💾 Archived View for bbs.geminispace.org › s › misfin › 19088 captured on 2024-08-31 at 15:42:45. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2024-08-19)
-=-=-=-=-=-=-
Hello, I call on this roll call in the effort of making sure all systems are still operational in terms of the Hashnix misfin server.
Recently I had helped a user delete their account, though in doing so I noticed a strange glitch occured between misfin and the directories for gemboxes and identities. After resetting the server, my account's gembox had been wiped. I must post this now in the interest of all users to make sure that hasn't happened across the board.
Is everyone's account still fine? Are your gemboxes untouched or have they been modified? Please let me know, I'll try my best to do what I can to remedy anything that's happened if it has.
Thank you always for your cooperation and your patronage. Have a blessed day.
Aug 15 · 2 weeks ago
My account is fine, gembox is intact.
All is fine here.
@flipperzero What do you mean by "resetting the server" ? Having a link in the frontend to delete the account would be helpful, will create a GL issue for that.
😎 flipperzero [OP] · Aug 15 at 20:50:
@gemalaya They had no access to their account and wanted to retry. I went into the host directory and did some test messages to find via latest modified date their gembox using ls -lht. I made sure I didnt target mine by cross examining the timestamps of the gemboxes as well as making sure whenever there was an update that it corresponded to received messages in the gembox. I found their gembox and moved it to a folder outside the host directory as well as their old cert. However, I had failed to stop the server before doing so (which would be systemctl stop misfin) and by that point the server was producing a 2 error code on pages. I ran systemctl restart misfin, wherein I found my gembox wiped.
☕️ hellfire103 · Aug 15 at 21:19:
My gembox is still full!
😎 flipperzero [OP] · Aug 15 at 21:36:
Ok well enough people have responded with "all good" that I' d say this probably only affected me. YAY! Remember folks: make a crontab to archive your shit ;/
On that note -- I will be doing so, as mentioned, for the sake of everybody's gembox and their messages' preservation. Thank you all, continue to have a good one. -|--- (it's a cross idk)
@flipperzero Ok. Weird. Backups are a good idea. You were using the default backend (MH) on your account ? I don't see anything in the code that could trigger this behavior. The /mailbox/clear endpoint is disabled (not routed). Is it possible that you somehow also moved your mailbox directory somewhere else ? In that case the frontend would recreate a mailbox automatically and you'd land on an empty gembox. I'll check again tomorrow.
😎 flipperzero [OP] · Aug 15 at 23:05:
@gemalaya upon further inspection, and some troubleshooting, what you suggest is EXACTLY wtf happened lmfaooooo.
I've recovered my messages now, and although I saved a file of the journalctl logs I was finding related to misfin, there's no need (unless you'd still like to inspect) as I had gone ahead and retrieved the gembox back to its directory as well as my messages within. The user who I helped, their account is now cleared for them to reset. Thanks for your help, still and always, cipres!
@flipperzero lmao
flipperzero: "one second, i'll fix your account"
user: "very kind, thank you"
....
flipperzero: "should be good now ?"
user: "not really"
flipperzero: "holy sh**, i've lost all my messages !"
😎 flipperzero [OP] · Aug 15 at 23:55:
classic hashnix