photog.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
A place for your photos and banter. Photog first is our motto Please refer to the site rules before posting.

Administered by:

Server stats:

244
active users

#singleuserinstance

0 posts0 participants0 posts today

Calling all FediFetcher users who are currently running FediFetcher as a GitHub Action:

Following recent changes by GitHub, and a poll, which showed that only a single respondent actually relies on GitHub Actions without any alternative way of running FediFetcher, GitHub Action support for FediFetcher has been deprecated.

If you are currently running FediFetcher as a GitHub Action, please refer to our new guide on migrating FediFetcher from GitHub Actions:

github.com/nanos/FediFetcher/w

You should have until 20th June to perform the migration, but if you don't migrate before then, FediFetcher will likely cease working for you shortly after.

If you have any questions, please free to open an issue on GitHub, or reply to this post.

And if you are that one person who doesn't have an alternative way of running FediFetcher: Please get in touch! I'd love to help you if I can.

GitHubMigrating From GitHub ActionsFediFetcher is a tool for Mastodon that automatically fetches missing replies and posts from other fediverse instances, and adds them to your own Mastodon instance. - nanos/FediFetcher
Continued thread

• Set ES_HOST=https://localhost in .env.production
• Set ES_CA_FILE=/path/to/selfsigned/cert

(b) I had to carefully set the maximum Java heap size, even though I thought I had plenty of RAM.

This last point confused me for sometime. I am running ES on the same VPS as my mastodon instance, which has 8GB RAM. So I I initial set -Xmx3g, but ES kept failing with exit code 137 (excessive memory consumption). I successively increased -Xmx to 4g, 5g, 6g, but with no luck.

After checking the logs, the heap size was not growing and the cap was being enforced properly.

Continued thread

For those interested, setting up #ElasticSearch was fairly straightforward. I mostly followed the official instructions at docs.joinmastodon.org/admin/el, with two minor modifications:

1. I installed OpenJDK 21 (latest version in the Ubuntu repository)
2. I installed ElasticSearch 8.

Two things tripped me up:

(a) ES v8 requires TLS by default, even when binding only to localhost. This means that:
• When using curl, you have to use --cacert to specify the path to the self-signed certificate generated by ES & prepend localhost with https://

docs.joinmastodon.orgConfiguring full-text search - Mastodon documentationSetting up Elasticsearch to search for statuses (authored, favourited, or mentioned), public indexable status, and accounts

❓Some questions for those of you on a #mastodon instance with full-text search enabled:

1. How often do you use it?

2. What do you use it for?

3. Does it offer something above being able to search for hashtags?

I don’t have it enabled at the moment, but I’m suffering from #FOMO. However, I’d have to beef up the VPS for my #singleuserinstance to enable full-text search and I’m not sure it’s worth it?

#mastoadmin

🚀 Boost for reach, please!

Service-Hinweis: Durch ein manuelles optimize table der Tabellen storage, contacts, apcontacts, post-content, item-uri, post-user, post-tag, post-searchindex, post-media, inbox-status, post-thread-user und contact-relation (das waren die speicherintensivsten) ist die Datenbank meiner Friendica-Instanz soeben auf 2/3 ihrer vorherigen Größe zusammengeschrumpft.

Service-Hinweis 2: Tabellennamen mit Bindestrich müssen in der MySQL-Konsole in Backticks stehen (z. B. `post-user`).

I’ve just released v7.1.0 of FediFetcher. This is an important release that will significantly improve resource usage and running time of FediFetcher, as well as help our users be good citizens of the fediverse.

I encourage you all to update to the latest version asap.

For full details including update instructions visit the release page:

github.com/nanos/FediFetcher/r

(Edited to point link to 7.1.1 - this is a bug fix, and you should update to 7.1.1 rather than 7.1.0 if you can.)

GitHubRelease v7.1.1 · nanos/FediFetcherThis release provides two new features which will make FediFetcher far more efficient and faster to run, and also benefit the Fediverse as a whole, as FediFetcher will be much friendlier to other ...

Question to the #FediFetcher and #MastoAdmin communities:

Should FediFetcher follow the robots.txt?

(Edit: I should probably rephrase this slightly: I'm mostly interested in whether you think FediFetcher should honour blanket bans through robots.txt.)

Arguments for are quite obvious imo. But there are a few arguments against as well: For starters I wouldn't consider FediFetcher a bot, as it doesn't really go out and crawl things, nor does it store or process anything.

What do you think?

I would really appreciate if you could leave a comment explaining your decision as well (either here, or on the github issue).

github.com/nanos/FediFetcher/i

GitHubFediFetcher doesn't respect robots.txt · Issue #84 · nanos/FediFetcherBy 33b5e5

I've noticed a lot of chatter about setting up Elasticsearch for Mastodon 4.2's new full text search over the last few days, including what hardware is required, how difficult is it, etc.

So I thought I’d write down my experience, including the hardware I'm running Elasticsearch on for my single user instance:

blog.thms.uk/2023/09/mastodon-

blog.thms.ukSetting up Elasticsearch for Mastodon 4.2.x - blog.thms.ukWith Mastodon 4.2.0 proper full text search has finally arrived. This post runs through the server requirements and setup of Elasticsearch for mastodon 4.2.x.

@jay Three things you can do to enliven your small-numbers instance (we do all of these):

1: Set up a "Librarian" account that follows others of general interest whether you in particular care about them or not. Our librarian follows a lot of news, art, and science, but use it with _particular_ abandon to fill out your Federated feed.

2: Check said Federated feed, have the Librarian follow people boosted and lively.

3: Use the relay functionality to relay hashtags via the hashtag relay server:

relay.fedi.buzz/tag/[TagYouCar]

Also check out generalist relays and add some. You might want to have a solid defederation list first to be safe, but we didn't have any real trouble.