Ultimate UNRAID Dashboard (UUD)


Recommended Posts

29 minutes ago, falconexe said:

 

Did you change any of those query settings? For these to work correctly, the interval must be set to "2h" (2 Hours)  on Week/Month/Year in order for them all to sync up correctly. It all comes down to not grabbing too many data-points and finding the most near-real time sweet spot. If we set the interval too low then the longer time-frames like Year will not update often enough to stay in sync. If we set the interval too high then the dashboard will literally eat up all your RAM and your browser will crash because of the Year time-frame trying to load thousands of data-points. Using math, I figured out that the least common denominator for interval that provides the best performance, but also keeps these in sync as near-real time as possible, is 2 hours. The SIZE of the actual panel can affect how many datapoints are pulled as well. If you shrunk these down, you may be chopping off data-points (because Max data points = Width of panel > so they should all be the same size too) for the year panel. Finally, it is critical that your are using the correct path. "/mnt/user0"

 

Please match my exact queries below. Refresh and let me know if that improves your outcome. Mine has been rock solid since version 1.2 where I implemented this approach. Many others have been successful with these settings.

 

Week:

image.png.fdc581e23e734464e672aa4f9cfcfcf5.png

 

Month:

image.png.0a26cba816f7b6c0844e7264a6c9de1b.png

 

Year:

image.png.b1e38d5a52c5fe969a5b4b07cd89d760.png

 

 

Results:

image.thumb.png.dfc0fe41d92e50bfb4abd9cfbcf1ecad.png

 

 

If you want to read more about how I work out the mathematics of this solution, you can read about it in the original forum topic where UUD was initially developed here:

 

 

For the year one I am seeing interval of 6hr. It is grayed out and can not be changed it looks like. Seems like I can change the amount of data points, but then it deletes the "Width of panel" from the top box. When I did this in a test it did seem to fix it though. Can you foresee any adverse effects of me adjusting this? 

 

image.thumb.png.2e42d11fa06b6c92bbb98c262a9e685a.png

Link to comment
37 minutes ago, Shbeda said:

For the year one I am seeing interval of 6hr. It is grayed out and can not be changed it looks like. Seems like I can change the amount of data points, but then it deletes the "Width of panel" from the top box. When I did this in a test it did seem to fix it though. Can you foresee any adverse effects of me adjusting this? 

 

image.thumb.png.2e42d11fa06b6c92bbb98c262a9e685a.png

Sorry, I should have been more specific. I mean "Min interval" (fixed in the original post). You can't change the interval. It is auto calculated based on the Min interval and time range.

 

@ShbedaI do see that your Max data points is 1286, which is different than mine. So you must have moved/resized that panel. If you ensure the Week/Month/Year panel all have the same settings, and that they are all the exact same size on the dashboard, that should keep you synced up, now and in the future. GLAD TO HEAR IT IS NOW WORKING!

Edited by falconexe
Link to comment
58 minutes ago, falconexe said:

 

For fans, we are using REGEX to pull all fans. Just to the right of the orange symbols you will see "/fan./". This says pull all IPMI values where name is fan#. Also, in version 1.4 there is bug fix to make the labels not all say "Fan 01". If you want to fix that now, change the ALIAS BY field to "$tag_name". This says set those labels dynamically based on the GROUP BY tag(name) variable.

 

image.png.48849343eef91d78b5bef95627a79413.png

 

Finally, you should be able to omit NULL values (fans not on) by setting the display area on the right of the query. You want Calculation set to "Last (not null)".

 

image.png.5683093f0d76e12465d1b1ff60ed9679.png

 

Result:

image.thumb.png.9c716da19d969805b3da32d262150c64.png

 

 

@Shbeda Alternatively, you can just remove the regex, and manually select the exact fan you want. Then copy query A, and change out the fan in query B. Rinse and repeat... This will giver you tighter control.

This info was exactly what I was looking for. I removed the regex as it was still showing the 0 rpm ones as they did not come through as null. I ended up liking this better as you stated, it gives me more control! 

 

Next up, tackling the CPU info as I have 48 threads!

 

Thanks for all your hard work on this. I have long dreamed of having all this information at my fingertips and these fill out my new 34" 1440p ultrawide quite nicely! 

  • Thanks 1
Link to comment
27 minutes ago, Shbeda said:

Next up, tackling the CPU info as I have 48 threads!

Yeah you'll need to adjust your Regex. I added some Regex resources on Post #3 last night. OR...you can remove the regex and select threads manually.

 

27 minutes ago, Shbeda said:

Thanks for all your hard work on this. I have long dreamed of having all this information at my fingertips and these fill out my new 34" 1440p ultrawide quite nicely! 

Glad you are loving it and thanks for the positive feedback. I'm rocking a 43" 4K HDR monitor (Acer Predator CG437K-P) and IT IS GLORIOUS!

 

image.png.a9bc6333d20db421119e3043e5db2999.png

 

It's why my dang screenshots are so high res. 😂

Edited by falconexe
Link to comment

What are your thoughts on having the temp gauges stacked vertically on the sides? I like that you can get the Array Growth charts on a single screen now, and they kind of frame/bookend the central panels. Just an option for 1.4... I kind of like it. And it seems to scale nicely for widescreen monitors.

 

image.thumb.png.995176324ed584737a2eb2309e9c1ade.png

 

 

Link to comment
37 minutes ago, falconexe said:

What are your thoughts on having the temp gauges stacked vertically on the sides?

Me likey.   A better use of the space IMO.

 

Just installed the Grafana-UnRAID-Stack docker from testdasi but I am using UUD as the dashboard.  I have everything customized and working except my APC UPS.  No stats there at all.  I suppose I need to do something in telegraf.conf as I see no UPS input in the file that downloads with the aforementioned docker stack.

Link to comment
1 hour ago, Hoopster said:

Me likey.   A better use of the space IMO.

Thanks. I'll be keeping it that way. It is growing on me.

 

1 hour ago, Hoopster said:

Just installed the Grafana-UnRAID-Stack docker from testdasi but I am using UUD as the dashboard.  I have everything customized and working except my APC UPS.  No stats there at all.  I suppose I need to do something in telegraf.conf as I see no UPS input in the file that downloads with the aforementioned docker stack.

Yes. In post 1 it talks about dependencies. It is possible @testdasi does not have the prerequisite settings on the telegraph docker. You will need to add them manually. If they are missing from his GUS, then hopefully he will add them in the future... You most likely have the plugin commented out and disabled in the Telgraf Config. Simply uncomment it, and stop/start all 3 dockers.

 

image.thumb.png.9544ca10966e4b8c7623031a9a1a4051.png

Edited by falconexe
Link to comment
1 hour ago, Hoopster said:

Me likey.   A better use of the space IMO.

Thanks. I'll be keeping it that way. It is growing on me.

 

1 hour ago, Hoopster said:

Just installed the Grafana-UnRAID-Stack docker from testdasi but I am using UUD as the dashboard.  I have everything customized and working except my APC UPS.  No stats there at all.  I suppose I need to do something in telegraf.conf as I see no UPS input in the file that downloads with the aforementioned docker stack.

Yes. In post 1 it talks about dependencies. It is possible @testdasi does not have the prerequisite settings on the telegraph docker. You will need to add them manually. If they are missing from his GUS, then hopefully he will add them in the future...

 

 

Link to comment
1 hour ago, ChatNoir said:

Like Hoopster says, it seems more space efficient.

 

Will have to confirm once I try it.

 

Since I run on 6.9, do you foresee any issues ? (maybe on my secondary pool ?)

Thanks for the feedback.

 

I am not running the beta yet, but I don't foresee any issues. I'll just have to keep adding support for new features. The problem is, I don't run my server with ALL features, so it would be hard to test. Some others have asked for NVIDIA GPU monitoring. This is all possible, but hard to develop for if I'm not rocking that architecture.

 

Perhaps I need to get myself a donation bucket or Patreon so we can start a development fund 😉 All of this development (and support, which I'm happy to do) takes an enormous amount of time, not to mention my day jobs.

Edited by falconexe
Link to comment
1 hour ago, falconexe said:

Yeah. Look back like 10 posts and you'll find the answer.

 

Here 👇

 

Yeah, sorry 'bout that... I'd been refreshing looking for updates, but was stuck on page 5 and hadn't noticed that you'd moved on. My bad... :(

 

I edited the Week panel and got this screen grab:

image.png.39af4719c9a56724abe568935e29282b.png

 

Then edited the Month panel and got this screen grab:

image.png.5ca3f4b39b8fb305ca730de86f36a838.png

 

So I believe all is good. However, I noticed, though it didn't register immediately that something looked weird with the Week panel. I went back in to edit on it and after looking at it for a few seconds, it changed to this:

image.png.52ba42bf468d3913d626254264368cd0.png

 

It looks as though for the Week panel only, it's picking up the current width of the panel, and since when I'm editing it the panel draws itself full width, that's nearly the full width of my monitor.

 

Your thoughts on this?

 

BTW- v1.3 looks good, and your mock up for V1.4 is looking good, too! I've got to go back and reread the directions on mapping a drive number to a serial number, and I'll be golden!

Link to comment
1 hour ago, falconexe said:

Yes. In post 1 it talks about dependencies. It is possible @testdasi does not have the prerequisite settings on the telegraph docker. You will need to add them manually. If they are missing from his GUS, then hopefully he will add them in the future

So, I am assuming that since I get this error on all UPS stats in UUD (each section with an error on line 1 but different char #) that the issue is with the testdasi GUS docker, correct?

image.png.cfbd38c6b6d2a86cb29528f9626a0341.png

 

I have added the following to telegraf.conf as well:

 

# Read metrics about UPS usage
# # Monitor APC UPSes connected to apcupsd
[[inputs.apcupsd]]
#   # A list of running apcupsd server to connect to.
#   # If not provided will default to tcp://127.0.0.1:3551
servers = ["tcp://127.0.0.1:3551"]
#
#   ## Timeout for dialing server.
timeout = "5s"

 

Link to comment

I know that you've done a huge amount of work on this, and I appreciate it!

 

I'm curious, now that the display is using variables to define the Cache, Parity & Array drives, are these stored in the InfluxDB database somewhere, or do they need to be reset each time we install an update? I noticed that the currency, kWh, UPS capacity, and all the data sources had been reset when I updated from v1.2 to v1.3.

 

You may also want to consider making these entry boxes a tiny bit smaller... I don't think most country's currency symbols or energy rates take quite this much screen real estate:

image.thumb.png.98420be12829016c1a24ea642b1809ad.png

If you make them narrower, they should fit on the top line after the CPU Threads box:

image.thumb.png.319089fb412402df71238f70ff5adeaa.png

and then all the drive selectors can occupy the next however many lines are necessary.

 

Might make the top a little tidier looking.

Edited by FreeMan
Link to comment
3 hours ago, Hoopster said:

Just installed the Grafana-UnRAID-Stack docker from testdasi but I am using UUD as the dashboard.  I have everything customized and working except my APC UPS.  No stats there at all.  I suppose I need to do something in telegraf.conf as I see no UPS input in the file that downloads with the aforementioned docker stack.

 

1 hour ago, falconexe said:

Yes. In post 1 it talks about dependencies. It is possible @testdasi does not have the prerequisite settings on the telegraph docker. You will need to add them manually. If they are missing from his GUS, then hopefully he will add them in the future... You most likely have the plugin commented out and disabled in the Telgraf Config. Simply uncomment it, and stop/start all 3 dockers.

 

@Hoopster: the latest version of GUS has all the configs enabled, including inputs.apcupsd. However, it won't delete the old Telegraf config. You just need to delete the telegraf.conf file and update to latest version (and restart the docker if already updated) and it should work.

Link to comment

I am loving GUS + UUD!  Thanks to both @testdasi and @falconexe for your amazing work!

 

I am experiencing a couple of hurdles.  First, I cannot figure out the CPU regexes.  I have 2 CPU's each with 4 cores. (quad core Xeons).  So they all show up in the CPU01 graph, which I am fine with just removing the 2nd one and having them all in one graph, but I'd prefer two graphs.  If anyone can get the regex for that (or point me in the right direction) I would be very grateful.

 

Secondly, I am trying to get UPS data and I'm getting query errors.  I set my UPS db variable correctly, apcupsd is running on unraid and it can connect to my UPS.

 

{
  "request": {
    "method": "GET",
    "url": "api/datasources/proxy/9/query",
    "params": {
      "db": "apcupsd",
      "q": "SELECT last(\"load_percent\") AS \"UPS Load\", last(\"load_percent\") * .01 *  AS \"Watts\" FROM \"apcupsd\" WHERE time >= now() - 12h GROUP BY time(30s)",
      "epoch": "ms"
    },
    "data": null,
    "precision": "ms",
    "hideFromInspector": false
  },
  "response": {
    "error": "error parsing query: found AS, expected identifier, string, number, bool at line 1, char 74",
    "message": "error parsing query: found AS, expected identifier, string, number, bool at line 1, char 74"
  }
}

 

Link to comment
6 hours ago, FreeMan said:

It looks as though for the Week panel only, it's picking up the current width of the panel, and since when I'm editing it the panel draws itself full width, that's nearly the full width of my monitor.

 

Your thoughts on this?

 

BTW- v1.3 looks good, and your mock up for V1.4 is looking good, too! I've got to go back and reread the directions on mapping a drive number to a serial number, and I'll be golden!

 

If you click in the min interval field and then click out of it in the black space, it will usually update the the max data points. I've seen some odd behavior, and it seems to do with the Telgraf polling and possibly jitter settings in the Telegraf config. Anyway, nothing to worry about. Once you go back to the dashboard, it will all auto adjust behind the scenes.

 

If you want to map a serial number to a drive number, you just need to add an Display Name "override" to the field. It is the right side panel.

 

image.png.bf2482cc63f9676ee5b618b92c65d003.pngimage.png.df024fc56b8998f80cca16a4e4e22cca.png

Link to comment
6 hours ago, FreeMan said:

I know that you've done a huge amount of work on this, and I appreciate it!

 

I'm curious, now that the display is using variables to define the Cache, Parity & Array drives, are these stored in the InfluxDB database somewhere, or do they need to be reset each time we install an update? I noticed that the currency, kWh, UPS capacity, and all the data sources had been reset when I updated from v1.2 to v1.3.

 

You may also want to consider making these entry boxes a tiny bit smaller... I don't think most country's currency symbols or energy rates take quite this much screen real estate:

image.thumb.png.98420be12829016c1a24ea642b1809ad.png

If you make them narrower, they should fit on the top line after the CPU Threads box:

image.thumb.png.319089fb412402df71238f70ff5adeaa.png

and then all the drive selectors can occupy the next however many lines are necessary.

 

Might make the top a little tidier looking.

Where do you adjust the size of the text boxes ?

Link to comment
6 hours ago, FreeMan said:

I know that you've done a huge amount of work on this, and I appreciate it!

 

I'm curious, now that the display is using variables to define the Cache, Parity & Array drives, are these stored in the InfluxDB database somewhere, or do they need to be reset each time we install an update? I noticed that the currency, kWh, UPS capacity, and all the data sources had been reset when I updated from v1.2 to v1.3.

Variable values are stored within Grafana on a per dashboard basis. They will definitely need to be set again upon importing a new dashboard. Also, EVERYONE will want to SAVE the value of current variables, once set, as the dashboard default. You need to click the following check-boxes when saving the saving dashboard. These check boxes ONLY appear if a variable has been changed since the last time the dashboard was saved. If you do not save these variables, and if you refresh the website, or close the tab and open it again, the variable values will reset to whatever was last saved. This is a pretty big deal, so be sure to check those boxes once you have everything setup perfect!

 

image.png.fc61ebade72949d150e85e700c36e31d.png

6 hours ago, FreeMan said:

You may also want to consider making these entry boxes a tiny bit smaller... I don't think most country's currency symbols or energy rates take quite this much screen real estate:

 

If you make them narrower, they should fit on the top line after the CPU Threads box:

 

and then all the drive selectors can occupy the next however many lines are necessary.

 

Might make the top a little tidier looking.

 

Since these variables are Type = Text box, there is no option to set a width. Believe you me, I'd love to make everything nice and tidy. If Grafana ever provides this option, I'll be the first to set it! I've already moved the drive variables to the last once you set so they appear last in version 1.4.

 

 

image.thumb.png.301e2032a60232953d40dbd1c1400d02.png

Edited by falconexe
Link to comment
2 hours ago, shwa87 said:

I am loving GUS + UUD!  Thanks to both @testdasi and @falconexe for your amazing work!

 

I am experiencing a couple of hurdles.  First, I cannot figure out the CPU regexes.  I have 2 CPU's each with 4 cores. (quad core Xeons).  So they all show up in the CPU01 graph, which I am fine with just removing the 2nd one and having them all in one graph, but I'd prefer two graphs.  If anyone can get the regex for that (or point me in the right direction) I would be very grateful.

 

For Regex help, see Post #3. I put some great resources in there. Or you can beg GilbN to help you out. If I had the time, I'd do it for you, but EVERYONE should really start trying to learn REGEX. The online lab that I linked is a GREAT way to learn!

 

2 hours ago, shwa87 said:

Secondly, I am trying to get UPS data and I'm getting query errors.  I set my UPS db variable correctly, apcupsd is running on unraid and it can connect to my UPS.

 

Those UPS query errors, and everyone who has encountered them so far, ALL seems to be people missing the dependencies. Please verify that you have the Telgraf Config set correctly for [[inputs.apcupsd]]. See Post# 1

 

image.png.c62630b30f45f952cf949f566ee72ac7.png

 

Edited by falconexe
Link to comment

Alright. I think I answered EVERY SINGLE outstanding support issue thus far. Let us know if you still need any help. You'll probably see a bunch of new REDDIT users coming over for help. I am going to ask that they come to this topic for any and all support. Moving forward, this will be the single source of support. So get the word out. If you see anyone asking about UUD issues/support on various social medias (mostly Twitter and Reddit so far), send them here. Thx.

 

 

Link to comment

Looks like the UUD is spawning all kinds of personal branches. I'm seeing our panels in all manner of custom dashboards out there in the wild. Just like I learned from GilbN, others are learning from me. Pretty cool to see, and I really appreciate the guy actually linking back to this forum topic and throwing credit our way. I feel like I have given birth 😂

 

 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.