Jump to content

BRiT

Members
  • Posts

    6,573
  • Joined

  • Days Won

    8

Posts posted by BRiT

  1. 15 hours ago, NLS said:

    BUT he is right. Seems that @jonp actually did make an update 16 days ago supposedly FOR 6.11.

    Maybe the plugin need to be re-evaluated? And why didn't the author of this plugin himself comment?

     

     

    Or maybe it was to ease some issues that come up over and over again because users never ever read or follow topics with update notes?

     

    I do not think this plugin should be revived. It has no active maintainer. The suggested replacement seems to be actively supported and maintained. The way forward is with that. Let this plugin rest in peace.

  2. 7 hours ago, exdox77 said:

    No OC settings have ever been enabled. The system has been running for years without issues with no changes to the BIOS. The memtest failed out because of too many errors. Running memtest on the new memory now. 

    Your screenshot shows XMP 3200MHz which is an overclock on the memory. You need to select the actual standard clocks and disable any and all XMP aspects.

  3. 55 minutes ago, frank-unr said:

    Thanks! That should work. I didn't know about this newperms command.

    There is also a Docker-Safe NewPerms command too, but I'm not sure where that's actually at and have only invoked it from the UI. I forget which CA-plugin provides it, if it's from the main "Community Applications" or one of the addons.

     

    As long as you don't run that newperms script against your "appdata" location or where you have dockers setup, you should be fine.

  4. 47 minutes ago, sonic6 said:

    This only applies to new installations. I didn't try it by my own, but someone from the german community runed into trouble when using docker containers with br0 and switching over to ipvlan.

    Not an issue since the users can change the setting if needed.

  5. 5 minutes ago, skyfox77 said:

    This was a CRITICAL bug and I almost lost 10Tb of data because of it, do not haste to add newer versions unless they are checked. This should really only happen on a BETA version.

    Edit: Well, we will see in 5 hours if I have lost anything or not, I am not out of the woods yet.

    Uhm details to share so others have an idea of what to look out for???

    • Like 1
  6. 13 hours ago, JoeUnraidUser said:

    However, there are things lacking.  Many tasks are not addressed by the server and need to be taken care of using utilities and scripts.

    Specific examples would be useful for future development efforts and prioritization. Not everyone is a mind reader to know what's needed or wanted.

  7. Article about power consumption:

     

    https://arstechnica.com/gadgets/2022/10/high-idle-power-use-on-intel-arc-gpus-requires-fiddling-with-bios-settings/

     

    Users will need to go into their PC's BIOS and configure a pair of advanced PCI Express power management settings—the "Native ASPM" (or active-state power management) setting should be enabled, and the "PCI Express root port ASPM" setting should be enabled and set to "L1 Substates." You'll also need to set the PCI Express Link State Power Management setting to "maximum power savings" in Windows' advanced power options settings.

     

    ...

     

    Testing from Tom's Hardware shows that with the settings enabled, Arc A750 power consumption at idle dropped from 37.3 W to 15.5 W, a significant drop. The same settings didn't seem to have an effect on an Arc A770 card, though it's unclear whether this is a motherboard bug, a GPU hardware or firmware or driver problem, or something else.

     

    Intel may be able to address the issue in the long-term with driver or firmware updates for the Arc A-series GPUs, but the troubleshooting article doesn't make it sound very likely. Intel says that the company "will be looking at making optimizations in future generations," which makes it sound like we'll need new hardware to address the issue decisively.

  8. 1 hour ago, tmchow said:


    just not as convenient since I have to then worry about mapping shares available to the docker, and that abstraction is inconvenient for various scripts I write. 
     

    I’d much rather have mediainfo available generally in unraid

     

    Ever thought to simply map full on /mnt/user/ to /mnt/user/ . That way the scripts shouldn't need changing and mounts shouldn't need adjusting.

     

     

     

×
×
  • Create New...