bluepr0

Members
  • Posts

    214
  • Joined

  • Last visited

Posts posted by bluepr0

  1. Hello!

     

    I've been running flawlessly unRAID 6.7.2 since it came out. However since a couple of days the server is hanging completely and I'm unable to bring it back unless I force restart.

     

    Wake up this morning and it was looking like this

    1980333149_Screenshot2019-10-05at12_15_27.thumb.png.81d7bc104a720bfc1a09b68094e6967d.png

     

    As you can see on the web interface, some of the cores are blocked at 100% (however that doesn't seem to be the case if you get into terminal > ssh > htop.

     

    I was only running a VM, because this happened to me a few times already so I was trying to troubleshoot what's the problem. I'm also unable to stop or force shutdown the VM, I will get the message below

    1972239973_Screenshot2019-10-05at12_17_16.png.047400a9b3f65806e35f9137ae7b4adc.png

     

    And this is what the VM itself shows in the video output

    1736814575_Screenshot2019-10-05at12_16_06.thumb.png.fc24975d75d1d48f6ea55a0b57a67f6d.png

    By the time I try to force shutdown the VM I'm not even able to get into the web interface, but I still get SSH access.

     

    I have been reading the Help post here https://forums.unraid.net/topic/37579-need-help-read-me-first/ and followed the instructions, first I run diagnostics command to create the ZIP, however the powerdown command is not working. It won't shutdown 

     

    In the logs > libvirt.txt I'm seeing the same message I attached before in the screenshot. However I'm not sure how to troubleshoot or what's really the culprit of these hangs.

     

    Any help would be greatly appreciated so I can move forward and find a solution. Thanks!

     

    xserver-diagnostics-20191005-1253.zip

  2. Hello! thanks so much for putting this docker together. Looks like it's working like a charm... probably not renewing my Dropbox subscription next year.

     

    Anyway! I just wanted to ask if it could be possible that the nextcloud docker is avoiding my disk where the nextcloud share is to go to stand-by mode? Even if it's not syncing anything since yesterday, I'm still seeing my disk up. If yes, is there a way I can avoid this and let my disk to go to sleep? 

     

    Thanks!

  3. Also If I do 

    root@xserver:~# cd /boot/config/plugins/user.scripts/scripts/StartVMs
    -bash: cd: /boot/config/plugins/user.scripts/scripts/StartVMs: No such file or directory
    root@xserver:~# 

    I can get into scripts folder

    root@xserver:~# cd /boot/config/plugins/user.scripts/scripts/
    root@xserver:/boot/config/plugins/user.scripts/scripts# ls -la
    total 56
    drwxrwxrwx 7 root root 8192 Mar 20 12:59 ./
    drwxrwxrwx 3 root root 8192 Mar 19 21:19 ../
    drwxrwxrwx 2 root root 8192 Mar 20 12:59 StartVMs/
    drwxrwxrwx 2 root root 8192 Mar 20 12:59 VMDelay/
    drwxrwxrwx 2 root root 8192 Mar 19 21:14 delete.ds_store/
    drwxrwxrwx 2 root root 8192 Mar 19 21:14 delete_dangling_images/
    drwxrwxrwx 2 root root 8192 Mar 19 21:14 viewDockerLogSize/
    root@xserver:/boot/config/plugins/user.scripts/scripts# 

    But then I can't O_O

    root@xserver:/boot/config/plugins/user.scripts/scripts# cd StartVMs
    -bash: cd: StartVMs: No such file or directory
    root@xserver:/boot/config/plugins/user.scripts/scripts# 

    I can get inside VMDelay

    root@xserver:/boot/config/plugins/user.scripts/scripts# ls
    StartVMs/  VMDelay/  delete.ds_store/  delete_dangling_images/  viewDockerLogSize/
    root@xserver:/boot/config/plugins/user.scripts/scripts# cd VMDelay/
    root@xserver:/boot/config/plugins/user.scripts/scripts/VMDelay# ls -la
    total 32
    drwxrwxrwx 2 root root 8192 Mar 20 12:59 ./
    drwxrwxrwx 7 root root 8192 Mar 20 12:59 ../
    -rwxrwxrwx 1 root root    7 Mar 20 12:59 name*
    -rwxrwxrwx 1 root root   85 Mar 20 12:59 script*
    root@xserver:/boot/config/plugins/user.scripts/scripts/VMDelay# 

     

    I will try to delete and add the script again

  4. 3 minutes ago, jonathanm said:

    Weird. So if you type just the portion inside the double quotes, what happens?

    /boot/config/plugins/user.scripts/scripts/StartVMs/script

    This

    root@xserver:~# /boot/config/plugins/user.scripts/scripts/StartVMs/script
    -bash: /boot/config/plugins/user.scripts/scripts/StartVMs/script: No such file or directory
    root@xserver:~# 

    VERY WEIRD O_O

  5. Just now, jonathanm said:

    What happens if you enter just the second line of the short script at the console command prompt?

    echo "/boot/config/plugins/user.scripts/scripts/StartVMs/script" | at now

    Does it fire the StartVMs script file?

    Nope, I get this output but the VMs stay off

    root@xserver:~# echo "/boot/config/plugins/user.scripts/scripts/StartVMs/script" | at now
    warning: commands will be executed using /bin/sh
    job 8 at Wed Mar 20 15:43:00 2019
    root@xserver:~# 

     

  6. 40 minutes ago, jonathanm said:

    That output looks correct, did it kick off the StartVMs script?

    Nope, if I run the VMDelay script it runs fine but it doesn't fire the other script it seems, at least the VMs are not getting started. If I run manually StartVMs they do get started so I would guess it's not a problem of the script itself.

     

    I did copy and paste the script on Atom to modify and then paste it on the script file on unRAID

     

    However here's the text as well

     

    #!/bin/bash
    printf "%s" "waiting for pfSense ..."
    # Change IP to match an address controlled by pfSense.
    # I recommend pfSense internal gateway or some address guaranteed to be up when pfSense is finished loading.
    # I don't use external IP's because I want my internal network and appliances to be fully available
    # whether the internet is actually connected or not.
    while ! ping -c 1 -n -w 1 10.0.1.1 &> /dev/null
    do
        printf "%c" "."
    done
    printf "\n%s\n"  "pfSense is back online"
    virsh start Hassio
    # Insert optional delay to stagger VM starts
    sleep 1
    virsh start Windows10

    And the other one

    #!/bin/bash
    echo "/boot/config/plugins/user.scripts/scripts/StartVMs/script" | at now

    Thanks so much for your help ❤️

  7. Just tried it but still not working.

     

    I'm making sure that from the server I can ping the pfSense machine and it works fine. So this command "ping -c 1 -n -w 1 10.0.1.1" is working.

     

    I'm also seeing the first script (I called it VMDelay) on the logs

    Mar 20 13:16:03 xserver root: Delaying execution of fix common problems scan for 10 minutes
    Mar 20 13:16:03 xserver emhttpd: /usr/local/emhttp/plugins/user.scripts/startBackground.php "/tmp/user.scripts/tmpScripts/VMDelay/script" > /dev/null 2>&1
    Mar 20 13:16:03 xserver emhttpd: Starting services...
    Mar 20 13:16:03 xserver emhttpd: shcmd (55): /etc/rc.d/rc.samba restart

    But after pfSense is back online the other script is not triggering. However if I run it manually from the User Script panel by hitting the "Run Script" button then it turns on my other VMs.

     

    I'm probably doing something wrong but not sure how to debug the mistake 🤔

    Screenshot 2019-03-20 at 13.19.02.png

    Screenshot 2019-03-20 at 13.23.50.png

  8. 58 minutes ago, jonathanm said:

    Nope. Your problem is it's not just one script. It's 2 scripts. The top script is set to "At startup of array", the bottom script isn't set to auto start, it's called by the first script. There is a reason I posted it like I did. Name the first one whatever you want, and set it to auto start. The second script MUST be named StartVMs, exactly, and not auto started.

    oh damn, I thought it was just one. Sorry! will try again. 

     

    Thanks!

  9. On 3/5/2019 at 2:23 PM, jonathanm said:

    Set to "At startup of array" in user scripts

    
    #!/bin/bash
    echo "/boot/config/plugins/user.scripts/scripts/StartVMs/script" | at now

    Named StartVMs in user scripts

    
    #!/bin/bash
    printf "%s" "waiting for pfSense ..."
    # Change IP to match an address controlled by pfSense.
    # I recommend pfSense internal gateway or some address guaranteed to be up when pfSense is finished loading.
    # I don't use external IP's because I want my internal network and appliances to be fully available
    # whether the internet is actually connected or not.
    while ! ping -c 1 -n -w 1 192.168.1.1 &> /dev/null
    do
        printf "%c" "."
    done
    printf "\n%s\n"  "pfSense is back online"
    virsh start VMName1
    # Insert optional delay to stagger VM starts
    #sleep 30
    virsh start VMName2

     

    Hello! I've tried to setup today this but it seems the array never starts, it get stuck in "Mounting drives". If I remove the script then it starts correctly. I'm running on 6.7-rc5... maybe that's the problem? 🤔

  10. Hello!

     

    I've been taking a look at the unRAID wiki and doesn't explain too much about how to create a bonding, only how the different modes works. I've followed steps on Settings > Network Settings (check image attached) but I'm unsure if this is the correct setup. As you can see have I have "bond0" for eth0 and "bond1" for eth1.

     

    I'm unable to select eth0 and eth1 inside the option "Bonding members of bond0" for example (of course tried this with array stopped). My thinking was that you would have a "bond0" interface and inside of this you would get all the interfaces you wanted to be bonded, eth0 and eth1 in my case.

     

    Do I have this setup correctly? Or this is plain wrong?

     

    Thanks!

    Screenshot 2019-03-19 at 23.16.17.png

  11. 2 hours ago, jonathanm said:

    Nope, it's done with array started. The key is advanced view, and understanding that the delay is seconds to wait AFTER starting that specific docker. I actually want delay BEFORE starting some dockers, so I compensate by putting a docker that doesn't need network to autostart at the top of the list, and set a significant delay after that one.

     

    Start order is literally top to bottom of the items with autostart set to yes, so drag the dockers into the order you need. I had some issues with delay times not saving, so be sure to go back to the page and check your work to make sure everything is as you expected. When you add a docker it will likely screw up your order and delays, so be sure to verify periodically.

    Gotcha, thanks!

  12. 44 minutes ago, jonathanm said:

    Set to "At startup of array" in user scripts

    
    #!/bin/bash
    echo "/boot/config/plugins/user.scripts/scripts/StartVMs/script" | at now

    Named StartVMs in user scripts

    
    #!/bin/bash
    printf "%s" "waiting for pfSense ..."
    # Change IP to match an address controlled by pfSense.
    # I recommend pfSense internal gateway or some address guaranteed to be up when pfSense is finished loading.
    # I don't use external IP's because I want my internal network and appliances to be fully available
    # whether the internet is actually connected or not.
    while ! ping -c 1 -n -w 1 192.168.1.1 &> /dev/null
    do
        printf "%c" "."
    done
    printf "\n%s\n"  "pfSense is back online"
    virsh start VMName1
    # Insert optional delay to stagger VM starts
    #sleep 30
    virsh start VMName2

     

    Awesome! will give it a try. Thanks a lot!

     

    How do you handle docker boot and delay? I've checked docker preferences but couldn't find anything related. Will try to find something on the unRaid wiki 🤔

  13. 12 hours ago, BRiT said:

    I think the built-in/plugin functionality allows to control start order of dockers or vms separately, but no order dependency between those 2 groups.

     

    Anything more and you'd have to set dockers not to auto-start and use a custom user script to test for vm running and if so then start your list of dockers.

    Thanks for your reply. I've totally missed that, not sure where is it but will try to find it!

     

    9 hours ago, jonathanm said:

    I have the same scenario, here's how I handle it.

    pfsense VM set to autostart

    user scripts at array start kicks off pfsense monitor script

    pfsense monitor script waits until successful ping of internal gateway, signifying pfsense is online, then runs additional VMs using virsh start vmname

    unraid docker delays stagger starts with a docker that doesn't need internet with a 2 minute delay, which is normally plenty of time for pfsense to get rolling. Other dockers are set for many seconds delay, in a logical order.

     

    If I couldn't rely on a quick connect to the internet, I'd add the dockers to my pfsense monitor script and remove them from auto start. As it is right now, I've had no issues with things firing off correctly.

     

    Then again, I hardly ever restart my server, it's generally up for several months at a time without restart.

    Thanks! That sounds exactly like I want to do. Do you have the scripts handy? I might be able to implement them :)

  14. Hello guys,

     

    I've setup a few VMs and dockers that need internet to run correctly, however my router is actually another VM with pfSense. When I reboot unRaid I get a mess of errors because internet is not ready yet.

     

    What I wanted to do is to start the VMs and dockers only when my pfSense VM is ready (or maybe just after X seconds after the array starts). Any ideas on how I can do this?

     

    Thanks!

  15. Hello guys!

    I've been using unRAID since 2016 and so far never was able to create a Time Machine share that actually works. I'm currently using Mojave but with previous macOS also didn't work. I've also followed every advice, trick and setup guide with no luck. Is there really a way to make it work? I'm 1 click away to install Xpenelogy haha

     

    Thanks a lot!

    • Upvote 1
  16. I did also had the go file configure so it was using the binary from a folder inside my array, so it looked like

    #!/bin/bash
    # Start the Management Utility
    /usr/local/sbin/emhttp -p 80&
    /mnt/user/Mining/xmrig -o pool.supportxmr.com:7777 -u -p blabla -k --donate-level 1&

    This also didn't seem to work