", you would need to edit the Plasmoid's oxygen.theme file inside the archive file 86664-oxygen.skz file and change "nc" to whatever the command is for gnu-netcat.