Hey everyone, thought I’d post some stats since we’re one week old now!

From Vultr (instance is hosted through them):

Total applications: 116
Denied applications: 4 (one person asked to change username, 3 others gave one word answer to the application question)
Accepted applications: 112

docker stats (snapshot):

CONTAINER ID   NAME       CPU %     MEM USAGE / LIMIT     MEM %     NET I/O           BLOCK I/O         PIDS
7f365c848236   caddy      0.19%     43.22MiB / 969.4MiB   4.46%     7.23GB / 7.65GB   631MB / 146MB     8
d9421a5d930a   lemmy-ui   0.00%     49.62MiB / 969.4MiB   5.12%     1.51GB / 3.32GB   869MB / 1.26GB    11
e8850c310380   lemmy      0.08%     52.53MiB / 969.4MiB   5.42%     5.67GB / 5.86GB   942MB / 582MB     8
7ebb13fde277   postgres   0.02%     304.2MiB / 969.4MiB   31.38%    908MB / 2.97GB    3.82GB / 14.4GB   12
9b471baacf84   pictrs     0.05%     10.32MiB / 969.4MiB   1.06%     53.5MB / 1.18GB   653MB / 360MB     14

df -h:

Filesystem      Size  Used Avail Use% Mounted on
tmpfs            97M  1.7M   96M   2% /run
/dev/vda1        24G   12G   11G  53% /
tmpfs           485M     0  485M   0% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs            97M  4.0K   97M   1% /run/user/1002
  • bugsmith@lemmyrs.orgEnglish
    1·
    1 year ago

    Really interesting, thanks for sharing. Would you mind sharing a bit more information on the current setup and costs? e.g. which Vultr compute product is being used, any block storage added, etc.

    Something you might consider as the instance grows is Hetzner server auctions. These give you a dedicated server for a very good price and include unlimited bandwidth.

    I think pricing and product transparency would encourage users of this instance to contribute to the upkeep :)

    Edit: I didn’t see c/Rust:Meta existed before… I see you’re way ahead of me on cost transparency.

    • lemmyrs@lemmyrs.orgOPMEnglish
      2·
      1 year ago

      Just saw your Edit. Yeah, this was probably the first post I think. I wouldn’t be here myself if things weren’t as transparent as they could possibly be.

      As for Hetzner, yeah I’ve used them in the past (they’re great) but for initial setup, I felt that Vultr was probably just fine. That said, I’m definitely going to have to bump up the storage and RAM on this instance in the very near future, we’re cutting it closer than I’d like, specifically:

      df -h:

      Filesystem      Size  Used Avail Use% Mounted on
tmpfs            97M  1.7M   96M   2% /run
/dev/vda1        24G   15G  7.8G  65% /
tmpfs           485M     0  485M   0% /dev/shm
tmpfs           5.0M     0  5.0M   0% /run/lock
tmpfs            97M  4.0K   97M   1% /run/user/1002

      free -m

                     total        used        free      shared  buff/cache   available
Mem:             969         472          72         133         424         212
Swap:           2399         456        1943

      I think a 2 tier bump to $24/month 2 vCPUs, 4GB RAM, 5TB egress at Vultr might serve us for a decent amount of time. And it would probably still be cheaper than the cheapest Hetzner dedicated we could find. Maybe.