jrollans.com is a Fediverse instance that uses the ActivityPub protocol. In other words, users at this host can communicate with people that use software like Mastodon, Pleroma, Friendica, etc. all around the world.

This server runs the snac software and there is no automatic sign-up process.

Site description
These are the voyag... uh, things I post about.
Admin email
jrollans@gmail.com
Admin account
@jrollans@jrollans.com

Search results for tag #selfhosting

AodeRelay boosted

[?]DeltaLima 🐧 » 🌐
@DeltaLima@social.la10cy.net

Check_MK 2.5.0 is a nice update. Feels waaay more snappier and the "facelift" looks nice :)

    [?]asmw [he/him|they/them] » 🌐
    @asmw@infosec.exchange

    I just sent my first mail from my new Yunohost VPS, and it did not get flagged as spam! :)

    Now to try some apps.

      🗳

      [?]weed stallman » 🌐
      @incentive@mastodon.circlewithadot.net

      [?]ay » 🌐
      @ay@polymaths.social

      Building a immich only box thinking of going debian and btrfs any other ideas or tips?

      It's a 8th gen nuc


      Tia

      #selfhosting #immich #server #debian

        AodeRelay boosted

        [?]Matt :nixos: :wayland: » 🌐
        @thelinuxcast@fosstodon.org

        [?]Paco Hope [He/Him] » 🌐
        @paco@infosec.exchange

        I built my pool controller on so that I could control it with . Worked great. I was so proud. It had been the least reliable bit of in the house. (Because it was the bit that I had the majority role building 😜)

        Then a combination power strike/UPS failure took out my running . That NEVER happens. So instead of being super convenient, I spend my weekend trying to rebuild the NAS.

        I lost one of the pair of NVMe boot drives and at least one of my SAS spinning rust drives is not working. “zpool import -F -n hope” runs for about 30 seconds and then I get a kernel panic and reboot. I am not a happy camper.

          AodeRelay boosted

          [?]Larvitz :fedora: » 🌐
          @Larvitz@burningboard.net

          New post: Monitoring our FreeBSD Mastodon instance.

          burningboard.net runs as a stack of Bastille jails on FreeBSD, so I watch it from a separate box with Prometheus, Grafana and Loki. The interesting part is the FreeBSD-shaped gaps: no native ZFS metrics, nothing for the Mastodon API or our S3 media bucket. A small textfile collector fills all of it.

          This is how I keep an eye on the very instance I'm writing this on:

          blog.hofstede.it/monitoring-a-

            [?]Paco Hope [He/Him] » 🌐
            @paco@infosec.exchange

            One of my (adult) sons is very technical. Chip off the old block. I'm troubleshooting a problem with my NAS, and he's helping. We identify one of the 2 boot NVME drives as a culprit. When it's in the system, it doesn't boot. Cool.

            So he says:

            "I'm going to mark this one with a sharpie until we're sure it's bad."

            "Ok"

            "Oh, this stupid sharpie doesn't work, I'm gonna go get one that works."

            "Good. And throw the broken one out."

            Son proceeds to throw the NVME drive in the trash.

            I figured it out when I asked "hey, where's that NVME drive you marked with a sharpie?"

              AodeRelay boosted

              [?]Rolle Laukkarinen » 🌐
              @rolle@mementomori.social

              What many people misunderstand about hosting your own content (like this social media instance) is thinking we somehow NEED a big audience or Big Tech involvement.

              I'm perfectly fine if the world faded away and it was just the thousand of us here. It's like the early days of the web when we had small forums, nobody missed Reddit back then. Federation is a big plus, not a requirement.

              It's the same with websites or IRC for me. I know people use Discord, but I still stick to IRC even if there are only about a hundred of us left. I know people use AI now and website visitors are dropping, but who cares? I still keep doing it for those who like to read.

              I don't need the whole world involved for this to feel worthwhile. It's mine, I own it, and I host it for as long as I breathe. After that, it won't matter to me anymore, but I hope other admins keep things running the way I did.

                [?]Geoff » 🌐
                @sternecker@infosec.exchange

                Velvet Ant's "Operation Highland" doesn't exploit a CVE for persistence, they replaced trusted binaries: trojanized pam_unix.so + backdoored ssh/sshd/scp that take a hardcoded password and log creds. No patch fixes a swapped binary; the defense is file integrity.
                On Debian:
                dpkg -V libpam-modules openssh-server openssh-client

                No output = your auth stack matches the package DB.
                Then confirm /etc/ld.so.preload is empty.

                Installing debsums enhances the file integrity checking!!

                Scripted it to check files they swap + flag any unowned .so in the PAM dir.

                github.com/sternecker/midil/bl

                My personal GIT Repo is not public.. that is on my ToDo list.

                  [?]jo the disgraced » 🌐
                  @jo@infosec.exchange

                  its only been a month since i put my new home server together, and a disk is already failing

                  good thing i have zfs?

                    AodeRelay boosted

                    [?]Matija Šuklje » 🌐
                    @hook@toot.si

                    Are you an operator?
                    Or are you interested in running your own server, but have questions and concerns?

                    There will be a full day XMPP OpMeet online meeting in autumn, where you are free to stay the whole day or jump in and out just for the sessions you are interested in.

                    Vote on which dates would work for you (vote open until 2026-06-20):
                    nuudel.digitalcourage.de/1IMTC

                    More info (heavily WIP):
                    wiki.xmpp.org/web/OpMeet/2026.

                      [?]Self-Hosted Feed » 🤖 🌐
                      @selfhosted_bot@fd.mrmave.work

                      ⚡ beam-cloud/beta9

                      Ultrafast serverless GPU inference, sandboxes, and background jobs

                      Runs serverless AI workloads at scale with GPU support, sandboxes and background jobs via a Python interface

                      ⭐ Stars: 1673
                      📅 Last Update: Jun 19, 2026

                      github.com/beam-cloud/beta9

                        [?]Brett » 🌐
                        @brett@bitflipp.social

                        It looks like by going to a self hosted single user instance I basically just lose access to trending posts.

                        I'm not seeing any other gaps and I'm not adding a bunch of relays to fill it in either so that's an acceptable loss but something to surface to others running self hosted mastodon.

                        Would be nice for the Ice Cubes News page to work but oh well.

                          [?]Dima » 🌐
                          @dima@dol.social

                          ⚠️ OVH Local Zone Zurich outage (~4h)

                          My VPS in OVH’s Zurich Local Zone (where this Mastodon instance lives) was completely unreachable for about 4 hours yesterday.

                          No status page updates, no incident notice, nothing from OVH (so typical).

                          From my side it looked like a clean Layer 3 issue - pings to the VPS returned Destination Host Unreachable from their ToR router inside the zone:
                          localzone-zrh-zur1-tor2-as4210011009.ch.eu (57.128.121.83)

                          Monitoring graph for the last 24h is attached.

                          Uptime monitoring dashboard for the last 24 hours. Large text shows “82.73% UPTIME”. A thin horizontal bar is mostly green with a distinct red segment in the middle indicating downtime. Below it, “0.82 APDEX (1.0)” with a row of small colored boxes showing hourly scores: mostly bright green at 1.0, but with a clear drop to 0.29, then several 0 values, and 0.62 (in red/pink) during the outage window around 16:00–20:00. Time labels on the x-axis go from 12h to 10h (next day). The red elements align with the OVH Zurich Local Zone outage.

                          Alt...Uptime monitoring dashboard for the last 24 hours. Large text shows “82.73% UPTIME”. A thin horizontal bar is mostly green with a distinct red segment in the middle indicating downtime. Below it, “0.82 APDEX (1.0)” with a row of small colored boxes showing hourly scores: mostly bright green at 1.0, but with a clear drop to 0.29, then several 0 values, and 0.62 (in red/pink) during the outage window around 16:00–20:00. Time labels on the x-axis go from 12h to 10h (next day). The red elements align with the OVH Zurich Local Zone outage.

                            AodeRelay boosted

                            [?]Yehor 🇺🇦 » 🌐
                            @yehor@mastodon.glitchy.social

                            Guess what? All my went offline today at 3:50 AM while I'm away from home. Is it a rule or something?

                            It happened right after "Cloud Gateway Ultra has updated to 5.1.19". I never had issues with updates previously. This sucks. Looks like I'm disabling automatic updates on my network devices.

                            There are two log messages in a screenshot from the UniFi console. The first says that Device Cloud Gateway Ultra has been updated to 5.1.19 at 3:19 AM, and the second says that Switch Lite 8 PoE went offline at 3:51 AM.

                            Alt...There are two log messages in a screenshot from the UniFi console. The first says that Device Cloud Gateway Ultra has been updated to 5.1.19 at 3:19 AM, and the second says that Switch Lite 8 PoE went offline at 3:51 AM.

                              [?]Owl Eyes » 🌐
                              @d1@autistics.life

                              @pojntfx Gee, all of a sudden it seems like a great idea to self-host a relay server

                                [?]ae [he/him] » 🌐
                                @ae@kpop.social

                                is there any kpop fans who self host their music or is it just me 😭

                                  [?]al » 🌐
                                  @alan@lighthouse.co.im

                                  Greetings from the actual mastodon web interface for once 😃 Joys of Sideliq taming

                                    AodeRelay boosted

                                    [?]Yehor 🇺🇦 » 🌐
                                    @yehor@mastodon.glitchy.social

                                    I made it! Object Storage cluster with load balancing and replication. Also configured a cache for the public S3 endpoint for as described here: docs.joinmastodon.org/admin/op

                                    Now we wait for Mastodon's cached service workers to accept new content security headers for 7 days, then MIGRATE MASTODON MEDIA storage from to my ! Sounds frightening to be honest.

                                    A screenshot of a console output for the "garage status" command. Displays two healthy nodes, each with 1.8 TB of capacity.

                                    Alt...A screenshot of a console output for the "garage status" command. Displays two healthy nodes, each with 1.8 TB of capacity.

                                      [?]Geoff » 🌐
                                      @sternecker@infosec.exchange

                                      Threw the script for possible memory injection detection in Linux into a GitHub repo named "midil"
                                      github.com/sternecker/midil.git

                                      #0-day

                                        AodeRelay boosted

                                        [?]Yehor 🇺🇦 » 🌐
                                        @yehor@mastodon.glitchy.social

                                        I took 2 SSD drives, put them into a in RAID1, and made it a data storage for object storage.

                                        So my Garage now is a single-node cluster with its downsides.

                                        And I just realised that I could use those two drives on two different machines to build a two-node cluster with redundancy, data repairs, and load balancing! That would be cooooooool!

                                        Good to realize this now, before I migrated my instance media storage there.

                                          AodeRelay boosted

                                          [?]Yehor 🇺🇦 » 🌐
                                          @yehor@mastodon.glitchy.social

                                          [?]Geoff » 🌐
                                          @sternecker@infosec.exchange

                                          I was chatting with a red teamer friend recently about Windows exploitation by process injection.
                                          My knowledge is probably very dated compared to what is current.
                                          I'd had a couple classes on the topic, but it wasn't an area that I show great skill in.
                                          But, the recent dialogue had me thinking about my entire personal infrastructure is Linux and I do not have commercial enterprise endpoint detections enabled on them.
                                          As I said, personal infrastructure.
                                          For Exploitation.. In Windows, I'd fire up vmmap/rammap from Sysinternals and look for processes with vulnerable memory.
                                          How do I do the same on Linux? It's all listed in /proc somewhere. Wow this lead me down a rabbit hole.

                                          The results of testing my local Linux daily driver are concerning!

                                          #0-day

                                            AodeRelay boosted

                                            [?]DeltaLima 🐧 » 🌐
                                            @DeltaLima@social.la10cy.net

                                            Mastodon 4.6.0 installed ✅

                                            Only thing not working so far, is the autocompletion for emojis. When typing `:check` it used to show a list of emojis with "check" in the name. This seems to be broken now :(

                                            Release notes confirm, that this is something this part of the software was touched.

                                              AodeRelay boosted

                                              [?]Tywèle [She/Her] » 🌐
                                              @Tywele@corteximplant.com

                                              For everyone not using a VPS that is connected via a VPN to your server to remotely access it. How do you secure your server if you have any services that are publicly available through open ports?

                                                [?]Larvitz :fedora: » 🌐
                                                @Larvitz@burningboard.net

                                                Our Mastodon instance "burningboard.net" is now running on the newly released FreeBSD 15.1-RELEASE :freebsd:

                                                Everything looking fine. No unpleasant surprises. Upgrade process went completely smooth!

                                                Our Stack now:

                                                OS: FreeBSD 15.1-RELASE with Bastille
                                                Mastodon: 4.6.0-rc.2
                                                DB: PostgreSQL 18.4 / Valkey 9.0.3
                                                Search: Opensearch 3.5.0

                                                @tux @Mathias @AlienJay

                                                  [?]Marius (windsheep) » 🌐
                                                  @windsheep@infosec.exchange

                                                  Hetzner price spikes.

                                                  Alternatives: Contabo, NetCup, Manitu, OVH to some degree

                                                  Outlook: Hetzner is a founder-owned entity, no private equity drivers (afaik, but I did not find one).

                                                  I don't expect a price hike before Christmas 2026, for cash-out effects due to migration time needed vs. temporary lock-in. I also don't expect an immediate price change to existing customers, because the old hardware isn't "AI ready" (too slow, i7, not enough GPU RAM).

                                                  The last price increases to existing customers were announced in a way, so that you could move over to competitors (Feb until May or smth).

                                                  In German law, you can dispute material contract changes. I doubt that you get lots of extra time, but certainly more than 30d. You will not need a lawyer for that. Just respond with: "such material changes require further consideration at my part. I am asking for a new term that allows me to xyz" (reasonable).

                                                  Not an option: real-self hosting for private contributors (power, network), Raspberry Pi on solar panel with DynDNS and Freifunk uplinks (also I have some ideas now)

                                                  What to really do as private / non-commercial contributors: consolidate, team up with others, share infrastructure with friends etc.

                                                  What not to do: panic, shutdown things as an existing customers, rant on the internet (like me :) ) or overthink this.

                                                    AodeRelay boosted

                                                    [?]BastilleBSD :freebsd: » 🌐
                                                    @BastilleBSD@fosstodon.org

                                                    RE: mastodon.bsd.cafe/@subnetspide

                                                    Bastille makes a great self-hosting platform! Look at this absolute list of self-hosted software ⬇️

                                                    nsd, unbound, acme, adguard, gitea, haproxy, homebox, mail, netbox, nextcloud, plex, rustdesk, samba, syncthing, tor, unifi, vaultwarden, and more on one box.

                                                    AodeRelay boosted

                                                    [?]subnetspider » 🌐
                                                    @subnetspider@mastodon.bsd.cafe

                                                    FreeBSD 15.1 is here, and my home server is already running it. :D

                                                    Screenshot of my home server running FreeBSD 15.1-RELEASE and all 22 jails currently running on it.

                                                    Alt...Screenshot of my home server running FreeBSD 15.1-RELEASE and all 22 jails currently running on it.

                                                      [?]Neil Brown [he/him/his] » 🌐
                                                      @neil@mastodon.neilzone.co.uk

                                                      New blog post:

                                                      "Speeding up static site generation with BSSG"

                                                      Some fantastic new features by @stefano to his excellent bash static site generator.

                                                      neilzone.co.uk/2026/06/speedin

                                                        [?]Wulfy—Speaker to the machines » 🌐
                                                        @n_dimension@infosec.exchange

                                                        Good news: Every time I check my GenSys dash, there are new, better features.

                                                        This evening I found this deterministically (No Ai) System Analytics panels. Useful

                                                        Bad News: Its only Tue and I've burned all the weekly Compute (resets Sat).
                                                        I swear it hasn't been this bad, so either turned down the screws again, of they use dynamic pricing to combat sneaky buggers like me who want all the comoute.

                                                          AodeRelay boosted

                                                          [?]Anthropy » 🌐
                                                          @anthropy@mastodon.derg.nz

                                                          Hetzner is doing another price increase.

                                                          this time it seems to only affect new orders, not existing ones .. but the price increase is pretty steep, like 2-3x or more

                                                          docs.hetzner.com/general/infra

                                                            [?]gadgetChecks.de » 🤖 🌐
                                                            @gadgetchecks@burningboard.net

                                                            [?]Geoff » 🌐
                                                            @sternecker@infosec.exchange

                                                            Happy Monday. Sharing an update. Historically, my process has been to patch monthly unless a Known Vulnerability became Exploitable. Various feeds and sites like CISA's Known Exploited Vulnerabilities (KEV) list have been helpful. In order to "shift left," I'm moving to checking daily CVE updates against my tech stack. This is "personal optimization" to reduce the time from CVE reporting, exploit reporting, and my remediation. I like having a stable environment and don't have personal Dev/Test/Prod environments for all things. I also run the majority of my apps as containers. This is a work in progress. I'll share the updates. Trying to remain in my

                                                              AodeRelay boosted

                                                              [?]Larvitz » 🌐
                                                              @Larvitz@mastodon.bsd.cafe

                                                              The webserver for my websites (blog.hofstede.it and others) is now runing on 15.1-RELEASE arm64 with PKGBase 🙂

                                                              Upgrade from 15.0 to 15.1was pleasantly boring:

                                                              docs.freebsd.org/en/books/hand

                                                              - Creating a BE for the new version
                                                              - Mounting the BE to /mnt/upgrade
                                                              - Executing :env ABI=FreeBSD:15:aarch64 pkg-static -c /mnt/upgrade upgrade -r FreeBSD-base"
                                                              - bectl activate -t 15.0-RELEASE
                                                              - Rebooting the system

                                                              Everything working fine, all lights green :freebsd_logo:

                                                                dansup boosted

                                                                [?]🏳️‍⚧️ Christin Löhner 🏳️‍🌈 » 🌐
                                                                @christin@lsbt.me

                                                                FediSuite - Fediverse Management Platform

                                                                Open-source platform for social media management and analytics

                                                                If you manage several Fediverse accounts, you're constantly juggling browser tabs, losing track of which input field belongs to which platform, and at some point you no longer know what you've already posted. brings everything together in one place.

                                                                Connect accounts from 19(+) platforms: , , , , , , , and more. The app detects your instance type automatically, loads the correct character limit and media rules straight from your instance, and sets up the composer accordingly. No manual configuration needed.

                                                                The analytics go way beyond plain follower counts: daily engagement charts, follower growth, your best posting times as a heatmap, hashtag performance, and a tips engine that evaluates your actual data and gives you concrete suggestions based on your own numbers.

                                                                Schedule posts down to the minute in your own time zone. Background workers handle publishing reliably, with resume handling for rate limits and atomic delivery.

                                                                FediSuite is free and under the GPL-3.0. Anyone can host their own FediSuite and get it added to the official list automatically.

                                                                If you find a bug, especially in the setup, feel free to report it. The project is being actively developed, and real-world bug reports are among the most valuable contributions right now. The CONTRIBUTING.md explains how it works.

                                                                The project lives on donations. Donations guarantee and make it possible for FediSuite to keep going and keep being developed. To support FediSuite, click the yellow button on the website.

                                                                More info: fedisuite.com