How do i get the Status Monitor (Temp,Load,Memory) back to Login Page

[Klaus]

These Informations are great for a sanity-check at a glance without beeing forced to login via ssh to check those parameters.
These parameters are shown as soon as you inserted the admin password.

[DanSchaper]

@Klaus

These parameters are shown as soon as you inserted the admin password.

Yes, and I believe they shouldn't be there at all. If you want something that will keep those stats and do a vastly better job at it then check out GitHub - XavierBerger/RPi-Monitor: Real time monitoring for embedded devices

[Starfoxfs]

I installed RPI Monitor so, thanks

[RedStabilo]

But you cannot ignore what several users want in your community. Or maybe you can, as you are one of the developers.

The reasons given are that it is not secure when pi-hole is accessible through the internet. It is textbook that you should not allow applications like these to be accessed through the internet so in my humble opinion it's not a valid reason. Exposing the pi-hole to the internet is insecure, not the graphs on the landing page. So why punish users who are doing nothing insecure in the first place.

Make it optional. Everyone wins. Users who expose pi-hole to the internet should make it more secure (with a VPN for instance), not the other way around.

[PromoFaux]

@RedStabilo

But you cannot ignore what several users want in your community.

@RedStabilo

So why punish users who are doing nothing insecure in the first place.

Just going to leave this here (tongue wedged firmly in cheek ):

Workflow

[DanSchaper]

@RedStabilo

The reasons given are that it is not secure when pi-hole is accessible through the internet.

Not really. The reasons given is that it's not information that should have ever been there in the first place. Again, if you want a system monitor then use something that is designed for monitoring systems.

The decision has really nothing to do with security. All of that metric information is going to be removed from the web interface in it's entirety so take this as a nudge to change your workflow now before the feature is completely removed.

Edit:

To add some more clarity. Our web interface shows you a snapshot of what is happening at that exact moment, you have to refresh to see new information and there is no history kept. Transient issues would never be seen and you wouldn't really ever know the true health of the server. Contrast that with RPIMonitor that uses RRDs to store all kinds of information about the health and status of the server. RPM does a vastly better job at monitoring so I feel that you should use the better tool and not one that is kind of accurate, maybe, if you're running on a platform that exposes the data.

[deHakkelaar]

Also not re-inventing the wheel, there is the monit package that can monitor all sorts of metrics and notify "on event" via mail (or do some other action).
Which is more practical instead of checking/F5-ing some webGUI:

pi@ph5b:~ $ apt show monit
[..]
Description: utility for monitoring and managing daemons or similar programs
 monit is a utility for monitoring and managing daemons or similar
 programs running on a Unix system. It will start specified programs
 if they are not running and restart programs not responding.
 .
 monit supports:
  * Daemon mode - poll programs at a specified interval
  * Monitoring modes - active, passive or manual
  * Start, stop and restart of programs
  * Group and manage groups of programs
  * Process dependency definition
  * Logging to syslog or own logfile
  * Configuration - comprehensive controlfile
  * Runtime and TCP/IP port checking (tcp and udp)
  * SSL support for port checking
  * Unix domain socket checking
  * Process status and process timeout
  * Process cpu usage
  * Process memory usage
  * Process zombie check
  * Check the systems load average
  * Check a file or directory timestamp
  * Alert, stop or restart a process based on its characteristics
  * MD5 checksum for programs started and stopped by monit
  * Alert notification for program timeout, restart, checksum, stop
    resource and timestamp error
  * Flexible and customizable email alert messages
  * Protocol verification. HTTP, FTP, SMTP, POP, IMAP, NNTP, SSH, DWP,
    LDAPv2 and LDAPv3
  * An http interface with optional SSL support to make monit
    accessible from a webbrowser

pi@ph5b:~ $ sudo monit status
Monit 5.27.2 uptime: 20d 1h 22m

System 'ph5b'
  status                       OK
  monitoring status            Monitored
  monitoring mode              active
  on reboot                    start
  load average                 [0.15] [0.10] [0.07]
  cpu                          2.5%usr 1.5%sys 0.0%nice 0.2%iowait 0.0%hardirq 0.3%softirq 0.0%steal 0.0%guest 0.0%guestnice
  memory usage                 110.4 MB [25.7%]
  swap usage                   24.8 MB [24.8%]
  uptime                       46d 3h 30m
  boot time                    Sat, 23 Jul 2022 19:25:42
  filedescriptors              1176 [0.0% of 2147483647 limit]
  data collected               Wed, 07 Sep 2022 22:54:40

Below one for a Raspi can also be incorporated:

pi@ph5b:~ $ vcgencmd measure_temp
temp=50.8'C

Or if not a Raspi (thermal zones can differ):

pi@ph5b:~ $ cat /sys/class/thermal/thermal_zone0/temp
50844

EDIT: As an example, I configured monit to notify via mail if a particular domain "test.domain.lookup" is being queried in the Pi-hole logs.
Its in Dutch though but you'll get the general idea:

https://gathering.tweakers.net/forum/list_message/72475024#72475024

[RedStabilo]

In a different thread one of your colleagues said this;

Short version of reasoning for the change is there are a lot of people out there that expose their Pi-hole web interface to the internet - behind a reverse proxy or otherwise (Just search Shodan!) The less information that is on those dashboards the more secure things are.

I cannot understand the unwillingness to add this as optional feature for people wanting this.

[PromoFaux]

@RedStabilo

I cannot understand the unwillingness to add this as optional feature for people wanting this.

It's OK. There is no need to understand. We made a design change to the dashboard, and some people do not agree with that change. That's OK, too. In fact we expected that some people would not agree. That is the nature of change.

Here is my longer message from that same thread:

Pi-hole dashboard accessible without login

TL;DR - I wouldn't put any money on the changes being reverted We're OK with some people being unhappy with change. It happens. The product is provided as-is, and for free. If we have made a decision to go in a certain direction, it's because that's the direction we want to go in, if a few feathers are ruffled in the process, then that is just the cost of change. A handful of people needing to adjust their workflow (like... just login, the information is all there. Use a password manager and …

[RedStabilo]

That is your prerogative as developers. I don't agree with that attitude that we users should accept it because it's free software. Maybe i've donated to your project. You don't know that. And yeah, donations don't give any rights whatsoever.

Anyway, thanks for taking the time to respond. Don't get me wrong; Pi-Hole is my favourite peace of software and all you people have done a great job. But I have a hard time understanding these changes as they don't seem logic team.

Luckily i've only upgraded my secondary Pi, and am still enjoying the graphs. Maybe one day i'll come to my senses and upgrade, but untill then.... That is my prerogative

[DanSchaper]

And in yet another post he said this:

@PromoFaux

In my honest opinion (and this is my own, it might not reflect that of the other devs), the Temp/Load/Memory shouldn't even be on the dashboard in the first place - as they're not really relevant from a Pi-hole point of view. Pi-hole is a DNS blocker, not a system monitor

Although I'm not quite sure why we are taking quotes out of context to prove a moot point but here's more

@RedStabilo

I don't agree with that attitude that we users should accept it because it's free software.

It's not an attitude that users should accept it, it's an attitude of "It's Free Open Source Software written by people in their spare time, volunteering their efforts." Maybe it's cliche but the source code is out there, feel free to make the changes to fit your specific use case.

We learned long ago that trying to make Pi-hole everything for everyone all the time just burns out the team. So we adopted the approach that Promo stated in his post

@PromoFaux

It's OK. There is no need to understand. We made a design change to the dashboard, and some people do not agree with that change. That's OK, too. In fact we expected that some people would not agree. That is the nature of change.

I guess you could say that's our prerogative. https://youtu.be/5cDLZqe735k

[deHakkelaar]

@DanSchaper

All of that metric information is going to be removed from the web interface in it's entirety so take this as a nudge to change your workflow now before the feature is completely removed.

Just to clear up a misunderstanding, its only the temp, load and memory metrics that are going to disappear with the v6 release am I right?
Query statistics on the the dashboard (when logged in) is going to stay am I correct?

[yubiuser]

You are. We are not that crazy.

[PromoFaux]

Hold my beer....

[jfb]

@PromoFaux

In my honest opinion (and this is my own, it might not reflect that of the other devs), the Temp/Load/Memory shouldn't even be on the dashboard in the first place

I agree.

[Blackbrook]

As mentioned by @deHakkelaar I have had a look at monit and got it working quite quickly just based on their manual and examples:

It's a web server running on a different port on the same machine as pi-hole, displaying current system stats, ssh details and - and I did this just for testing purposes without any need - the first ten entries returned by top. This has been set up in less than 30 minutes.

Maybe this is a solution for all those using for the stats being gone?

All the best

[jBlues]

Changing an interface to a new style and having no option to get back to the old (better: well known) style was, is and will be not a wise plan. Why? There're many good reasons for that statement. Just picking one: Changing the interface imposes the - many - users changing their habits and force them to forget, what they've learned (some folks call this skills) and start from scratch again. Have you ever thought, how many hours were spent to make yourself comfortable w/ a software GUI? And if you've reached the point, that you finally know where certain commands are located, then somebody changes the whole thing again to something different and you've to start to learn all over again, would you really like to spend applause for that? I guess not. So, make a long story short: Please, give the PiHole user community the option to choose, which version of the GUI they want to use. Many thanx in advance.

[jfb]

@jBlues

Changing the interface imposes the - many - users changing their habits and force them to forget, what they've learned (some folks call this skills) and start from scratch again.

We have changed the interface many times over the past 5 years, along with many options available from the web admin GUI.

I can't think of a software package I use regularly that hasn't had an interface change.

[deHakkelaar]

With a little DuckDuckGo magic I was even able to monitor the GPU temperature on my Raspi.
And I adjusted settings a bit so I can login from a remote machine:

pi@ph5b:~ $ sudo nano /etc/monit/monitrc
[..]
set mailserver localhost
set alert my@mail.address not on { instance, action }

set httpd port 2812 and
   use address 0.0.0.0
   allow 0.0.0.0/0.0.0.0
   allow admin:monit

We need a script for this:

pi@ph5b:~ $ sudo mkdir /etc/monit/scripts
pi@ph5b:~ $

pi@ph5b:~ $ sudo nano /etc/monit/scripts/cputemp.sh
#!/bin/sh
CPUTEMP=$(vcgencmd measure_temp | sed "s/temp=//; s/\..'C//")
exit $CPUTEMP

pi@ph5b:~ $ sudo chmod +x /etc/monit/scripts/cputemp.sh
pi@ph5b:~ $

pi@ph5b:~ $ /etc/monit/scripts/cputemp.sh; echo $?
49

A nice separate config file for monit:

pi@ph5b:~ $ sudo nano /etc/monit/conf-available/cputemp
check program CPU-temp with path "/etc/monit/scripts/cputemp.sh"
    if status > 45 then alert

Symlink in the enabled folder:

pi@ph5b:~ $ sudo ln -s /etc/monit/conf-available/cputemp /etc/monit/conf-enabled/
pi@ph5b:~ $

Activate changes:

pi@ph5b:~ $ sudo monit reload
Reinitializing monit daemon

And soon after:

pi@ph5b:~ $ tail -F /var/log/monit.log
[..]
[2022-09-11T15:48:23+0200] error    : 'CPU-temp' status failed (49) -- no output

pi@ph5b:~ $ sudo monit status CPU-temp
Monit 5.27.2 uptime: 30m

Program 'CPU-temp'
  status                       Status failed
  monitoring status            Monitored
  monitoring mode              active
  on reboot                    start
  last exit value              49
  last output                  -
  data collected               Sun, 11 Sep 2022 15:48:23

A mail in my inbox:

Subject : monit alert -- Status failed CPU-temp
Sender : monit@ph5b.home.dehakkelaar.nl
Recipient: my@mail.adress
Date : Today 15:48

Status failed Service CPU-temp

Date: Sun, 11 Sep 2022 15:44:33
Action: alert
Host: ph5b
Description: status failed (49) -- no output

Your faithful employee,
Monit

And webGUI status:

It keeps mailing me every 4 minutes or so tough
But I am able to disable monitoring temperatures via the webGUI.

And oc I changed the temperature threshold to a sensible 80'C.
I believe a Raspi starts throttling the CPU/GPU clock at about 90'C ... ish.
It took me little over 30 minutes ... ish

EDIT: Ow the mailing stopped ... pew.

[yubiuser]

I agree with your statement that users need to adapt to the new interface and acquire new "skills". However, it's quite uncommon to offer options to go back to old versions for every change in the UI which is made. At least I don't see an option to get back the old Win 3.1 style I loved to much