Jump to content

xwiggen

Members
  • Posts

    136
  • Joined

  • Last visited

Posts posted by xwiggen

  1. 2 hours ago, Johanes said:

    as i said, the installation is native and fresh. also, i have removed all useless services, such as avahi*, unattended-upgrades, etc.  only ssh and ntp are running. so my question: where are the canonical and other undefined connection attempts coming from?? where can i find the responsible code in the system? 

     

    On 11/8/2021 at 7:29 AM, Heisath said:

     

    *.canonical.com are all Ubuntu Servers, so you are using an ubuntu system and it is doing some background stuff (Updates? Status?)

    littlericket.me seems to be some kind of Telegram bot. No idea if you are using that in any way https://botostore.com/c/messagestatisticsbot/ (see subscribe/unsubscribe url)

     

    Those are all NTP requests to a network pool, check the sizes (61b/15b), nothing wrong here move along

  2. On 10/31/2021 at 12:12 PM, destroyedlolo said:

    I'm suspecting a change/optimisation within GCC 10.

    I tried to use -rdynamic which is supposed to add all symbols, including not used ones ... but it doesn't change anything.

     

    I don't find a solution and it's blocking : I'm return back to Buster for my dashboard but I'm still investigating on other board.

     

    odd that works, I would link:

    cc -o .. ... -L<path-to-so> -lSelDRMCairo -l<other-sel-so>-lcairo .....

     

    as a shared object is not linked to any library but simply a position independent object file.

  3. On 9/18/2021 at 11:46 AM, ProstoUser said:

    Hi all .

     

    Before i buy Orange Pi 3 ( 64-bit 1.8GHZ ARM Cortex™-A53 , 2GB RAM ) , i want to  know:

     

     

    I wouldn't buy it for retail ~$80 in any case, because it's not much of an upgrade over an RPi 3.

     

    For that price you can get (and probably want) 4GB LPDDR4 instead of 2GB LPDDR3, go with either RK3399 as mentioned (M4V2) or RPi4.

  4. 4 hours ago, sockscap said:


    The latter web page provides the key as ".asc" files, but I cannot find how should I do with these ".asc" files. Could someone point out the official recommended method to fix such NO_PUBKEY errors? It's a bit strange that I have to consult ubuntu keyserver to fix Debian signing key issues.

     

    You can download and add the key with

    apt-key add <file.asc>

     

    The keyservers are capable of synchronization and I suspect (wouldn't know for sure) the ubuntu service is.

  5. On 8/16/2021 at 6:31 PM, ChrisO said:

    Fun? :-) Because it is so sophisticated? Built-in RAID/LVM? Snapshots?

    On all my Linux computers where there is space for more than one disk I use LVM on top of software RAID. Now I wanted to give ZFS a try as it seams to be really ready.

    And also because I spent several years of my life as admin of Solaris systems with zones on ZFS and it was fun.

     

    AFAIK ZFS for linux will never be ready unless Oracle resolves licensing issues. BTRFS has these features (except some RAID modes perhaps) and is stable.

  6. On 7/30/2021 at 11:52 PM, lanefu said:

    Lol i got a lot of feedback about https being an undesirable default.    Major reason being those on small connections heavilly rely on caching proxies to make life better.

     

    And yes persisting protocol based on request would be ideal.  How are you at python?

     

    https://github.com/armbian/dl-router/issues/2

    Same script, one for http mirrors (port 80) and one for https mirrors (port 443).

  7. 1 hour ago, Z06Frank said:

    I know this is an older topic but I have one question as I'm setting up some details to the HHD's on my OMV (5.5.20-1 Usul) based Helios64 NAS. I'm on the latest kernel 5.9.4 rockchip64. I have my primary OS on an M2 SATA drive and 3 8TB HHD's using Snapraid and UnionFS (2 in pool & 1 in parity).

     

    I have not set up HD-Idle yet....but plan to shortly. Question is why turn off write cache in the OMV "Disks" interface?

    PS...set APM=127 and spindown_time=240 on all three HHDs

    you want buffers to be flushed directly so the drive does not spin up whenever the kernel decides it needs to flush.

    IMHO I wouldn't recommend it (esp in RAID), consumer disks aren't as loud as 20 years ago and spinning up is the point where drives fail most. Early USB disks where prone to early catastrophic fails due to USB power mgmt.

  8. odd, both zeropi (http://ix.io/2DtK) and opipc+ (http://ix.io/2DEm) have ok temps (though 7z does not finish) and do not throttle below 80; buster is somewhat leaner than focal (~2C).

    have you run without X (acceleration?)                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                          ?

  9. 6 hours ago, Igor said:

    Fastest build server we will need to purchase - we still need a major CPU upgrade on cross compilation server. Something in range of Threadripper 3970x / 3990x + board + pci4 ssd in raid0 + 128GB memory which will costs us ∼6000 EUR.

    might consider Epyc with 8 channels, 16 dimm slots and better I/O, memory is "cheap"/upgradeable upto 1TB. 

     

    Best of luck!

×
×
  • Create New...

Important Information

Terms of Use - Privacy Policy - Guidelines