[–] [S] 3 points 1 week ago

Oh, ok. That's really useful, thanks. Right now I have a reverse proxy on my NAS, where all the services are running. So photos.mydomain.com, music.mydomain.com, search.mydomain.com etc are all included in AdGuard rewrite rule and all point to the NAS. So I can just add those to the extra_records and they will continue to resolve I guess.

  • source
  • parent
  • context
  •  

    Hi! I've been thinking of setting up a Headscale control plane on my VPS and adding my NAS as a node so that I can install the tailscale client on my phone and laptop and access my services from outside of my network. I have navidrome, Immich and a few more things.

    Right now I have AdGuard Home as my DNS server with DNS rewrites for my domain. *.mydomain.com points to my NAS and there I have NPMplus running that routes everything to the correct service/port.

    Do I still need that with headscale? I've read the docs but I still don't understand the "MagicDNS" stuff. Does that bypass my DNS server when the client is connected to the tailnet? Should I still leave it on for local traffic? And if I'm connected to the tailnet from outside, how do I propagate all the addresses?

    If anyone has a pointer or experience with a similar setup I'd really appreciate some help.

    [–] [S] 2 points 1 year ago (2 children)

    Thanks for all the replies! I'll try ebay probably. I kinda agree that posting buy/sell posts here might dilute the sparse content we have.

  • source
  •  

    Sorry if this goes against the community rules, I haven't included links or descriptions to anything just in case.

    I have a few smart switches that I "tasmotified" and an air purifier that I re-flashed for ESPHome. I'm moving overseas and I feel like selling them on craigslist or facebook would be a pain if they end up in the hands of someone that's not into home automation.

    I don't know if it would be ok to post them here or if someone knows of a more dedicated community where they would be a good fit. Would appreciate the help.

     

    I've got a bunch of old (1992) photos shot on 110 film with a point and shoot. I scanned the negatives with a cheap film scanner (from LIDL) a long time ago, and to be fair most of the 35mm film I had came out ok. But there are a couple of 110 rolls that I don't even know what to do with. The exposure is pretty bad and I think the film was exposed to light or not developed properly. I've played around with the photos trying to first get white balance right and then pushing the curves, but I can't get them right.

    Any advice? I don't have the negatives with me, otherwise I'd try scanning them again just in case. Is there any workflow or app that I can use to try to recover them?

    [–] [S] 2 points 1 year ago (1 child)

    Ok, so I have removed all the flags and all the mods. I also shut down the upstream velocity proxy, in case it was an issue with polling the server or something. Even running by itself the consoled gets completely spammed by messages. I know that these are not errors in themselves, but spamming the console just makes catching any real errors or relevant warnings that much more difficult. As a side issue, I'm using puffer panel to manage all the worlds and installs and the spam to the console makes the admin panel unresponsive. I have no idea where to trouble shoot next.

  • source
  • parent
  • context
  • [–] [S] 2 points 1 year ago*

    Yes, we already had that in place. The storage is in FreeNAS and we have snapshots every half an hour. We’ve already had to use the backups because after a paper update the whitelist turned itself off and when we logged in there were 3 random persons basically destroying everything. Btw, tangent, you find an open server and your first instinct is to just start tearing everything down? Why? Anyway, we just restored to the previous snapshot and it was back to normal. Zfs snapshots are great.

  • source
  • parent
  • context
  • [–] [S] 2 points 1 year ago (3 children)

    Oh, and in case it's relevant, I'm running the jvm with these flags (I just copied over the flags I had on my paper server): -XX:+UseG1GC -XX:+ParallelRefProcEnabled -XX:MaxGCPauseMillis=200 -XX:+UnlockExperimentalVMOptions -XX:+DisableExplicitGC -XX:+AlwaysPreTouch -XX:G1NewSizePercent=30 -XX:G1MaxNewSizePercent=40 -XX:G1HeapRegionSize=8M -XX:G1ReservePercent=20 -XX:G1HeapWastePercent=5 -XX:G1MixedGCCountTarget=4 -XX:InitiatingHeapOccupancyPercent=15 -XX:G1MixedGCLiveThresholdPercent=90 -XX:G1RSetUpdatingPauseTimePercent=5 -XX:SurvivorRatio=32 -XX:+PerfDisableSharedMem -XX:MaxTenuringThreshold=1 -Dusing.aikars.flags=https://mcflags.emc.gs/ -Daikars.new.flags=true

    This is a 1.21.4 server.

  • source
  • parent
  • context
  • [–] [S] 1 point 1 year ago

    Hey! thanks for the offer. I've pasted the output here: https://pastebin.com/X7Wfw9JU (this is just a fragment; the server has been on for a couple of hours and I could paste a Mb of text easily. It is very verbose)

    These are the mods installed I have installed:

    • fabric-api
    • instantgroup
    • voicechat
    • worldedit
    • tick-stasis
    • lithium
    • krypton
    • cloth-config
    • spark
    • ferritecore
    • viewdistancefix
    • FabricProxy-Lite
  • source
  • parent
  • context
  • submitted 1 year ago* (last edited 1 year ago) by to c/minecraft@lemmy.world
     

    I run a couple of minecraft worlds for my daughter and a few friends (yeah, ok, and for me :D ). Up to now we were running velocity and paper servers. I kept hearing that fabric servers were supposed to be more vanilla, and there were a couple of mods not ported to paper, so I wanted to try to replicate my servers with fabric.

    The server runs fine, but the output of the terminal is extremely verbose compared to paper, and I don't know if this is expected or if I've got something misconfigured.

    Most of the output is about "netty"

    "Netty Epoll Server IO #3" at io.netty.channel.epoll.Native.epollWait(Native.java:213)
    os_prio=0 cpu=6.33ms elapsed=1517.77s tid=0x00007068e00160d0 nid=1862005 runnable at io.netty.channel.epoll.Native.epollWait(Native Method)`___`
    

    for example.

    Does someone have any ideas? I wouldn't mind posting a large chunk of the output if someone thinks there might be an issue. Thanks!

    [–] [S] 4 points 2 years ago

    I think I got it working. I've got this in /etc/default/spamd:

    OPTIONS="-x -q -u spamd --max-children 5 -H /etc/mail/spamassassin/"
    

    and this in /etc/default/spamass-milter:

    OPTIONS="-e example.com -u spamass-milter -i 127.0.0.1"
    

    And it definitely reads user prefs from the database because I can change the 'required_hits' to 6.7 for a given user and I see this in the logs:

    spamd[523771]: spamd: clean message (-0.2/6.7) for user@example.com:133 in 1.2 seconds, 24427 bytes.
    

    I don't know how many different combinations of config flags I've tried, but these seem to do the job.

  • source
  •  

    Hi,

    I'm trying and failing to get spamassassin to load user prefs from a mysql database. I'm using spamass-milter and I can't find any way in the docs to send anything. spamass fails to parse the recipient as the user and just uses it's running user in it's call to spamd. The database is properly configured and I can connect and set settings from roundcube. The sql config is added to local.cf.

    I know that you can use spamd as a pipe, and then you can pass more variables, but I can't figure out the correct config for this setup.

    This is what I have in /etc/default/spamd: OPTIONS="-Q -x --max-children 5 -D sql,bayes -H /etc/mail/spamassassin/" I've also tried multiple combinations with the flags -q, without -x...

    And this is what I have in /etc/default/spamass-milter OPTIONS="-u spamass-milter -x -i 127.0.0.1" Where again I've tried without -u, with -e domain.com to explicitly set the domain.

    If anyone has any advice or can point me to a recent tutorial for Ubuntu 24.04, I would be really grateful!

    view more: next ›