-
-
Notifications
You must be signed in to change notification settings - Fork 26
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
No results anymore #95
Comments
Is that instance public? I have the same on my instance which is public, but I did not update... I think someone is spamming again. |
I hardly use it anymore. Its my default browse search engine but I wont lie it hasn't returned any results in like a month now. I went to update it to see if it would fix it, but idk. |
Could you block requests incoming for a while and check after? Do you keep |
Do you have any feedback? Also, do you use DuckDuckGo or Google? |
I use DuckDuckGo, but it fails with either google or duckduckgo, we dont keep an access log, the goal was no logs, total privacy. Could I have my instance delinked to see if less traffic comes to it? Starting to think all these "privacy frontends" are horrible ideas when you share a master instance list in the repository because they ALL can be abused to circumvent rate limiting. |
Disable instance fallback for a bit, or just 403 everything in NGINX and check after like a day of denying everything. |
I will be adding a new instance soon if that works out and doing some IP rotation so that will fix at least a little traffic :P |
Could we yank results from searx and whoogle instances also? Maybe adding a way to supply a custom API key for googles custom search engine aswell. And interopolating other sources like where it can yield results from bing, duckduckgo, and google all at once |
I am personally against yanking from other instances, because I think that'd be just stressing them and not really helping anyone... Are Google API keys free? Might be worth looking into... There is already a big issue open on adding other search engines, see #96. |
Can you remove the supernets instance and onion from the instances.json Seems like everyone is jumping ship on being included on this instances lol... |
Well, we all like instance fallback, and we all dislike no results. Did whatever I told you above change anything? |
If not, I have no clue what might be the issue, since I heard no complaints from other people and LibreY runs fine on 3 separate servers. |
#96 isnt completely finished just yet, but it seems promising, finally am able to get search results on my instance most of the time. If you want to help test it then feel free to checkout my branch. Make sure you set the To checkout my branch, in a LibreY repository:
then edit the config with the changes described above. Remember to switch back to main after this PR is completed. Let me know if this improves the ability to fetch results or not. If you find any bugs or issues, please comment on #96 directly so I can fix them asap. |
Yeah I would still like my instance removed. I havent seen results on this shit in like 2 months now lol....please rm my instance. Funny how issues get closed on this repo without even being fixed.... |
Was pretty inactive and #96 was pretty much done and works like a charm on my instance and a few others. I almost always have results. I think this is not a LibreY-related problem to be honest. I will remove your instance, however. Also: have you tried my earlier mentioned tests? |
Make sure you've updated to the latest version of LibreY, and change the If this doesn't work for you, could you check if Google/Duckduckgo/other engines are blocking all of your requests from your server's IP? You can do this using links/lynx or a different web browser (curl output is pretty hard to read sometimes). You should see a page asking you to complete a captcha or something similar. If the engines are always blocking your requests then there might not be much that can be done.
Fair enough, if its not listed then hopefully you might not get as much traffic, so this might solve the problem. May I know which instance is your's so I can remove it from the list? @Ahwxorg I think we should keep this thread open until the issue is confirmed as resolved, or unitl a workaround/solution is found. Also |
I already removed it in my pr so liv created merge conflicts. I will update |
Oops, sorry... |
We could also take results from Yandex and Bing. The issue I often get is this: |
Bing seems a bit difficult to scrape, though it could be possible. We've already implemented Yandex results and it seems to work somewhat. I haven't seen it very often though, so it could be that their ratelimiting is more harsh than even google's. |
Just |
Most semi-updated distros have PHP 8+ packaged (even Debian!), and there could also simply be problems with the package itself. It doesn't make sense to investigate this issue because PHP 7 is now unsupported. |
I often observe this behaviour on my instance. It's more due to spam, while instance fallback or another engine solves the problem with a quick page reload. |
I'm starting to see this now on basically any libreY instance that I try now; somehow only my instance is working though that might be due to me disabling its API. I'll look into how to improve the fallback system over the next couple of days, if that might solve it, if not then there might be other ways to avoid being constantly ratelimited |
My API is also disabled but the VM that runs LibreY is now in a datacenter, so that could be an issue. |
Disabling the API was the magic bullet to make my instance work since it's the target of constant API abuse. |
I did a fresh update, set it up according to the wiki editing my config.php and opensearch.xml.
We are getting zero search results now. Not sure what happened. I can get image results but zero text results show using either google or duckduckgo.
The text was updated successfully, but these errors were encountered: