RationalWiki:Technical support/Archive9

This is an archive page, last updated 27 October 2013. Please do not make edits to this page.

Archives for this talk page:

<1>, <2>, <3>, <4>, <5>, <6>, <7>, <8>, <10>, <11>, <12>, <13>, <14>, <15>, <16>, <17>, <18>, <19>, < List>, <20>, <21>, <22>, <23>, <24>, <25>, <26>, <27>

, (new)(back)

Invalid Raw Signature

What makes this [[User:Man_of_Perspective|<span style=color:white>The Invisible Man</span]][[User_talk:Man_of_Perspective|<span style=color:silver>^{''I spoke to Him''}<span>]] an invalid raw signature. It says to check HTML tags but I can't see what the problem is.

P.S. this is what it is with wiki markup: [[User:Man_of_Perspective|<span style=color:white>The Invisible Man</span]][[User_talk:Man_of_Perspective|<span style=color:silver>^{''I spoke to Him''}<span>]] [nowiki'd because your span tag was still open]

L'homme de la Perspective ^Discusez? 09:31, 31 March 2013 (UTC)

Try <span style="color:white;"> and see if it's happier? - David Gerard (talk) 09:53, 31 March 2013 (UTC)

If this was copied verbatim, the obvious problem would be "</span". You're missing a ">". Wèàšèìòìď^{Methinks it is a Weasel} 13:12, 31 March 2013 (UTC)

Thanks for the help guys but I just can't get it to work, I think I'll stick with this sig then... L'homme de la Perspective ^Discusez? 22:23, 2 April 2013 (UTC)

MediaWiki requires valid HTML/CSS in raw signatures. Here's your code (white and silver are awful):

[[User:Man of Perspective|<span style="color:white">The Invisible Man</span>]][[User_talk:Man_of_Perspective|<span style="color:silver"><sup style="font-style:italic">I spoke to Him</sup></span>]]

-----> The Invisible Man^{I spoke to Him} <-----

00:44, 3 April 2013 (UTC)

Thank you so much NR! The Invisible Man^{I spoke to Him} 09:40, 3 April 2013 (UTC)

NP 12:39, 3 April 2013 (UTC)

Who was running WIGObot?

It appears to have been inactive for a month or two now - David Gerard (talk) 23:07, 2 April 2013 (UTC)

I think Pi wrote it but I asked for the code several times a while ago and got no responses. Who has the code? 00:46, 3 April 2013 (UTC)

Is it possible to make a template which auto-substituate ?

the french for quotes is «  », and I have made the template Template:fr" to make them easily, but I think it would be better If {{fr"|a}} was the same as {{subst:fr"|a}}, and I was wondering if it was doable. Jean 5 5 (talk) 07:07, 4 April 2013 (UTC)

A better option would be to add them to the editing toolbar or the box-of-characters below the editing field. Either globally, or by using a user script that adds them to your toolbar. I think that this was possible in previous versions of MediaWiki.--ZooGuard (talk) 17:02, 8 April 2013 (UTC)

Dealing with links to the Huffington Post in WIGOs

Could someone add an Abuse Filter rule that detects when someone tries to add a link to the HuffPo to WIGO:World and then displays a warning and asks for confirmation to save the page? Something like "Ceiling cat has noticed that you are using a link to a non-reliable website. Consider replacing it a link to the original source or a better news website, such as the BBC."--ZooGuard (talk) 17:05, 8 April 2013 (UTC)

Cute. 18:07, 8 April 2013 (UTC)

What makes HuffPo any less reliable than the BBC? I read it regularly (along with many other sites) and its news is fairly reliable. Sure its opinion pieces can be dubious, but (a) that's the case with most media outlets and (b) any opinion piece cited in WIGO should be cut or moved to Blogosphere anyway (as is the case with the "kittens" post on WIGO at the moment). VOX^HUMANA 21:48, 8 April 2013 (UTC)

Almost nothing in HuffPo is original; most of it is bloggy rewrites of content from elsewhere. (The exceptions are original bloggy posts.) Same thing as any Gawker site - no original content, there's an original source which is pretty much always better - David Gerard (talk) 23:15, 8 April 2013 (UTC)

Can't say I agree with that - for international news I'd agree (they tend to use AP), but most of their news on USA topics seems original. The article on how a rep from North Carolina was drafting a bill to make Christianity the state religion is a case in point. VOX^HUMANA 23:36, 8 April 2013 (UTC)

That's why I proposed asking for confirmation, not outright blocking the links. :)

HP is prone to sensationalism and click-baiting and I don't have much faith in their investigative/editorial skills.--ZooGuard (talk) 07:51, 9 April 2013 (UTC)

Capturebot

Any reason why capturebot does not seem to be working at the moment ? Maybe trent needs to replace the hamster. Ghost (talk) 04:16, 11 April 2013 (UTC)

502 errors

Not of any real concern to me - I'm mentioning it just in case it is of interest to anyone. Since the server switch yesterday I'm getting fairly regular 502 errors. I hit refresh and they go away. VOX^HUMANA 06:58, 11 April 2013 (UTC)

Holydaze's calculation of dates

How does Template:Holydaze determine when to show a message? Because I'm getting a template targeted for 12 April, and according to both my computer and RW's MediaWiki it's still the 11th. Unless the template thinks that the beginning of the day is when it starts in the UTC-12 timezone...--ZooGuard (talk) 15:43, 11 April 2013 (UTC)

If memory serves it starts at T-12 GMT (or UTC or whatever) & finishes at T+12. To allow NZ & Samoa(?) to have a full day. Scream!! (talk) 19:57, 11 April 2013 (UTC)

jQueryUI

Can we has it? 19:20, 14 April 2013 (UTC)

Probably. What's it need? Is there a handy page on mediawiki.org? - David Gerard (talk) 21:56, 14 April 2013 (UTC)

Just needs to be added as an external script call. You just add <script src="//ajax.googleapis.com/ajax/libs/jqueryui/1.10.2/jquery-ui.min.js"></script> next to wherever the main jQuery declaration is in the template. If you're not already using the Google API to pull the minified regular jQuery, you should. Is fast. 00:18, 15 April 2013 (UTC)

Apparent performance slowdown

I've got some actual data on performance from an editing standpoint. I've been making lots of small edits getting the section headers labelled for the Dover transcript. Below are the edit histories for two pages. Editing on each was from the same machine on the same network.

• This is the edit history from 03 April - as you can see, with the small section header edits I was average 7-8 edits per minute.
• This is from today - I'm now only making 2-3 edits per minute and I'm spending a lot of time waiting for the server to respond.

VOX^HUMANA 23:32, 17 April 2013 (UTC)

Just logged in. Bloody hell, load average is 30! Site's still ticking along, though. Something is actually hammering the site No, there's not many actual requests - but the requests there are are sopping up extravagant amounts of CPU. Not sure what's going on, will keep looking. At least we're totally fine for memory - David Gerard (talk) 08:39, 18 April 2013 (UTC)

It's calmed down now. My first guess is someone sending a pile of complex requests of the sort that send MediaWiki off to think for 30 seconds (obscure complicated diffs, that sort of thing). Not sure how to alleviate this one. I am pleased that the site mostly just keeps ticking along - David Gerard (talk) 12:21, 18 April 2013 (UTC)

It's obviously L33T TE4M HAXXOR getting their Guy Fawkes off. ^bomination 13:17, 18 April 2013 (UTC)

Capturebot

This discussion was moved here from RationalWiki:Saloon bar#Capturebot.

Any idea on when it will be working again ? Seems that a gentleman beginning with K has taken to oversighting all the links on the WIGO:CP talk Ghost (talk) 05:56, 14 April 2013 (UTC)

If you tell it to capture manually it complies, after a good while, but it doesn't seem to want to do it automatically for some reason. Peter ^mqzp 08:51, 14 April 2013 (UTC)

In the meantime, screenshot manually, preferably before posting, if you can tear yourself away from hanging on his every doing. He's really leading you a merry dance, isn't he? Sophie^Wilder 21:58, 14 April 2013 (UTC)

Ugh who has the code. 00:22, 15 April 2013 (UTC)

There are some links at the bottom of User:Capturebot2, but they might not be all of it. Peter ^mqzp 00:29, 15 April 2013 (UTC)

So the sandbox is capturing properly, as is the console, but wigocp isn't and nor do other pages added. Can somebody take a look from the server direction? Peter ^mqzp 09:00, 21 April 2013 (UTC)

Slow.

Like, 20-30 seconds to load a page slow. If the thunder don't get you/Then the lightning will. 16:07, 20 April 2013 (UTC)

Yup. And at this moment the server isn't showing high load or complex requests. I am puzzled - David Gerard (talk) 18:02, 20 April 2013 (UTC)

Tech blog post. Says nothing useful, but says a lot of it! - David Gerard (talk) 19:21, 20 April 2013 (UTC)

Better now. Possibly related: there was a Brasov-style vandalism edit last night my time. Any chance there's a bot of some sort running? If the thunder don't get you/Then the lightning will. 22:53, 20 April 2013 (UTC)

If you mean this then no, that wasn't "Brasov-style." The site's been ok for me for the last few minutes, but it was pretty bad yesterday. Peter ^mqzp 23:01, 20 April 2013 (UTC)

It's not quite better now - load around 7 (which is a LOT), Apache processes busy sopping up extragavant amounts of CPU again - David Gerard (talk) 23:49, 20 April 2013 (UTC)

From Futurama:

Cubert: We'll show 'em. All Awesome Express needs is a sturdy interstellar delivery craft.

[He shows Dwight an advert for a hovercraft.] Voilà! You got a quarter?

[Dwight pulls out a coin. Cubert rips out the ad and puts it in an envelope with the quarter and puts the envelope in the mailbox. It flies off down a message tube.]

Dwight: Man, the ad said to allow four to six seconds for delivery.

[Cubert snorts.]

Cubert: More like seven!

[A mail tube shoots out of the box and knocks Cubert off his feet.]

The video's funnier, but it's not on Youtube.--Krej^talk 02:24, 21 April 2013 (UTC)

Account Creation Issues?

A would-be-user came into the irc channel tonight complaining that account creation from his country (Indonesia) is blocked, and, seeing as he was obviously not a spambot, I offered to help him create an account, but found that I was continually getting errors related to a confirmation code being incorrect when I tried (from China) to make the account. Is this a known error? How can I help him out? FeiRuoWa (talk) 15:16, 22 April 2013 (UTC)

Account creation from his country? What happens when he tries to edit as an IP? Is his ISP's proxy open?

Do you have precise timestamps (and time zone) for when this happened, so I can see if the logs show anything? - David Gerard (talk) 16:09, 22 April 2013 (UTC)

I tried a handful of times between 19:12 and 19:14 yesterday, China time. I don't have contact info for him, so I've poked him on irc and am waiting for him to come back in (or check in here himself) to find out about editing as an IP. FeiRuoWa (talk) 09:30, 23 April 2013 (UTC)

I tried to create an account approximately 1-2 hours ago from an iPhone at a friend's house using her Comcast cable wireless network. Got the "confirmation code incorrect" error. Tried on my laptop at my home Comcast wireless network at approximately 19:00 PST 4/22/13. I was able to save an edit, which required a confirmation code, but couldn't create an account, even though the confirmation code question and answer were the same. - ^{— Unsigned, by: 98.234.94.233 / talk / contribs}

I know someone had troubles creating an account from their phone, because the phone company proxy was in the open proxy block list. I'm wondering about the open proxy block list and if it should be switched off again - David Gerard (talk) 07:46, 23 April 2013 (UTC)

Moar Account Creation Issues

Can one of you guys help me. I cannot create an account. When I try, it reads "Troll-B-Gone". What should I do?--184.6.214.237 (talk) 14:10, 25 April 2013 (UTC)

What it says. B-Gone.--Fergus Mason ^{If in the first act you have hung a pistol on the wall, then in the following one it should be fired. Otherwise don't put it there. - Anton Chekov} 14:29, 25 April 2013 (UTC)

I've been told by the community to get an account. You guys know how to fix it?--184.6.214.237 (talk) 14:43, 25 April 2013 (UTC)

I suggest waiting for some of the regular tech people to chime in. (Hint: Fergus Mason is not one of them.) I've checked the parameters of the last blocks you have received, but it looks like nothing should be blocking account creation.

I suggest clearing your cookies and trying again. What exactly is the error message you receive?--ZooGuard (talk) 15:35, 25 April 2013 (UTC)

This is exactly what it said (I copied it to here):
"Login error
Warning: This action has been automatically identified as harmful. Unconstructive edits will be quickly reverted, and egregious or repeated unconstructive editing will result in your account or IP address being blocked. If you believe this edit to be constructive, you may click Submit again to confirm it. A brief description of the abuse rule which your action matched is: Troll-B-gone"

Note: When I submit again, it says:

"Login error

This action has been automatically identified as harmful, and therefore disallowed.

If you believe your edit was constructive, please report this error. A brief description of the abuse rule which your action matched is: Troll-B-gone"

I have not cleared cookies yet, but i'll try that now.--184.6.214.237 (talk) 15:43, 25 April 2013 (UTC)

Just tried after I cleared cookies. Still gives the error message and won't let me through.--184.6.214.237 (talk) 15:51, 25 April 2013 (UTC)

It looks like your registration attempts had triggered the Abuse Filter. Looking at its log, it thinks that you have already created successfully an account as "AnAlternativeView" and you've been trying to register another one.

Try again to log in with that user name, do not click on the "Create an account" link.--ZooGuard (talk) 15:58, 25 April 2013 (UTC)

Okay, standby...--184.6.214.237 (talk) 16:01, 25 April 2013 (UTC)

I just tried to sign in under my username and it said:
"Login error
There is no user by the name "AnAlternateView". Usernames are case sensitive. Check your spelling, or create a new account."--184.6.214.237 (talk) 16:03, 25 April 2013 (UTC)

Oh... wait one second, I think I typed the wrong username. Standby...--184.6.214.237 (talk) 16:04, 25 April 2013 (UTC)

Even with the username that you said, it still says:
"Login error
There is no user by the name "AnAlternativeView". Usernames are case sensitive. Check your spelling, or create a new account." Any ideas?--184.6.214.237 (talk) 16:07, 25 April 2013 (UTC)

No, it was me who made a mistake. And yes, apparently there is no such an user. The edit filter log probably lists the user name from the attempted registration.

I think I got the root of your problem: the filter probably reacts to AnAlternateView. :) Try using spaces between the words or another username.--ZooGuard (talk) 16:14, 25 April 2013 (UTC)

You'd have the same problem at Conservapedia with that filth. Генгис^pillaging 16:18, 25 April 2013 (UTC)

When I tried to create an account with a different username, it returned:
"Login error
Incorrect or missing confirmation code."--184.6.214.237 (talk) 16:22, 25 April 2013 (UTC)

Make sure you get the letters right/count the number of brackets correctly/answer whatever other skill-testing question it gives you correctly. Shall we go, you and I while we can/Through the transitive nightfall of diamonds? 16:35, 25 April 2013 (UTC)

Hallelujah!!! Just got it to work! Thanks guys for all your help! Have a great day!--The Debater (talk) 16:39, 25 April 2013 (UTC)

Praise be. Shall we go, you and I while we can/Through the transitive nightfall of diamonds? 16:40, 25 April 2013 (UTC)

502 Bad Gateway The server returned an invalid or incomplete response.

Been getting this error message 1/3 to 1/2 of the time when I hit the main page bookmark in my browser ever since the update the other day. Did he doubt/Or did he try? 17:44, 11 April 2013 (UTC)

I am going to attempt to tweak a few things. Please let me know if you continue to see them. Tmtoulouse (talk) 19:50, 11 April 2013 (UTC)

I'm still getting tons of these--"Shut up, Brx." 16:29, 12 April 2013 (UTC)

Any improvement? Tmtoulouse (talk) 20:59, 12 April 2013 (UTC)

I've gotten a few 502's in the last few hours, which I never used to get. Never. A browswer refresh clears it. My bookmark lands me on Recent Changes. Most of the time there is no problem. Sprocket J Cogswell (talk) 21:07, 12 April 2013 (UTC)

I've yet to see one, which is somewhat more annoying than if it were reliable and reproducible. Trent, what are you tweaking? - David Gerard (talk) 22:06, 12 April 2013 (UTC)

I got two in the last hour. Do you want us to start sending you emails with the time it happened each time? Might clog up your inbox, but I know you asked for that last time there were issues like this. Also, pages might seem to be loading more slowly for me. But I'm not sure if it's just my imagination.--"Shut up, Brx." 03:36, 13 April 2013 (UTC)

Edit and save response is definitely slower. Again, not the end of the world, just noting for the record. VOX^HUMANA 04:03, 13 April 2013 (UTC)

I just got one, but before that I hadn't seen any in days. Peter ^mqzp 02:13, 14 April 2013 (UTC)

Seems to be browser/platform dependent. On Chrome/tablet: 502 every time; elsewhere & even with Chrome: OK. Scream!! (talk) 10:07, 14 April 2013 (UTC)

I've had them with Win7/Chrome, Mac OSX/Chrome and XP/Chrome. Haven't tried any other browsers though (I'm a bit of a Chrome fan) VOX^HUMANA 10:57, 14 April 2013 (UTC)

Chromium here, but haven't had one in a few days. If the thunder don't get you/Then the lightning will. 16:54, 14 April 2013 (UTC)

I'm still getting these, though, oddly, only when my desktop browser (Firefox 19/20 on Win7) tries to access it via bookmark and on mobile (Dolphin 9.3.2 on Android 2.3) when it initially tries to load as my homepage. I did get it to load properly as a homepage on mobile once by clearing my cache, then the problem came right back. Radioactive afikomen ^{Please ignore all my awful pre-2014 comments.} 18:03, 17 April 2013 (UTC)

Is it possible that it's choking on the redirect (rationalwiki.org -> rationalwiki.org/wiki/Main_Page)?--ZooGuard (talk) 18:14, 17 April 2013 (UTC)

The bookmark/homepage in both browsers is to my watchlist. Radioactive afikomen ^{Please ignore all my awful pre-2014 comments.} 21:42, 17 April 2013 (UTC)

I've had a couple of 502's. The wiki has also been intermittently very slow for me. Wonder if we're getting hammered by something - David Gerard (talk) 22:15, 17 April 2013 (UTC)

As above. Loads of 502s and slow wiki Scherben (talk) 21:58, 2 May 2013 (UTC)

As above, except the 502 errors are less frequent for me--P3A58NT86 22:20, 2 May 2013 (UTC)

Note to Web Developers: Main page redirection

Hi guys. I have a little tip for you (not sure if you're able to use it for RationalWiki). When going to RationalWiki.org, the browser redirects you to the main page (like it should). The only problem with that is that it uses an old redirection method that is noticed by the browser as an automatic redirection, and some people (like me) set up their browsers so that it blocks and asks whenever that happens. If you are able, you should set up a way for a file named 'index.html' to directly reference the main page (without 'index.html' actually existent. Just a server-side script that directly references the main page). I believe that this would fix the problem. (Note: i'm not an experienced server-side-scripter, but I do a bit of html work)--The Debater (talk) 21:16, 27 April 2013 (UTC)

Either the method shown above, or a little piece of javascript within the file 'index.html' would work as well:
<script="text/javascript">{window.location.href = '(the main page url)';}</script>--The Debater (talk) 16:44, 28 April 2013 (UTC)

Is ANYONE awake here??? I haven't received a reply in 2 days?!--The Debater (talk) 17:02, 29 April 2013 (UTC)

Someone!!! Wake up around here!!!--The Debater (talk) 18:51, 29 April 2013 (UTC)

Chill out. There are only a couple of people with server-side access & they may not always be immediately available for non-urgent issues. In the meantime, I doubt this issue is actually causing any disruption. ΨΣΔξΣΓΩΙÐ^{Methinks it is a Weasel} 19:16, 29 April 2013 (UTC)

Thank you sooo much!!! Another living human!!! Sorry about freaking-out, didn't mean to. Could you fix the issue my friend?--The Debater (talk) 19:22, 29 April 2013 (UTC)

The mustelld is not a tech. Генгис^{RationalWiki GOLD member} 19:32, 29 April 2013 (UTC)

Could you tell a tech about this?--The Debater (talk) 20:42, 29 April 2013 (UTC)

I have never heard of having to do such a thing. Do you have links showing anyone else on earth would benefit from this? - David Gerard (talk) 22:34, 29 April 2013 (UTC)

No, but if you want your website to be up-to-date, then you can do this. Plus, you will probably run into other problems if you leave it the way it is (outdated methods). Is it that hard to make a SSS or a file named index.html on the server? Couldn't you telnet over there and fix it remotely?--The Debater (talk) 15:40, 30 April 2013 (UTC)

I can't say I've heard of it, either. Surely, if it was "outdated" as so a major problem, the internet would be practically swimming in these complaints? ^postate 16:07, 30 April 2013 (UTC)

Bad link on blog network main page

The main RW blog network page has a link to the WIGO blog, but it seems to be missing a trailing slash. Clicking on it, I get an empty window. Sprocket J Cogswell (talk) 17:54, 2 May 2013 (UTC)

Either somebody just fixed it, or the black box by my left knee suddenly decided to start behaving itself. Sprocket J Cogswell (talk) 18:03, 2 May 2013 (UTC)

I literally fixed it just this minute. Well spotted! - David Gerard (talk) 18:05, 2 May 2013 (UTC)

Nice to know our best people are on the job... Sprocket J Cogswell (talk) 18:08, 2 May 2013 (UTC)

Stupid Squid tricks

The squid3 setup on rationalwiki.org is infuriating me. It lets through rationalwiki.org, it doesn't let through rationalblogs.org, and I have no idea why.

It's Ubuntu 12.04, with the distro squid3. We have a load balancer, which talks to two Squids, which talk to one Apache server. Here's the /etc/squid3/squid.conf . 192.168.162.183 is the Apache server as seen from the Squids.

http_port 80 accel defaultsite=rationalwiki.org vhost
cache_peer 192.168.162.183 parent 80 0 no-query originserver name=rwiki
acl our_sites dstdomain .rationalwiki.org
acl our_sites2 dstdomain .rationalwiki.com
acl our_sites3 dstdomain .rationalwiki.net
acl blog_sites dstdomain .rationalblogs.org
acl privateip dst 192.168.0.0/16
http_access allow all
cache_peer_access rwiki allow our_sites
cache_peer_access rwiki allow our_sites2
cache_peer_access rwiki allow our_sites3
cache_peer_access rwiki allow blog_sites
cache_peer_access rwiki allow privateip
cache_peer_access rwiki deny all
cache_dir ufs /home/squid 15352 16 512
cache_mem 512 MB
icap_uses_indirect_client off
logformat combined %{X-Forwarded-For}>h %ui %un [%{%d/%b/%Y:%H:%M:%S +0000}tl] "%rm %ru HTTP/%rv" %>Hs %<st "%{Referer}>h" "%{User-Agent}>h" %Ss:%Sh
logformat pagehits [%{%d/%b/%Y:%H:%M:%S +0000}tl] %ru
logfile_rotate 999
strip_query_terms off
access_log /var/log/squid3/access.awstats.log combined
access_log /var/log/squid3/access.pagehits.log pagehits
forwarded_for transparent
visible_hostname squid1.rationalwiki.org

What simple stupid thing am I doing wrong?

DNS for rationalblogs.org just changed - it should now resolve to 69.164.223.61 (the load balancer). If you see 50.22.105.163, the update hasn't reached you yet. The Squids are still seeing rationalblogs.org as 50.22.105.163 - could that be it? (I tried putting the correct IP in /etc/hosts, no difference.) - David Gerard (talk) 20:26, 27 April 2013 (UTC)

OK, further testing suggests the load balancer needs testing. Now waiting on Trent - David Gerard (talk) 11:14, 28 April 2013 (UTC)

So can anyone else rouse Trent? 'Cos I evidently can't - David Gerard (talk) 20:36, 28 April 2013 (UTC)

Trent roused, Nutty set G*D*ddy on fucking fire, and now it should resolve for at least some people. HOORAY! - David Gerard (talk) 06:49, 29 April 2013 (UTC)

Thank you - and Trent, and Nutty - for resolving this with such verve.--^talk 23:40, 13 May 2013 (UTC)

External links wandals

This action has been automatically identified as harmful, and therefore disallowed. If you believe your edit was constructive, please report this error. A brief description of the abuse rule which your action matched is: External links wandals

Getting this when I try to make any edit. ^{— Unsigned, by: CHRISTMASLIST / talk / contribs}

It's content based. We had a vandal spamming external links. It's a shitty filter that gets hit very very rarely so I'm disabling it for you. 21:54, 15 May 2013 (UTC)

Thank you. :) ^{— Unsigned, by: CHRISTMASLIST / talk / contribs}

YW 21:58, 15 May 2013 (UTC)

Spambots

Whatever was doing such a great job of keeping spambots off the wiki is failing spectacularly tonight. Faced with mysteries dark and vast/statements just seem vain at last. 03:52, 18 May 2013 (UTC)

Didn't we have a link filter? (See two sections above.) Peter ^mqzp 04:43, 18 May 2013 (UTC)

No, that's not what it did. I turned it back on, but it only checked 3 or 4 external sites, one of which CHRISTMASLIST was trying to link to. It's not a Good Idea to prohibit all linking for unconfirmed editors on a wiki that doesn't require credentials. 05:19, 18 May 2013 (UTC)

503s all day

Any reason for the persistent 503 errors today? I see other people have complained on the Tech Blog too.--ZooGuard (talk) 19:26, 23 May 2013 (UTC)

The web server was down - Trent rebooted it - David Gerard (talk) 11:21, 24 May 2013 (UTC)

Get SSL

You guys don't want people to plain text their passwords, there's a few sites which offer free certificates let's make rationalwiki secure: https://en.wikipedia.org/wiki/CAcert.org https://www.cacert.org/ https://startssl.com/ http://www.openssl.org/ https://www.sslshopper.com/article-free-ssl-certificates-from-a-free-certificate-authority.html https://www.godaddy.com/ssl/ssl-open-source.aspx http://www.makeuseof.com/tag/free-ssl-certificate/ etc ^{— Unsigned, by: Ethanwashere / talk / contribs}

Linode has directions for creating self signed (free) SSL certificates. That means there's not a lot of point on highly dubious and possibly dangerous services. One of them requires entering your home address, and explicitly forbids using a company address. The problem with self signed SSL and presumably any of these services, is that it generates attack website warnings in browsers. They're only recommended for things like email or other non browser related services. –Meine Ehere heißt Toleranz (Talk | Contribs | Ragebox) 18:12, 19 June 2013 (UTC)

Cannot create account

Hi. I keep getting the following error when trying to create my account (mschlindwein) here:

Login error

Incorrect or missing confirmation code.

What does the system mean by "confirmation code"?

I'm an active Wikipedia user, and I have relative good know-how on how Wikis work. And, of course, I'm sure I typed my password correctly every time and properly answered the (changing) question to avoid bots. ;o) I also saw above that other users had the same issue in April.

Could you please help? Thanks,

mschlindwein^{— Unsigned, by: 78.233.158.215 / talk / contribs}

I think this is the message one gets after giving a wrong answer to the anti-spam question.--ZooGuard (talk) 14:26, 17 May 2013 (UTC)

If you're using a proxy, you might want to consider using a clean IP address at least to set your account up. I think we're still preventing account creation and editing from a staggering number of proxies because of our last serious attack. 14:37, 17 May 2013 (UTC)

I'm not using a proxy, and I'm giving the correct answer(s) to the anti-spam question(s), which are the same to be able to write here ;o). Any idea? Thanks. mschlindwein

Hmmm... unfortunately, the sysadmin is moving house and out of action for some time so we can't really check if the anti-spam is borked on that end. ^sshole 15:00, 17 May 2013 (UTC)

Worked fine for me there. I now have yet another sock account I'll never use. Reject (talk) 15:04, 17 May 2013 (UTC)

Is "what is this wiki's name" case sensitive? Because officially it's CamelCase but that doesn't seem to be accepted. Reject (talk) 15:05, 17 May 2013 (UTC)

I type the answers always as they are, ie, case sensitive. And this was not the only question I had, I tried all the other 2 (number of brackets and the first and last letters). I tried with Firefox and with Safari, and I live in France. I have no idea what it can be. Thanks. .

If it can help, I can try to register "mschlindwein" with a default password and set it to you. ^theist 15:55, 17 May 2013 (UTC)

Yes, Armondikov, please. Please send the password to my e-mail address: marcelo_schlindwein@yahoo.com. Many thanks, best regards! mschlindwein

I'm having the same problem exactly - same error message, refreshing my browser doesn't help, using TorBrowser to get a new IP address gives the error: Your IP address is listed as an open proxy in the DNSBL used by RationalWiki. You cannot create an account.

I'd like to have an account and contribute, but I'm at a loss. Could someone create an account for me and email the password to chrismlanguage@gmail.com .. account name chris_lang .. I would appreciate it. I'll check back here for any response as well.

I have disabled the CAPTCHA for now. Please see if that allows you to create accounts. 206.192.168.26 (talk) 21:13, 17 May 2013 (UTC)

Yes, it worked for me and I could (finally) create my account. The CAPTCHA seems to be the culprit... Thanks a lot! Mschlindwein (talk) 21:22, 17 May 2013 (UTC)

Same problem here. Tried 2 different browsers (no proxy), always getting the same message as above. --93.71.83.153 (talk) 19:23, 14 June 2013 (UTC)

Anybody? Also, is it normal that I'm always getting the same captcha question? --93.71.83.153 (talk) 19:26, 15 June 2013 (UTC)

Helloooo... anybody home? I still can't sign up... --93.71.83.187 (talk) 15:51, 16 June 2013 (UTC)

Same Captcha for successful first attempts on Firefox, Safari, and Chrome. I've tried reproducing this on the very rare occasions it's mentioned and cannot. It's possible that there are multiple Captchas with the same question, although different (incorrect) answers but I doubt it. Sorry buddy. 16:03, 16 June 2013 (UTC)

Isn't there at least a way to force a different question? I've been getting the same one for days now (also for editing, but there it works). --93.71.83.187 (talk) 17:16, 16 June 2013 (UTC)

For people who don't like red links when users don't have user pages

You can remove red links in your CSS page... –Александр(а) Ehrenstein (Talk | Contribs | Ragebox) 05:54, 29 May 2013 (UTC)

User:Username/common.css

a { color: #0000FF; }

–Meine Ehere heißt Toleranz (Talk | Contribs | Ragebox) 02:33, 20 June 2013 (UTC)

That doesn't tell anyone what to actually do. Add this to monobook.js if you want to get rid of red links, although there's a reason they exist. ... $("a.new").css( { color: "blue" } ); 02:56, 20 June 2013 (UTC)

I've always followed this adage: Don't like red linked userpages??? Simple: GET OVER IT!!!! Reckless Noise Symphony (talk) 08:59, 20 June 2013 (UTC)

^This. Генгис^{You have the right to be offended; and I have the right to offend you.} 12:48, 20 June 2013 (UTC)

But it wouldn't be RW without a solution to a problem that doesn't exist. 13:21, 20 June 2013 (UTC)

Image disparity in Zionism article

After repeated squid refreshing, the Zionism article still shows a thumbnail of an ocean-going vessel near a crowded shore. It's supposed to be a D9 'dozer with defensive fittings. Sprocket J Cogswell (talk) 15:27, 27 June 2013 (UTC)

It shows a D9 Cat when I look at it. Генгис^{You have the right to be offended; and I have the right to offend you.} 16:16, 27 June 2013 (UTC)

I get the ship as well. Balaam (talk) 17:33, 27 June 2013 (UTC)

I also see a ship. EDIT: I was looking at it earlier and saw a sheep, but now that I look at it again it a tractor. What gives? - ConservapediaMarkman (talk) 17:43, 27 June 2013 (UTC)

Still getting a ship after local cache purge. Sprocket J Cogswell (talk) 18:04, 27 June 2013 (UTC)

The ship image was uploaded over the dozer image. Our image page failed to update, somehow, while the image use updated across the wiki. Hipocrite (talk) 18:57, 27 June 2013 (UTC)

(PS - there is a bug somewhere.) Hipocrite (talk) 18:58, 27 June 2013 (UTC)

Weird. I don't think I'm dreaming when I remember looking at the image page and seeing two versions of the dozer in the upload history, but no seaside scene. All better now, thanks! Sprocket J Cogswell (talk) 21:31, 27 June 2013 (UTC)

illegible to vote

I am illegible to vote, can this be changed please?--Patmac (talk) 11:09, 3 July 2013 (UTC)

It's automatically based on the time you have been registered and the number of edits you have made. So it's not something anyone else can change. I'm sure someone else can point you to the exact values. Oh, and BTW, illegible means "Not clear enough to be read." Генгис^{Our ignorance is God; what we know is science.} 21:09, 3 July 2013 (UTC)

Yeah, my spelling is crap, thanks for the clarification though--Patmac (talk) 21:14, 3 July 2013 (UTC)

Eligibility criteria are being at least 3 months old and having at least 75 edits, neither of which you satisfy, unfortunately. Blue (is useful) 21:40, 3 July 2013 (UTC)

If this editor is under three months old yet has managed to not only read and write but can also use wiki software then perhaps an exception should be made!--Weirdstuff (talk) 09:50, 4 July 2013 (UTC)

IP editing question typo

aaaaaa just go to sleep already me

Just thought I'd point this out: the "name of website" check has "webiste". 98.27.34.197 (talk) 13:40, 6 July 2013 (UTC)

bible tag not working correctly

There seems to be a problem with the <bible> tag - it's supposed to provide a quote and a link to the appropriate section of the Annotated Bible (see RationalWiki:Annotated Bible#Bible Generator), but the link doesn't work - it points to the page on which the tag is used instead. For example:

Genesis 1:1
In the beginning God [Elohim] created the heaven and the earth.

--ZooGuard (talk) 19:24, 23 May 2013 (UTC)

Yeah, I noticed that too. Will investigate if I have a tick today. (I am hampered by having no broadband at home, and work being a bit busy) - David Gerard (talk) 11:21, 24 May 2013 (UTC)

I think it may have something to do with the recent change in verse section naming.--ZooGuard (talk) 08:06, 29 May 2013 (UTC)

That'll be it. (I have internet back at home, so might actually look at it.) - David Gerard (talk) 08:53, 13 June 2013 (UTC)

Any progress with this?--ZooGuard (talk) 06:34, 7 July 2013 (UTC)

Of course not *cough* Er, soon probably! - David Gerard (talk) 17:51, 8 July 2013 (UTC)

Global JS script includes

DG or Trent: Are they accessible on-wiki or only from the code? I want to experiment with something grand. 01:24, 9 July 2013 (UTC)

Ninja flag

I seem to have lost the ability to turn Ninja off. The button is greyed out on my user rights page (although Ninja is still turned on). Any ideas? VOX^HUMANA 13:56, 6 July 2013 (UTC)

And of course, my now-permanent ninja flag means no-one ever notices when I leave a comment on Tech Support. Oh well. VOX^HUMANA 10:23, 8 July 2013 (UTC)

I've un-Ninja'd you, that's as far as my wiki-prowess goes. Генгис^{Gum disease} 18:01, 8 July 2013 (UTC)

Umm, you might want to re-ninja him. The issue is not ninjosity, but botness. Vox is still a bot, hence invisible in the default view of recent changes. No longer being a ninja, he cannot change that status on his own. Sprocket J Cogswell (talk) 18:09, 8 July 2013 (UTC)

I see Nutty has taken care of it. Генгис^{Our ignorance is God; what we know is science.} 05:47, 9 July 2013 (UTC)

The deeper problem is that I will want to be a ninja again, lest my hundreds of small edits to the Dover project blow RC to oblivion. Ninja is still turned on, it is greyed out so I can't change it... I'm pretty mystified by it all. VOX^HUMANA 09:36, 9 July 2013 (UTC)

You can't change "ninja"—it's the "bot" group itself that's important. Ninja just lets you become a bot, having it doesn't change anything else directly. Peter ^mqzp 09:59, 9 July 2013 (UTC)

Spambot for rent?

Before being deleted, User:CarltonDaniels claimed to link a site offering a recaptcha solver for rent, with a 60% yield. Interesting, or not? I don't need an answer, but thought someone might like to know. Sprocket J Cogswell (talk) 12:10, 9 July 2013 (UTC)

You can do a lot better buying from catchpa clouds - ref http://www.captchabrotherhood.com Hipocrite (talk) 13:32, 9 July 2013 (UTC)

DDOS-alike

http://rationalwiki.blogspot.co.uk/2013/07/wiki-slow-were-being-hammered.html - David Gerard (talk) 23:24, 12 July 2013 (UTC)

Did you piss a crank off again, David? 02:44, 13 July 2013 (UTC)

Still running really slow here. We still getting hammered? The bus came by/and I got on. 14:54, 14 July 2013 (UTC)

User rights list?

Is it possible to see a list of everyone who is "autopatrolled" but not a sysop? I suspect that there might be a few users worthy of mops and buckets that we can't see because they don't leave red bangs in RC. PowderSmokeAndLeather: Say something once, why say it again?. 21:33, 24 July 2013 (UTC)

Yes. 22:28, 24 July 2013 (UTC)

That gives everyone who is autopatrolled including the folks who are sysops. Anyway to winnow it down for ease of navigation? PowderSmokeAndLeather: Say something once, why say it again?. 22:36, 24 July 2013 (UTC)

Do/Should WIGO Pages (not talk) auto-trim/archive?

As topic. WIGO:World might seems to be a bit long to load properly at times. [[User:K61824|]][[User_talk:K61824|]] 17:51, 30 July 2013 (UTC)

Is WIGObot down again? - David Gerard (talk) 07:43, 31 July 2013 (UTC)

Entries in May are still up, so does that count as a yes? [[User:K61824|]][[User_talk:K61824|]] 13:23, 31 July 2013 (UTC)

So who runs WIGObot, anyway? - David Gerard (talk) 13:42, 31 July 2013 (UTC)

Update: Entries in May are still up with the start of August entries. [[User:K61824|]][[User_talk:K61824|]] 11:12, 1 August 2013 (UTC)

And WIGO CP/talk seems to get trimmed faster than necessary these days. ħuman 05:18, 10 August 2013 (UTC)

People keep changing the time limit. Генгис^{You have the right to be offended; and I have the right to offend you.} 06:45, 10 August 2013 (UTC)

Perhaps the time limit should be changed to a kb limit or a message number limit? ħuman 05:13, 11 August 2013 (UTC)

The question is broken

Sometimes, "What is the name of our webiste?" does not accept the correct answer - many, many times. And if it is supposed to catch spam, it shouldn't be the same question every time.--CIA (talk) 21:14, 7 August 2013 (UTC)

Spammers, good news and bad.

Bad news, we're seeing an influx of spammer accounts. Good news, they seem pretty much unable to make edits. Thanks to the folks that make the latter possible. PowderSmokeAndLeather: Say something once, why say it again?. 17:28, 10 August 2013 (UTC)

site outage ?

I kept getting a 500 message about 30 minutes or so ago. something about no server available ? was there a problem or was it just my link ? Hamster (talk) 05:04, 11 August 2013 (UTC)

Me too, a couple of 503s, but obviously it is all better now. I hope. ħuman 05:11, 11 August 2013 (UTC)

Upgraded some aspects of the server requiring a shutdown. Tmtoulouse (talk) 05:11, 11 August 2013 (UTC)

all better then. ħuman 06:14, 11 August 2013 (UTC)

Linode has also been upgrading the servers. We recently got twice as much storage space at no cost. –Александр(а) (Talk | Contribs | Ragebox) 05:40, 13 August 2013 (UTC)

References on talk pages

Can something be done in the edit filter to prevent <ref> being posted on talk/discussion pages, or to make it render as nowiki text rather than an actual footnote? I'm tired of seeing people pasting content with references on talk pages without tidying it first, the resulting red "missing tag" message at the bottom of the screen, and the references sections some users then shove in at arbitrary points in the page. Wëäŝëïöïď^{Methinks it is a Weasel} 15:48, 11 August 2013 (UTC)

You're being serious, aren't you? 15:56, 11 August 2013 (UTC)

Yes, why wouldn't I? Footnotes don't belong on talk pages. Wẽãšẽĩõĩď^{Methinks it is a Weasel} 15:57, 11 August 2013 (UTC)

Just add a <references/> tag at the end of it. That way we can still see what the references are. –Александр(а) (Talk | Contribs | Ragebox) 15:59, 11 August 2013 (UTC)

Of course there are legitimate uses; for example, see here. ^talk 16:01, 11 August 2013 (UTC)

(EC) At the bottom of that, I see "Refs from the above section", but if anyone adds more refs below this section, the footnote would still appear in that same section. It's much better not to use footnoting on talk pages at all, and to put any references or links alongside the text e.g. in brackets. Hence why automatically nowikiing any ref tags on talk pages would be a good idea. Ŵêâŝêîôîď^{Methinks it is a Weasel} 16:11, 11 August 2013 (UTC)

Because I know you're a smart person and can figure out that (a) like Sterile says, there are legit uses, (b) you can fix it yourself instead of simply talking about it as usual, and (c) I'm sure you can imagine that breaking the wiki isn't going to be high on any developer's priority list. Hope this helps, Weaseloid. Time to start putting your money where your mouth is and doing something instead of offering your sage opinion. 16:07, 11 August 2013 (UTC)

Wow, sarcasm is not actually what we call "technical support". ħuman 01:57, 12 August 2013 (UTC)

I don't have the access rights or technical knowledge to do what I'm suggesting. Hence why I'm talking about it. Wěǎšěǐǒǐď^{Methinks it is a Weasel} 16:11, 11 August 2013 (UTC)

There are easy ways to create groups of references to avoid that "problem." ^talk 16:13, 11 August 2013 (UTC)

I also haven't run into problems with multiple references tags. –Александр(а) (Talk | Contribs | Ragebox) 16:18, 11 August 2013 (UTC)

I've seen a few clunky examples of what the mustelid refers to, but of course they all get archived pretty quickly. ħuman 01:57, 12 August 2013 (UTC)

Configuring a wiki to make all users sysops

What changes should I make for this to be safe? I figure that some them will be restricting system message editing and editing of user css and js pages. –Александр(а) (Talk | Contribs | Ragebox) 04:43, 13 August 2013 (UTC)

You can set capabilities in MW in a fairly fine-grained manner. What RW does is give some of the more potentially-damaging powers to bureaucrats-or-the-equivalent - David Gerard (talk) 10:53, 13 August 2013 (UTC)

Spammers

I'm going to write up a script to deal with them. I'll start with a new set of questions for confirm edit and possibly create a package of anti spam extensions with pre configured settings. We've been getting quite a few of them. –Александр(а) (Talk | Contribs | Ragebox) 22:16, 13 August 2013 (UTC)

I just wrote a short series of new questions that should help hold back the spammers. While writing your own questions that are unique to your own website works very well at keeping out bots, someone will eventually program the answers into the spam bots, especially if the website is popular.

Here's the script I wrote:

https://github.com/Inquisitor-Sasha/ConfirmEditQuestions

It includes the line require_once( "$IP/extensions/ConfirmEdit/QuestyCaptcha.php");, so that should be removed from LocalSettings.php in addition to the existing questions.

I'll work on a larger anti spam package with multiple extensions and settings to keep out spam. The DNS blacklist that I had added from the Brasov attack is good for most vandals, but it doesn't have enough blacklists to be good against spammers.

The extension can be installed by adding it to the /extensions/ folder and adding require_once("$IP/extensions/ConfirmEditQuestions/ConfirmEditQuestions.php"); to LocalSettings.php.

–Александр(а) (Talk | Contribs | Ragebox) 23:10, 13 August 2013 (UTC)

Knowledge-based questions really aren't a wise idea, plus there are multiple valid answers for most or all of those. Wẽãšẽĩõĩď^{Methinks it is a Weasel} 23:13, 13 August 2013 (UTC)

I can change them then. I also chose questions that people here would be likely to know. If you are interested in contributing any, you can create an account at GitHub. They have a web based editor there, so you don't need to install git. –Александр(а) (Talk | Contribs | Ragebox) 23:18, 13 August 2013 (UTC)

Verification questions aren't targeted at people here. They're targeted at people using the site for the first time. Site visitors shouldn't be obliged to correctly answer some random question about religion before they're allowed to edit or comment. That would do no good for the site's reputation or growth. Wéáśéĺóíď^{Methinks it is a Weasel} 23:32, 13 August 2013 (UTC)

I can change it, but I intended questions that the sort of people coming here (who aren't religitard trolls) would know. If you think that not everyone who comes here would know the answers than I can change the questions. –Александр(а) (Talk | Contribs | Ragebox) 17:10, 14 August 2013 (UTC)

You're missing the point. We shouldn't use knowledge-based questions as we shouldn't restrict use of the site only to people who know certain things or make generalisations about what things site readers and potential editors are interested in. We do already have a verification question which is something like "what is the fourth letter of our site name?" and which is effective in stopping almost all automated spam. Weaseloid^{Methinks it is a Weasel} 17:49, 14 August 2013 (UTC)

I would also add that the answers to those questions are wrong, esoteric, subject to multiple spellings, or weird. David Shepard wrote numerous additional questions that serve the purpose of deterring people scraping the page and automating signups. I'm sure he'll do what's appropriate, but at this point I don't gather that he thinks people creating accounts they do nothing with is that big a deal. You can't make it more difficult for real humans to sign up. 18:55, 14 August 2013 (UTC)

I see the problem and so I'll change the questions to all be about things that can be seen from the account creation page. The problem with questions is that while they're unbeatable by bots for a long time, programmers will eventual give spambots the answers to the questions. I did notice the issue of multiple answers, since technically Darwin wasn't the first person to propose evolution; I can't remember the actual details. Multiple spellings though are a problem. I'll redo the questions and also find additional DNS blacklist domains for our existing blacklist. –Александр(а) (Talk | Contribs | Ragebox) 19:25, 14 August 2013 (UTC)

We already have two developers. 21:33, 15 August 2013 (UTC)

I'm just helping to contribute. –Александр(а) (Talk | Contribs | Ragebox) 05:24, 16 August 2013 (UTC)

There are data analysis and visualization projects if you're up for it. Trent and David Gerard have the wiki stuff covered. 02:07, 18 August 2013 (UTC)

I'll take a look. If I can handle it, I'll work on it. –Александр(а) (Talk | Contribs | Ragebox) 08:10, 18 August 2013 (UTC)

We could also pray for the spam to stop. –Александр(а) (Talk | Contribs | Ragebox) 06:48, 14 August 2013 (UTC)

Protection

maybe it's just me being a tech klutz, but I just tried changing the length of time an article was protected. The software said "expiry time invalid" and I tried again and it worked. However, on the second attempt I didn't give a reason. Is something borked (ie you can't change protection terms and give a reason) or is it just me being Miss Fail again? Sophie^Wilder 21:09, 15 August 2013 (UTC)

Did you maybe write your reason in the "other time" field? I've done it loads of times by mistake as it looks like it should be a comment box. Ŵêâŝêîôîď^{Methinks it is a Weasel} 21:32, 15 August 2013 (UTC)

Ahhh yes, that'll be it. Klutzella strikes again. Sophie^Wilder 09:47, 16 August 2013 (UTC)

:) ħuman 01:29, 18 August 2013 (UTC)

Sitenotice

Uh, I blanked MediaWiki:Sitenotice over 36 hours ago, but it's clearly still appearing as not-blank on some people's browsers. Is there a way to fix this?

We're still getting donations and emails, and while I obviously love that this is happening, I don't want to wear people out by making it appear that the fundraiser is going for too long, you know? Radioactive afikomen ^{Please ignore all my awful pre-2014 comments.} 19:45, 17 August 2013 (UTC)

Its the squid cache will dissipate fairly quickly. Tmtoulouse (talk) 19:48, 17 August 2013 (UTC)

Alrighty. Radioactive afikomen ^{Please ignore all my awful pre-2014 comments.} 19:55, 17 August 2013 (UTC)

Blog upgrade, plugins reenabled

I've reactivated plugins on RationalBlogs, and upgraded WordPress to 3.6. The only plugins switched on at present are Akismet ('cos I'm sick of the FUCKING SPAM EMAILS) and WP-SuperCache; let's see if we get the white-screening again ... - David Gerard (talk) 14:03, 18 August 2013 (UTC)

Problem with Picture Uploads

My Toolbox doesn't show the 'upload pics' option. I recently made a new article on Bill Wiese, seen here, and I wanted to add two pictures to the article, this one [1] and this one [2]. The first one with Bill would be under the Bill first section, and would say 'For a guy that was tortured by an asshole God, Bill sure does smile a lot...' The second picture would be in the 'Bill's Experience' section and would say 'Actual pictures of the demons' I'm writing this in hopes that someone could be help me and upload the pictures in question, as I can't do it -User:TokenSkepticMagician 1:31 20 August 2013 (UTC)

See if it appears now. Tmtoulouse (talk) 17:08, 19 August 2013 (UTC)

Got it. You're officially Awesome! -User:TokenSkepticMagician 2:26 20 August 2013 (UTC)

rationalwiki.bit

Just for the halibut I just pre-registered rationalwiki.bit (at the low low cost of 0.1 Namecoins, or about 6 cents) before some domain squatter grabbed it. Quick question for one of you techie types so I don't screw this up (each attempted update/screw up costs me another 0.005 Namecoins and cuts into my retirement stash, the horrors!). What's the correct way to point it to rationalwiki.org? Just a simple mapping the domain to 69.164.223.61? Or is there anything more complicated I need to know first such as custom subdomains, or the webhost not accepting unknown TLDs like .bit? Secret Squirrel (talk) 17:46, 24 August 2013 (UTC)

From RationalWiki:Technical_support/Archive9#Stupid_Squid_tricks I would guess the server is set up to accept rationalwiki.org, .com, .net, and rationalblogs.org, and to deny any other domain name someone points to it, correct? Secret Squirrel (talk) 18:52, 24 August 2013 (UTC)

The setup is fairly complicated 69.164.223.61 is actually our node balancer which uses a round robin system to send traffic to one of two squids, which then contact the apache server backend. Each of these will need to be setup, another easier solution is to setup rationalwiki.bit as an alias of rw.org so that traffic to rw.bit is forwarded to rw.org. That can be done all on your end. Tmtoulouse (talk) 20:57, 24 August 2013 (UTC)

Hmm..the namecoin client (thorough which .bit domains are registered) doesn't seem to have aliasing as an option yet, although according to what I was able to find on forums it is planned for the future. There is a "translate" option which is supposed to do a translation step before delegation, so the syntax looks like this:

{"translate": "rationalwiki.org", "ip":"69.164.223.61", "map": {"*": {"ip":"69.164.223.61"}}} for mapping the domain directly to 69.164.223.61 and doing a translation step to rationalwiki.org first, or

{"translate": "rationalwiki.org", "ns": ["69.164.223.61"]} for doing the translation step and pointing to 69.164.223.61 as a DNS server.

Or, either of the above two can be done omitting the "translate" step, which would keep the domain as rationalwiki.bit.

In any case I'll figure something out - may just leave the domain parked for now and point it to rationalwiki.org if/when I can. Sorry...I've been playing around with the namecoin client and it still seems rather undeveloped, although I'm intrigued by the idea. Right now .bit seems to be a haven for cybersquatters grabbing domain names because they are so cheap - as mentioned the entire process costs about 6 cents per name if done directly through the namecoin client. Secret Squirrel (talk) 22:30, 25 August 2013 (UTC)

My watchlist...

...is broken. ħuman 04:25, 10 August 2013 (UTC)

Nothing? My watchlist is how I have tracked the site for 6+ years, now it is broken. Reasons, explanations, excuses? ħuman 05:12, 11 August 2013 (UTC)

Oh but nothing here. Why does a simple wiki button no longer work? ħuman 06:15, 11 August 2013 (UTC)

It was weirdly broken a while ago - did it start working between then and now, and stop working again? - David Gerard (talk) 11:35, 11 August 2013 (UTC)

I don't think so, no. I stopped working between then and now, I think ;) Thanks for stopping by, I appreciate it. ħuman 01:49, 12 August 2013 (UTC)

See if the page loads. Tmtoulouse (talk) 16:21, 12 August 2013 (UTC)

Dang, no it didn't, and my hopes were high. Thanks for trying, any more ideas? ħuman 03:56, 13 August 2013 (UTC)

IS the error still nothing but a blank screen? Tmtoulouse (talk) 03:57, 13 August 2013 (UTC)

Yes, it still a cold blank screen, no error messages, simply nothing. ħuman 08:06, 13 August 2013 (UTC)

When I tried diagnosing this before, it was MediaWiki running PHP out of memory. You can see it right there in the php_errors log (which is in the MediaWiki directory). I duplicated the list and the duplicated list didn't do it. It'll be something obscure in Human's settings doing it, though I didn't get as far as duplicating everything else about the account - David Gerard (talk) 10:51, 13 August 2013 (UTC)

And how many pages art thou watching? ^talk 10:55, 13 August 2013 (UTC)

I dunno, it's hard to tell without a watchlist. About a third of the wiki, perhaps? Is there a limit or problem there? ħuman 08:51, 15 August 2013 (UTC)

Tumbleweeds... why should how many pages I watch matter? It's a simple feature of MW, and it's broken. It's been broken for a long time now. Any chance anyone who cares can fix it? ħuman 04:06, 21 August 2013 (UTC)

At this point about the only thing that could be done is for me or david to take over your account and see if we can replicate it. Tmtoulouse (talk) 04:23, 21 August 2013 (UTC)

Feel free to knock yourselves out in your copious free time. I can use one of my dozens of obvious sockpuppet accounts until you let me back in to mine. Just pmail me or something if you are going to take it over for a bit, need the password, whatever. Thanks for the reply! ħuman 03:32, 22 August 2013 (UTC)

Still hoping for salvage. ħuman 03:21, 26 August 2013 (UTC)

It is still dead. This is stupid and embarrassing. I cannot use my watchlist? Please fix, ASAP. ħuman 03:48, 29 September 2013 (UTC)

Who do I ask about getting a file off the server?

Before you all freak out, there's a problem going on over at NaturalNews that's breaking all the references. I think I have the way to fix it, but it would require a hatch at the PHP code (I have exhausted using templates since "a" isn't an allowed HTML tag). I think I need the file named Cite.php for this. Anyone willing to let me take a stab at this? Zero (talk) 18:58, 23 September 2013 (UTC)

Cite.php is from an extension [3] take a look and make sure its what you think you need and we can work out a patch. 206.192.168.26 (talk) 19:23, 23 September 2013 (UTC)

I would suggest that you need to take it with these people here.--Bob^{"I think you'll find it's more complicated than that."} 19:31, 23 September 2013 (UTC)

That's Trent. 23:05, 23 September 2013 (UTC)

If you are talking about rel="noreferer", how many browsers implement it? And why do you need to modify the cite extension? Links in references are provided by the standard MediaWiki link mechanism, not the ref tags themselves. And unreferering all the links in the footnotes is probably overkill.--ZooGuard (talk) 07:46, 24 September 2013 (UTC)

Not for special cases like NaturalNews, arguably - what Mike Adams is unhappy about is his own words being used as references for the batshit stuff he says. Zero, have you looked on mediawiki.org for a suitable extension that already exists and is maintained? Most of our nightmares about RW upgrades concern our hacky little pile of RW-exclusive extensions (like the voting mechanism) - David Gerard (talk) 07:58, 24 September 2013 (UTC)

I was thinking about how you make that a lot easier - it's not fun the first time, but it prevents the upgrade problems you're talking about: can you make an extension that creates a private API? That way the JS in common.js that gives stuff like voting its functionality browser side can make its AJAX calls with you only needing to make sure your extension is installed. No patching necessary. 15:05, 24 September 2013 (UTC)

Worth a look. Zero (talk) 15:10, 24 September 2013 (UTC)

Oh sorry - I should have said it's irrelevant to your issue and probably not posted here at all. 15:30, 24 September 2013 (UTC)

Oh, you just want Template:Unref - David Gerard (talk) 08:09, 24 September 2013 (UTC)

Answers in order: Browsers that support HTML 5 (which is about anybody not using Windows Vista or earlier at this point, IE10 for Win7 supports) support noreferer. I wasn't thinking of setting the entire system to unrefering, just include the ability to specify an unrefer when wanted.

I'm still in the research phase. I haven't started developing a solution. I'll come back when I've got more interesting things to say about this subject. I think I'll need a testbed of MediaWiki somewhere on my own site. Zero (talk) 14:09, 24 September 2013 (UTC)

One more thing. The Unref template doesn't work for anybody behind a filter of some sort since the site is considered a "bypass" and is blocked by default. Zero (talk) 14:49, 24 September 2013 (UTC)