Viaduct

Members
  • Posts

    76
  • Joined

  • Last visited

Everything posted by Viaduct

  1. Hi. Yes, I have the same mobo. I had terrible trouble with reliability with the Marvell chipset. Disks dropping off into thin air and even smart errors. Maybe it was my particular setup, and asrock did release a firmware update for the Marvell controller which I failed to apply. (Probably my lack of technical knowledge). But I lost confidence in the controller and got an lsi hba card to use more ports instead. Not looked back since. Currently on 9 disks attached. YMMV. However, love the Unraid ecosystem and the forum support here. Sent from my iPhone using Tapatalk
  2. I'm struggling to get telegraf working. Here is my docker run command: docker run -d --name="telegraf" --net="host" --privileged="true" -e INFLUXDB_URL="http://10.0.1.66:8086" -e HOSTNAME="tower" -e HOST_PROC="/rootfs/proc" -e HOST_SYS="/rootfs/sys" -e HOST_MOUNT_PREFIX="/rootfs" -e HOST_ETC="/rootfs/etc" -e TZ="Europe/London" -v "/mnt/cache/appdata/telegraf":"/etc/telegraf":ro -v "/mnt":"/mnt":ro -v "/var/run/docker.sock":"/var/run/docker.sock":ro -v "/mnt/cache/appdata/telegraf/telegraf.conf.tpl":"/etc/telegraf/telegraf.conf.tpl":ro -v "/proc":"/rootfs/proc":ro -v "/":"/rootfs":ro -v "/sys":"/rootfs/sys":ro -v "/etc":"/rootfs/etc":ro jjungnickel/telegraf Keeps exiting without doing anything. I have the standard telegraph.conf file in as telegraph.conf.tpl Is there anything special I need to add into the telegraph.conf? Any idea what's wrong? I can get the admin screen of influxdb OK, but nil gets into it.
  3. Thanks for docker. Works great. Any chance of hddtemp plugin? Would this work via docker? Thanks. Sent from my iPhone using Tapatalk
  4. Yikes. I've played about with netdata a bit, but only short periods. Have just run: Find . -name "*.aes" from the command line at /mnt and found none. Phew! I'll keep netdata off for now just in case. Sent from my iPhone using Tapatalk
  5. Netdata docker is amazing. Simple to install and monitor in webpage. Sent from my iPhone using Tapatalk
  6. Ah! Makes sense now. Yes smartctl measures my precleared hot spare on unassigned devices as 31. The value does not feature in the ui at all! Thanks for the info. Was getting confused by the readings. I guess my hot spare is the hottest drive in my case! Thanks again for very useful plug in. Sent from my iPhone using Tapatalk
  7. Last hour in the log is between 32 and 33 for HDD temp, but the readings pane has steadily deduced from 30 to 26 degrees. Incidentally, the temp polling time drop down doesn't do anything. I can't select it to change anything. Stuck at 3 min. It's just keeping the hd fan spinning higher than ideal when most drives except cache are spun down. Sent from my iPhone using Tapatalk
  8. Noted something weird. The fan control log shows a different hd temp to the readings tab and the footer value. E.g. Currently the readings tab has shown a HDD temp of 30 for the last 10 min or so, but the fan control log has varied between 32 and 33. Have spun up all drives too to make sure the data is refreshed. Any thoughts on the difference? Sent from my iPhone using Tapatalk
  9. Good stuff. I liked zukhof container because it had all in one, database and dash in one. With librenms would I also need a database container? It just gets more complicated? Did you get the hd temps to show in librenms? And the share free space? Thanks Sent from my iPhone using Tapatalk
  10. Thanks. Still works good but without the apparent new function of being able to add custom oid. I'll try it in a full Ubuntu vm I think before venturing with custom Dockers. Thanks for pointers though. Sent from my iPhone using Tapatalk
  11. Is it easy to update the docker container to the latest version of observium CE? Can one do a 'docker run' type script with the update from: http://www.observium.org/docs/updating/#community-edition I'd like the option to add custom oid which I think is in the latest one.
  12. That would be great. Sent from my iPhone using Tapatalk
  13. Nice! I hope those SATA breakouts aren't Marvell based... Sent from my iPhone using Tapatalk
  14. Using both turnkey and docker versions of observium, I'm not getting the hard drive temps and share free metrics to show. I guess it's not seeing the mib? Any tips? Or another dashboard to try? Working great otherwise, although maybe the CPU and mb temps are the wrong way round? Mb temp follows the core temps more than CPU temp! Sent from my iPhone using Tapatalk
  15. Doh. Finally found it at root rather than where I expected it in the script folder. All ok now. Is it worth setting the current directory to a set output directory since many scripts will send output to the current folder? I looked in /tmp/ initially but eventually found it in the root folder. I didn't think of looking there first. Silly me. Sent from my iPhone using Tapatalk
  16. Ran the diskspeed script in here, but the html output didn't appear. Am I doing something wrong? Sent from my iPhone using Tapatalk
  17. I'm interested if drivers for scsi controllers are in the Unraid kernel capable of reading tape drives. I have some old lto2 tapes and wondered if I could attach a lto2 drive either via sas or scsi and copy the content into a unraid share such as: dd if=/dev/st0 of=/mnt/user/share/tape Any ideas welcome! Sent from my iPhone using Tapatalk
  18. There are reports of a firmware update for Marvell controllers which may or may not help. Sent from my iPhone using Tapatalk
  19. Which Asus mobo? There was a recall on some Sandy bridge mobos due to dodgy sata controller. I was affected and had a swap out. Sent from my iPhone using Tapatalk
  20. I looked at the fan script again and if you set low to auto it is like setting it to 0°C. I have default values of 30 for low and 40 for high. But as the script is written those are only used if a value doesn't exist. But 0 is a value with isset so I'll have to fix that. So if you have low set to auto your linear scale is 0°C to 40°C or whatever your high is set to. And your fans will work off this scale. So that's why they are at 500 rpms. The fan speed minimum is for keeping your fans from stopping not for setting a point that they stop. The range is from 1 to 64 because that is what the ipmi raw range is. 0 = smartfan. If you wanted to keep your fans from stopping or keep them at a minimum speed you would increase the value from 1 till you get the rpm you desire. Thank you. Makes sense. Sent from my iPhone using Tapatalk
  21. Setting low to auto for me makes the fan spin up. I think it's because the CPU temp is high-ish (40), so when the hd is below the high threshold it operates in auto mode and tracks the CPU temp. Incidentally my fans do not switch off, they idle at 500rpm with a setting of 1. Is a setting of 0 possible for off? Sent from my iPhone using Tapatalk
  22. Which sata ports are you using on the mobo ? I have same case and mobo. I had loads of problems with the 4 port Marvell controller, the 9230 one, to such an extent I have lost faith in them and added an lsi card to use the remaining hd slots. I had similar issues with loss of disc and read errors. I only had unassigned discs on the controller at the time so didn't affect the array/parity for me. (The lsi card I got is just very slightly too big to get away with not losing the 8th slot BTW. Might need some modding!) There are reports that the Marvell firmware fix possibly sorts out some issues. It is available on the asrock site. I couldn't get the USB to boot in my case, but probably user error. I didn't persevere due to my personal loss of confidence in the Marvell controller. YMMV. Another thought is the power to the discs because it is distributed by the back plane to several discs, there might be a dodgy connection or capacitor if the discs are powered by the same connector. Those capacities stick out a bit and I'm always fearful of damaging one when I fettle in the case! Hope you get it fixed. V Sent from my iPhone using Tapatalk
  23. The 4 port one. The 2 port one works fine. I'm on latest firmware. I tried to re flash the Marvell but couldn't get the USB to boot. I didn't persevere because some on line reports still suggested it was a bit dodgy and to be honest I don't trust it anymore. Sent from my iPhone using Tapatalk
  24. I have lost trust in the second Marvell controller! Got lots of preclear errors with it and read errors. Currently have 7 spinning and 2 SSD. So have 6 on board and 3 off lsi. Working well so far but it has only been a couple of days. If I need another spinny such as when 6.2 comes with dual parity I'll need to rethink. Either shave a bit off the plastic insert or get a new case. Can't really shave the card since the sas socket is right there! Sent from my iPhone using Tapatalk
  25. Just added an lsi card and unfortunately it is just a few millimetres too wide to fit without having to take the plastic drive slide out. Effectively taking out a drive slot. Might have to get the dremel out to shave a notch in it! Sent from my iPhone using Tapatalk