Jump to content

civic95man

Members
  • Posts

    224
  • Joined

  • Last visited

Posts posted by civic95man

  1. 1 minute ago, ne10g said:

    So with that in mind, I just pass the 2080ti to the Win10 VM and leave the 1650 for unRAID's sole use?  

     

    This may sound silly, but how would I use the Win10 VM, direct on the server? Just want the one machine for all.

    I don't know if you've even setup your server yet, let alone booted unraid yet.  I would suggest looking up spaceinvaderOne's videos.  He has videos on just about everything which are extremely helpful.

     

    You would probably want to start with getting the nvidia build (through CA).  Then you would want to isolate the 2080ti card that you want the VM to use (again, look up his videos) - this just tells unraid not to use this card/ ignore it.  Then, when you setup the VM (again, watch the video) you are given the option to use the 2080ti card.

     

    Now for general unraid use, with the nvidia build, you will be able to manage unraid via the 1650 card (assuming you are booting into the gui mode) and still use it for transcoding.  This will give you an option to manage the system if you don't have another computer available.  

     

    In my situation, i have a monitor with two inputs - one from my win10 VM and another from the igpu that unraid uses.  So I can manage unraid with one input and interact with windows on the other input.  I also pass through several USB ports to the vm and attach the keyboard and mouse to those for win10 use.  I run my unraid "headless" in a sense where I don't really use the monitor and have no keyboard for unraid use.

     

    Hope this gives you a little information to get started.

     

    Good luck!

    • Thanks 1
  2. 4 minutes ago, ne10g said:

    So I guess in a VM, there could be more options

    I was referring to using everything (plex, sonarr, radarr, etc) within docker so that your system (unraid) can better utilize the resources. 

     

    5 minutes ago, ne10g said:

    which I wont mention as I read the notes in the nvidia build thread

    I'm assuming you are referring to bypassing the 2 transcode limits, yeah, you are on your own, or get the p2000 card to handle unlimited transcodes.  It really depends on what you are trying to transcode and how many simultaneous streams you expect to encounter. If you don't anticipate any 4k stuff then gpu/cpu would be fine (2 via gpu, the rest with cpu).

    9 minutes ago, ne10g said:

    Or....could I just let the CPU (i9-10940X) do hardware transcoding in the plex docker? If it even can?

    you can but gpu is always better than cpu (unless you REALLY care about quality, but thats another debate)

     

    10 minutes ago, ne10g said:

    Sorry for all the queries, please bare with me!

    no problem, we were all there at one point.

    • Thanks 1
  3. 1 minute ago, ne10g said:

    Would there be any issues with using the 1650 for hardware transcoding in Plex if I was to run natively with docker instead of an Ubuntu VM?  I'm not precious about the VM, docker would be better if it would do the job.

    1650 only supports 2 transcodes, thats the only issue. and you would need to use the nVidia build of unraid (many of us do that).

     

    with a VM, you are setting aside X cpu cores and Y memory for that virtual computer; whereas with docker, you share the system's cpu and memory between unraid and the different containers (plex, sonarr, etc). so those processes can use as much or less as needed without wasting memory and resources. 

  4. 15 minutes ago, ne10g said:

    2) Ubuntu VM - for Plex, radarr, sonarr etc - I would like this to use the 1650 GPU - maybe display on monitor using separate port (not essential)? Use its own NIC, 1 of the 2 available.

    # 1 is the easy one.  For #2, why not run these natively in unraid with docker?  That way you aren't tying up resources for that VM.

  5. While it could be a serious hacking attempt, you mentioned you have netgear router, which is probably performing vulnerability scans as previously mentioned.  Check the router's admin page and see if you can disable "armor".  That should disable that "feature" of the router, if you don't want it.

    • Thanks 1
  6. You would probably need to 'copy' the source file to the destination using the --preserve=all option followed by deleting the source file (move [mv] doesn't allow to preserve timestamps).

     

    i.e.

    cp --preserve=all [source] [destination]

    rm [source]

     

    I'm not big on making scripts so perhaps someone else can help with that, or you can attempt it yourself.  Might want to not remove anything until you are sure it works, or create a dummy directory and files to test on.

  7. 1 hour ago, BodoTheWaterGuy said:

    Maybe.  I caught it chugging away early this morning (yea 5 am alarms...) and it was working on 'deep analysis'.  This isn't something I need PLEX to do.  So, I turned off that option.  When it was doing that, PLEX only had 300MB of memory tied up.

    Plex itself could be consuming only 300MB of memory; however, if it needs to create any temporary files while doing it's work, it might be doing that in your /tmp.  That resides only in RAM and is counted separately from what plex reports.

  8. You only have 4 GB of memory, have you considered adding more?  4 GB is okay if you want to *just* run unraid but if you start doing more (dockers, vm, etc) you should consider adding more memory.  

     

    You could try killing xserver to shutdown the GUI but it would probably be best to reboot into non gui to save resources.  Also, you could try disabling anything you aren't using, such as VMs and Docker (I don't know if you are using any dockers, don't see any reference to anything in the diagnostics) - such that they no longer appear in the top menu bar.

     

    Good luck

  9. You seem to have a lot going on inside your server... Several plex transcoded streams, boinc, jitsi meet, VM, etc.  I don't know much about jitsi meet, but it seems to be node/java based and *could* be a resource hog.  If plex is transcoding to /tmp, that will take up memory as well.  Boinc is also using a good amount of memory.  Have you considered adjusting the resource utilization within boinc to use less memory?  

     

    Those are my thoughts, don't know if it will be helpful to you

  10. 10 minutes ago, MarkRMonaco said:

    That's not necessarily a true statement. That particular generation of Ryzen runs fine on RAM speeds between 3000 and 3200 MHz (+/-). In fact, I've had Unraid running on that combination for over a year and its been fine.

    I'm just quoting what is in the facts... and any system will run fine on an overclock until it doesn't.  If your'e trying to troubleshoot an unstable system, its always best to remove any overclock and go from there.

    10 minutes ago, MarkRMonaco said:

    That particular generation of Ryzen runs fine on RAM speeds between 3000 and 3200 MHz (+/-)

    runs fine where? windows? or unraid?

     

    That linked faq for unraid has this table to identify the supported memory speeds based on configurations

    image.png.ae627a45f89e61b4e8e3a65346647716.png

  11. 2 hours ago, ZosoPage1963 said:

    so, since I believe I have 1st gen Ryzen

    I believe you have a second gen Ryzen, but then again AMD's naming convention confuses me.  Set the memory speed to 1866 in the bios and try running a few cycles of memtest with all of your RAM installed and see how that goes.  If it passes then that was probably the cause of all of your issues.

     

    2 hours ago, ZosoPage1963 said:

    Or, even better... dump Ryzen and go with Intel and swap motherboard and cpu out...

    Since you already have it, might as well stick with it and see if the above fixes your problems.

     

    7 hours ago, JorgeB said:

    Yeah, I like AMD and just recently bough my first AMD CPU since the Athlon 64X2, but wish AMD did the same as Intel for RAM speeds, i.e. 10th gen Core CPUs support up to 2933MHz, doesn't matter if you have one or all DIMMs slots used, Ryzen 3rd gen supports "up to" 3200Mhz, most users then don't know that max speed is based on number of DIMMs/ranks used.

    My last AMD system was an Athlon 64x2 as well.  And I would never have known about the Ryzen memory speed issues if it weren't for this forum.  I suspect windows just handles these problems better when the memory is overclocked. either that or people just figure windows is being windows when it crashes.

  12. 7 minutes ago, ZosoPage1963 said:

    Awesome, thank you for the help.  I thought that the speed of the RAM was reflected in the number after DDR4... in my case, it is 3200.  RAM has always been difficult for me, and I thought I had it worked out, but obviously, I haven't and need to read a bit more.  Thanks for the schooling.

    No prob. and that is the advertised speed that the memory itself can operate at - but it is dependent on if the CPU and motherboard support that speed as well.  In the case of Ryzen, there are a lot of factors that come into play.

  13. 12 minutes ago, ZosoPage1963 said:

    Did I miss something stupid or is it just an experience thing?

    Just looking through your diagnostics says everything about your configuration, which is nice to spot things like this.  Looking at system\meminfo.txt in the diagnostics, it lists details about your memory. Two of the sticks say this:

    Speed: 2133 MT/s
    Manufacturer: Unknown
    Serial Number: E0F78509
    Asset Tag: Not Specified
    Part Number: BLS16G4D32AESB.M16FE
    Rank: 2

    While two other say this:

    Speed: 2133 MT/s
    Manufacturer: Unknown
    Serial Number: 00000000
    Asset Tag: Not Specified
    Part Number: F4-3200C16-8GVKB
    Rank: 1

    Then referring to this for the proper speed

     

  14. looks like you're mixing single rank with dual between those two types of sticks.  Don't know if that is considered a big no-no.  But since you're running dual rank in [some] configuration with 4 sticks, try lowering the RAM speed to 1866, since 2133 might be considered an overclock.

  15. 7 minutes ago, trurl said:

    You might be able to delete old logs (syslog.1, syslog.2, ...) in /var/log. Doubt you can delete the current syslog.

    I had to do this a few weeks back when my logs filled up and didn't want to reboot.  If I remember right, I think I was able to delete the current syslog as well.

     

    9 minutes ago, trurl said:

    Why?

    Why? In my case, not entirely sure but it was related to the webgui server and some issues with it - probably because I left a browser window open on a computer for too long. I also had to nuke the web server and restart it.

     

    In hind sight, I probably should have just moved the logs off onto the array in case I needed to troubleshoot something.

     

    Anyway, I find this to be a good tip in case you don't want to or can't reboot. ymmv

  16. Need new diagnostics to know for sure.  Did the network cable get moved from one port to another? The interface assignments should be saved to /config/network-rules.cfg on your flash drive. So the interface assignments will remain static and shouldn't hop around from one (re)boot to the next.  Might need to verify that file.

     

    Without knowing more about your setup, you should just be able to connect your cable into the eth0 assigned port.

×
×
  • Create New...