September 23, 20205 yr 16 minutes ago, MrLondon said: I seem to have gone backwards as now my panel is not reporting anything I am getting everywhere NO data and in the telegraf logs all I see is the following lines /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found /bin/sh: 1: apk: not found using the alpine container?
September 23, 20205 yr I accidentally changed it from alpine to latest, that seems to have caused this issue, but still get some n/a in the interface
September 23, 20205 yr Just now, GilbN said: Please share all your setup configurations for Telegraf and Grafana sorry the docker configuration or the conf files for both?
September 23, 20205 yr 3 minutes ago, MrLondon said: sorry the docker configuration or the conf files for both? docker and confs
September 23, 20205 yr Hey, this looks awesome. Is there a Tutorial on how to install this? i dont have unraid installed yet (waiting on my hardware) I woul like to Install this as soon i have my unraid up and running Greetings
September 23, 20205 yr Thanks again GilbN for the assistance, I found the conf file for telegraf but I am not sure where the conf file for grafana is stored as it is not under appdata/grafana telegraf.zip
September 23, 20205 yr 15 minutes ago, MrLondon said: Thanks again GilbN for the assistance, I found the conf file for telegraf but I am not sure where the conf file for grafana is stored as it is not under appdata/grafana telegraf.zip 55.14 kB · 0 downloads you have two /run/udev mounts, remove one. Not sure if that can create conflicts. And pastebin you telegraf.conf please.
September 23, 20205 yr Incase any body finds this helpful I always found telegraf config file full of so much that is not used. This is mine where it just has the inputs that I need, makes it easier to work on. Just note I don't have a ups so I don't use that setting. # Global Agent Configuration [agent] hostname = "box" flush_interval = "15s" interval = "15s" # Input Plugins [[inputs.cpu]] percpu = true totalcpu = true collect_cpu_time = false report_active = false [[inputs.disk]] ignore_fs = ["tmpfs", "devtmpfs", "devfs"] [[inputs.io]] [[inputs.mem]] [[inputs.net]] [[inputs.system]] [[inputs.swap]] [[inputs.netstat]] [[inputs.processes]] [[inputs.kernel]] [[inputs.diskio]] device_tags = ["ID_SERIAL"] skip_serial_number = false [[inputs.docker]] [[inputs.net]] [[inputs.sensors]] remove_numbers = true [[inputs.smart]] attributes = true # Output Plugin InfluxDB [[outputs.influxdb]] database = "telegraf" urls = [ "http://192.168.1.58:8086" ] username = "*******" password = "*******"
September 23, 20205 yr I found the reason for the no data as it was pointing to /mnt/user0 which is not present on my system, mine is all /mnt/user so changed that and now it shows the total sizes and growth.
September 23, 20205 yr 1 hour ago, MrLondon said: I found the reason for the no data as it was pointing to /mnt/user0 which is not present on my system, mine is all /mnt/user so changed that and now it shows the total sizes and growth. Do you not have any drives in the array?
September 23, 20205 yr 4 minutes ago, MrLondon said: I have 14 array drives and 2 parity drives. Does it say /mnt/xxx or /rootfs/mnt/xxx in Grafana for you. A bit wierd you dont have user0 in the disk.path column
September 23, 20205 yr I manually edited them all to /mnt/user as on my array I don't see that directory with 6.8.3
September 23, 20205 yr Just now, MrLondon said: I manually edited them all to /mnt/user as on my array I don't see that directory with 6.8.3 Ah, do you have any dedicated cache devices?
September 23, 20205 yr Yes I have a NVME cache drive, I have selected the drives at the top of the grafana for cache/array/parity.
September 23, 20205 yr 8 minutes ago, MrLondon said: Yes I have a NVME cache drive, I have selected the drives at the top of the grafana for cache/array/parity. I think its this line in your config. ignore_fs = ["tmpfs", "devtmpfs", "devfs", "iso9660", "overlay", "aufs", "squa shfs"] "squa shfs" is probably also ignoring fuse.shfs Edited September 23, 20205 yr by GilbN
September 23, 20205 yr 1 minute ago, MrLondon said: how should I change this or should I now just leave it alone? See if ignore_fs = ["tmpfs", "devtmpfs", "devfs", "iso9660", "overlay", "aufs"] adds the user0 path
September 23, 20205 yr I got this all set up and it is mostly working. I want to understand a little more about how this is all set up though so I can customize a little for my specific setup. For instance..... Fans: It is showing all fan headers on my MOBO. I Currently do not use Fan 4, Fan A, and Fan B. They are all showing on the dashboard, but show 0 RPM. Is there somewhere I can go do not have them pull data from these unused headers? Array Growth: Set this up yesterday and daily, weekly, monthly were all the same, but the annual is 50gb behind. The daily reset and midnight and has added data for today. Annual is still showing a 50gb difference. Thanks in advance for any guidance you can provide!
September 23, 20205 yr Author 42 minutes ago, Shbeda said: Fans: It is showing all fan headers on my MOBO. I Currently do not use Fan 4, Fan A, and Fan B. They are all showing on the dashboard, but show 0 RPM. Is there somewhere I can go do not have them pull data from these unused headers? 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. 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)". Result: @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. Edited September 23, 20205 yr by falconexe
September 23, 20205 yr Author 1 hour ago, Shbeda said: Array Growth: Set this up yesterday and daily, weekly, monthly were all the same, but the annual is 50gb behind. The daily reset and midnight and has added data for today. Annual is still showing a 50gb difference. Did you change any of those query settings? For these to work correctly, the Min 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: Month: Year: Results: 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: Edited September 23, 20205 yr by falconexe
September 23, 20205 yr Author 16 minutes ago, hogfixer said: Did the JSON get taken down? Says unavailable. Thanks Not that I'm aware of.... Let me try it. Be right back. Update: I just downloaded it fine. It is at the bottom of post 1. Try this direct link. https://forums.unraid.net/applications/core/interface/file/attachment.php?id=91063 Edited September 23, 20205 yr by falconexe
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.