Abzstrak

Members
  • Posts

    110
  • Joined

  • Last visited

Posts posted by Abzstrak

  1. obligatory "I'm new to unraid..."

     

    how and where do i change files? currently I need to modify fstab and have it persistent across reboots.  I'm familiar with Linux and would prefer CLI over some plugin, just don't know where unraid keeps things.

     

    why I want to change it-

    -add a tmpfs mount for plex trancode.  the default shm mounts using defaults, I don't want that.

    -----i want to do it to RAM to prevent a drive from spinning and reduce writes to ssd's..... not for performance/speed reasons.

    -add a swap file, due to using tmpfs, insurance to prevent memory from depleting. This will be on a spinner, I'll set swappiness to 5 or 10 to prevent it from spinning up unless really needed.

     

    the swap plugin is crap and old, I don't care to try to fix it, it's easy normal changes to make to a normal system... this boot from usb, running from RAM thing is just new to me.

  2. 6 hours ago, saarg said:

    Do you use /mnt/cache, /mnt/diskX or /mnt/user for the appdata?

    I was originally using /mnt/user but now am using /mnt/diskX which makes it corrupt less often, lasts like 4 days this last time instead of 12 hours.

    I'm tempted to try using cache, as I do not currently use any, and see if it works better with appdata on there, but adding another variable isn't really good for troubleshooting.

     

    Can I boot off of a 6.6.7 usb and expect to be able to mount and use my docker (plex and syncthing are the only ones)?  Can I expect my data to work? Will it have to recalc parity?  I'm just not sure of the changes from 6.6.7 to 6.7.0 and what would be involved since I'm so new to unraid.  I suspect that other than resetting up unraid, the data would all be fine, yeah?

  3. I have just been using unraid a couple of weeks, but I've been running plex for a long time... never in my life have I had the database corrupt until coming to unraid, and now it's been 9 times in 3 weeks.  I knew nothing of versions and just downloaded the latest stable, which was 6.7.0.

     

    something is seriously not right with this fusefs or release or something. I see others complaining and many suggesting downgrading "back" to 6.6.7.  I started with 6.7.0, so how can I try to go to the older version (and keep my plex docker and data)? 

     

    My other option to to scrap the plans to buy unraid as it just isn't working well at all... the serious hit in performance from this weird parity methodology is fine for this machines' purpose, but I can't have it corrupting files every 12-36 hours.  I can't even record OTA plex tv shows without expecting the db to become corrupt. 

     

    My hardware is an i3-8300, 8GB RAM.  6x 4TB SATA drives (dual parity), no caching.

  4. I'm sure this is a noob question, but I can't seem to find info on this that works... How do I setup a persistent Cron jobs that survives reboots?  Also, where can I put scripts since the normal /usr/local/bin also doesn't survive reboots?

  5. I figured it out, in case anyone has similar issues.

     

    Between my Cisco switch and pfsense, somehow dhcpcd was coming up before the port was up on the switch entirely... I found that if I left it for a while at the CLI login (like 7 minutes) it would eventually get the IP and start working... however, at that point, the check was already made for internet access (for the trial) so it wouldn't start the array.

     

    Simply fix, I set a static IP.  I usually don't use these in lieu of DHCP reservations, but it immediately solved the issue, everything works as expected.

  6. when I don't sign in, no IP.  It doesn't appear to make a DHCP request until after I login.  Obviously this means I cannot access the web console from elsewhere on the LAN unless I login from the gui mode. 

     

    I've switched to another virtual console from the login screen and confirmed.  I can run dhcpcd manually on the bond and get an IP, but that's not really getting me anywhere as it still require manual intervention. This also doesn't mount anything or start nginx.

     

    I must be missing a setting somewhere...

     

     

  7. I ave the trial working, but unless I boot the GUI mode, and sign in, it doesn't get and address from DHCP and doesn't mount the arrays or anything.  Is this by design for the trial?  Or is there a setting I'm missing somewhere?

     

    I want to run headless, but having to manually sign in on the console any time I reboot is holding me back from purchasing.

  8. I'm trying to migrate from OMV to unraid.  So far I've run into a few issues.

     

    The most annoying of which is that the usb only boots correctly in gui mode, and only mounts and does dhcp requests after a local login.

     

    If I log manually into nongui mode, I can manually run dhcpcd against bond0 and get the correct IP from my DHCP, however nothing is mounted, etc... If I boot into GUI mode, and leave it at the login screen and switch to another virtual console, however it still hasn't received an IP or "finished" the normal boot process unless I go back to the gui and login.  Once logged in, all is good.

     

    So, I have to boot to gui mode and I have to login, after that, everything mounts, dhcp gets the correct IP and apache/nginx or whatever starts up and I can connect over the LAN to the web console.

     

    So, how can I get this running headless the way I want?  or is this normal while on the trial?