friendica (DFRN) - Link to source

Mark my words. The Epstein/Techbro totalitarians are intentionally trying to thin out the population!

Princess Vimentin PhD | Cancer Biologist (@princess-vimentin.bsky.social)

Recall that the CDC lab that tracked outbreaks like cyclospora was hollowed out to a point of being non-functional. More than 17,000 people have been sickened with cyclospora. This is deliberate, part of Project 2025. 🧪 https://www.politico.
Princess Vimentin PhD | Cancer Biologist (@princess-vimentin.bsky.social) (Bluesky Social)

friendica (DFRN) - Link to source

Rogue Project's Friendica Online

After about 8 hours or more of work I have finally gotten the Instance back online. Unfortunately there was some data loss. Though all user accounts are still available and everything appears to be up and running smoothly now.

One huge hurdle I have had is that the developers of Friendica use Ubuntu by Canonical. I will not use anything from Canonical personally. I prefer the RHEL side of the spectrum. My workstations are Fedora which is RHEL upstream. My VPS for this Instance is Rocky Linux 9. The issue here is that Debian based distro's such as Ubuntu use AppArmor which is arguably less secure or at least less configurable than SELinux. So Friendica upstream has no concern for SELinux (to be fair they do warn that this is their position and only support Ubuntu so it is what it is) and how it effects a server using this heightened approach to security. It usually leaves me with headaches after every upgrade of Friendica. Not only because of their updates, but because of the updates Rocky Linux makes to SELinux Context causing me to have to mass re-label and configure files and directories. The changes are needed for security and not just designed to give me a headache ;-).

To any of my users that may have bumped into my Maintenance page the answer is no. No, I am not going to scrap Rogue Project's Friendica. Though it was tempting earlier today when I was looking at the hours of work I had to make sense of what was going on. The losses were minimal and I hope no one is to disappointed. At least this go around I did not lose the entire site and have to build over from scratch.


I have made some changes that I hope to be worth the effort. I have gone away from using worker.php from the cron and now use daemon.php via systemd. This should help with the SELinux issues if nothing else because the context issue will not be so wide spread. Hopefully this also helps with the memory leak issues of the worker.php. It will show just how much of a change good or bad it will have in the future. As usual time will tell.


I hope everyone has a great day! Take Care!

Unus Nemo


Tips & Donations


#SERVER UPDATE #INSTANCE RECOVERY #ONLINE AGAIN

#SERVER UPDATE #INSTANCE RECOVERY #ONLINE AGAIN
in reply to Plan-A̵̛͈̬̥̿͋̓͛̕

friendica (DFRN) - Link to source

Unus Nemo

@Plan-A̵̛͈̬̥̿͋̓͛̕

At least at this point I have the db corruption and its cause sorted out. I have fixed the big issue, and db bloat seems to have dwindles away as the db is now less than half its size in two days. There are still some SELinux context issues but I will sort them out soon. Then all will be smooth riding. I have learned a lot today as I usually do when the Instance goes down.

For right now it is just a matter of watching for SELinux alerts and seeing if they are valid or if I have to edit yet another context ...

@Plan-A̵̛͈̬̥̿͋̓͛̕
This entry was edited (yesterday, 8:14 AM)
⇧