RationalWiki talk:Technical support

Archives for this talk page:

Unable to view any pages that transclude Sirius’ signature (including tech support itself, hence this being posted on the talkpage)[edit]

Found I couldn’t load tech support, went on page information, clicked on the transclude templates in turn and found I couldn’t view @Sirius’ sig. Went on special:whatlinkshere/User:Sirius/sig, clicked some links, none worked. If a page contains the sig, it consistently refuses to load, producing a 503 error if I wait long enough. I’ve copied the source code below (I can view source, but can’t edit, I tried to edit tech support). nowiki’d for obvious reasons.

Sirius hasn’t edited her sig recently (history also works fine, last edit was on 17th February), and this is a new problem. Perhaps a problem with the images? They all take a while to load, longer than they should, with Pencil.png being the worst, loading all three at once could be too much. They’re all from wikimedia commons.

Christopher (talk) 15:40, 13 May 2021 (UTC)

That's weird, I'm not seeing it on my end. For reference, I'm using Firefox on a windows 10 computer. ☭Comrade GC☭^{Ministry of Praise} 15:50, 13 May 2021 (UTC)

[EC]Submitting that edit also gave me a 503 error (as did submitting this one the first time, which didn’t work), even if it worked in the end. The original problem has completely gone away, I can now view Sirius’ sig and transcluding pages fine. Other wikimedia common images were taking a bit too long to load, I didn’t properly test file types other than png. Should’ve mentioned this before, but I’m currently using safari on an iphone. Christopher (talk) 15:53, 13 May 2021 (UTC)

Has @Tim Starling been up to anything lately re: the server upgrade? Files not uploading (see tech support and the saloon bar), whatever this is, and other people are reporting general slowness. The test server is also very slow for me. Christopher (talk) 16:06, 13 May 2021 (UTC)

Checking now, I got the same error viewing some pages on and off. I also had a short-lived problem viewing one page after editing it earlier today. But it changes while writing and old examples now work other than the main technical support page. Most of the 50 first pages on the list with Sirius' sig work, not all. --ApooftGnegiol (talk) 16:30, 13 May 2021 (UTC)

Also, we do have a vague threat from Raven today here, where he claims, "Also, deleting this page and banning me will do nothing. I told you. I've hacked the website. Sweet sweet, revenge." In an earlier revision, he claims to have had help to do so, but nothing more about what he's supposedly done. --ApooftGnegiol (talk) 16:38, 13 May 2021 (UTC)

Don't think he'll actually do much. I can only imagine script kiddie shenanigans at worst for something so petty like that. --It's-a me,

🎄Lefty Green Mario!🎄 21:14, 13 May 2021 (UTC)

Some super-long pages are taking aaaaages to render, until the Varnish gives up (as it's supposed to). Some fancy sigs are taking ages - Fuzzy's for example, even though #choose shouldn't be expensive. Running "top" on the Apache server shows lots of memory free, but lots of CPU in wait state - i.e. it's like the disk is slow. All of this is frankly bizarre. Normal pages and editing work, but super-long pages with a ton of fancy template tricks seem to be giving the problems right now. (I tried simplifying Fuzzy's sig as a test, but it didn't help, so I self-reverted.)

I have no frickin idea what's up with this thing. But I'll be keeping a vague eye on it. As I sit here getting over my second Covid shot - David Gerard (talk) 21:11, 13 May 2021 (UTC)

I still have no idea what was wrong, but it's come good now - David Gerard (talk) 09:27, 14 May 2021 (UTC)

Link trolling[edit]

Sick of the alexa crap. Created new filter that prevents new people from adding "alexa.com" to anything. Let me know if you guys have any issues as a result of this. CorruptUser 21:56, 26 October 2021 (UTC)

"Up vote bomb" on WIGOCP[edit]

Pretty much every entry on the WIGOCP got 50 up votes out of the blue within a couple of hours. Not a big deal - very few people care about it nowadays anyway, and I'm not sure if techs can do anything about it, but I still felt like reporting it here since it could be either a bug or some sort of weird raid attempt that could happen again on the other WIGO pages. GeeJayK (talk) 04:45, 6 January 2022 (UTC)

Sidebar icons[edit]

For some reason I'm not seeing sidebar icons anymore. I just see a hyperlink saying "Icon war2.svg". Chef Moosolini’s Ristorante Italiano^{Make a Reservation} 18:44, 15 May 2022 (UTC)

It seems to be an issue affecting File:Icon war2.svg specifically (and maybe other files). The image doesn't display on the file page itself, although it does still exist. I tried to upload a new version but it seems that issue hasn't been resolved yet. Christopher (talk) 19:08, 15 May 2022 (UTC)

The image disk is full so the software can't create thumbnails - David Gerard (talk) 18:35, 17 May 2022 (UTC)

Can you run deleteArchivedFiles.php and cleanupUploadStash.php. You can also run findOrphanedFiles.php to see what files go unused. Note that the latter script may not neccessarily exist on our current install (upgrades!). deleteArchivedFiles.php removes any images we've since removed from the site and cleanUploadStash gets rid of any failed temporary uploads. -- Techpriest (talk) 19:31, 17 May 2022 (UTC)

I was cleaning up some of the oldest unused files on the wiki, and noticed that an old bot named Capturebot uploaded thousands and thousands of CP screen shots from 2007 to May 2009. Most (all?) of these files are unused and seem to rather useless to keep. The bot's uploads can be seen by doing a search for "wigo*.png". While we're in cleanup/maintenance mode, a good argument could be made for bot-removing all this cruft. —cosmikdebris ^talk stalk 19:47, 17 May 2022 (UTC)

I can fire the -unusedfiles generator with User:Inferno Bot but this would remove every file listed on Special:UnusedFiles; this is at the time of writing 19958 files. This will include personal images of editors that currently go unused. -- Techpriest (talk) 19:58, 17 May 2022 (UTC)

But if the image is on said editor's home page, it won't be removed, right? Bongolian (talk) 20:01, 17 May 2022 (UTC)

Correct; if the image is used on someone's personal page, it would be kept. However, images like File:Chef Moosolini.jpg would be removed. -- Techpriest (talk) 20:03, 17 May 2022 (UTC)

Personally, I think it would be a minor loss. Bongolian (talk) 20:13, 17 May 2022 (UTC)

Related statistics; -titleregex:"Wigo.*" -unusedfiles produces 3445 pages, while -titleregex:"Capture.*" -unusedfiles (antoher frequent uploader bot of unused files) produces 14800 files. I think combining these two would be a better use of my time. -- Techpriest (talk) 20:39, 17 May 2022 (UTC)

Paywall template[edit]

I'm not quite sure if this is the right place for this, but here goes. I know we have a template for labeling dead links — sites and pages that are no longer accessible at all for whatever reason. But do we have a template for links to pages behind paywalls (i.e. a fee of some kind is required to be allowed to read the page)? If we don't, we probably should. I would make one myself, but I'm not really sure how template documentation and other such minutiae work. --Luigifan18 (talk) 18:24, 19 December 2023 (UTC)

Look at what Wikipedia does. They've recently added something where you can mark citations as: free to access, free with registration (sometimes a limited free trial), and paywalled. Keep in mind that paywalled sources are admissible as citations, just nonpreferred assuming a citation of equal quality/authority/whatever is not paywalled. I don't know how hard it'd be to import the stuff Wikipedia uses to mark citations as paywalled, free, or some intermediary between those. Chillpilled (talk) 18:28, 19 December 2023 (UTC)

Yeah, I know, scientific journals and news outlets often have paywalls (which is part of the reason why I brought this up). I suppose they have to pay their bills somehow…

I wasn't aware that Wikipedia had a paywall alert system — I don't spend a whole lot of time there nowadays. That's good to be aware of. --Luigifan18 (talk) 18:37, 19 December 2023 (UTC)

See here. Chillpilled (talk) 19:11, 19 December 2023 (UTC)

For the paywalls of news outlets I use archive.today to get around them. Carthage (talk) 19:22, 19 December 2023 (UTC)

Wikipedia appears to do this in the Citation template, which we don't have here. There are parameters that tag a citation as free-to-access, subscription-required, etc.

RW has various older undocumented versions of the Cite template, none of which appear to support the tags needed for specifying the type of citation. I suppose some energetic soul with lots of time on their hands and expertise in such things could port the citation template here, but this doesn't look like a trivial task.

In the case of news media sources that have soft or hard paywalls (e.g., The Washington Post, The New York Times, Wall Street Journal), the Internet Archive or archive.is usually has a snapshot of the article in question. The academic journals hiding their content behind paywalls are less likely to be available on the archive sites. —cosmikdebris ^talk stalk 19:25, 19 December 2023 (UTC)

┌────────────────────────────────────────────────────────────────────────────────────────────────────┘ Template:Cws has as part of its formatting links to archive.today and the Internet Archive. I'll give an example of how to use this template below:

{{cws|https://www.pinkfloyd.com/|Pink Floyd Homepage|Pink Floyd Official Website}} Carthage (talk) 19:30, 19 December 2023 (UTC)

Just 'disable javascript'. Scream!! (talk) 20:07, 19 December 2023 (UTC)

I'm not sure how practical that is on a tablet browser… and it still sounds like a good idea to make a paywall notification template to notify users. I'm pretty sure we can just make it a variation of our existing dead link template. --Luigifan18 (talk) 20:14, 19 December 2023 (UTC)

Are you trying to make it a maintenance category? Or something that pops up in-text? Because either of these would be kind of obtrusive. If it's just to appear within citations, that's understandable. Chillpilled (talk) 20:48, 19 December 2023 (UTC)

Again, the idea is that the paywall template would be like the dead link template, which IIRC is usually placed within citations. --Luigifan18 (talk) 21:34, 19 December 2023 (UTC)

Apparently, my homepage doesn't think I exist.[edit]

Greetings,

I just signed up this afternoon and posted a Talk page suggestion to an article I read without incident.
I then went to my homepage and was given a canned message: "You've followed a link to a page that doesn't exist yet."
I proceeded to prove my existence by typing a rebuttal, stating: "I DO exist! See? I here I am typing. Do the non-existent type?"
But when I went to enter the rebuttal (and correctly identified the reCAPTCHA busses), I got a: "Warning....[ something to the effect that I was detrimental to society, but maybe I am misremembering the exact wording as I wasn't allowed to copy it ]."

How do I prove to my homepage that I exist? Bear in mind that I am a Dumb Bunny - should have made THAT my user name - and will not understand jargon.
Thank you for your help, Golgafrinchin Ark B (talk) 19:03, 14 December 2024 (UTC)

[PS: Is there a tutorial where I can learn how to modify my signature? I would like to try that.]
[PPS: I know to use the < br > command to create full line breaks, 1.0 size. However, I wonder if there's a command where I can create half-breaks, 0.5 size?]

By the way, just before I typed my rebuttal comment, I copy-pasted the "You've followed a link to a page that doesn't exist yet." message so people would know what my rebuttal was referring to. Was THAT my problem? Your new confused member, Golgafrinchin Ark B (talk) 19:10, 14 December 2024 (UTC)

It was a problem with the spam filter, won't say exactly what you did to trigger it but I have given you autopatrolled so it will no longer be an issue for you. Plutocow (talk) 19:36, 14 December 2024 (UTC)