Please do not post any new comments on this page. This is a discussion archive first created in 2024, although the comments contained were likely posted before and after this date. See current discussion or the archives index.
Why bot encodes decoded links?
Latest comment: 8 months ago4 comments2 people in discussion
Latest comment: 8 months ago4 comments3 people in discussion
I have experienced problems with the IABot not functioning as normal, and timing out. I hope this can be investigated and resolved. Marshelec (talk) 06:32, 12 January 2024 (UTC)
Same issues encountered here, I have been experiencing 504 Gateway Timeout Error for 95% of my run since 2–3 days ago even though the IABot just needed to archive 1–3 sources for my every run, looks to me on surface that it's iterating through every single sources regardless if it has been archived. Paper9oll (talk) 07:25, 14 January 2024 (UTC)
Hubba, this appears to be due to geo-restriction interfering with our U.S.-based link checker. I have added that domain (and the root domain and www. variant) to our permalive list. Harej (talk) 21:01, 27 March 2024 (UTC)
Latest comment: 8 months ago3 comments2 people in discussion
Please see this diff. I'm not sure what's going on, but InternetArchiveBot keeps adding incorrect archive links pointing to a googleads.g.doubleclick.net page that doesn't seem to exist rather than to the kyodonews.net link that's actually present in the reference. (It's also edit-warring with Citation Bot, which correctly removes the bad archive links.) This appears to be a bot problem rather than an Internet Archive problem, as the proper link does exist in the Internet Archive. Jay8g (talk) 00:20, 7 January 2024 (UTC)
it:Special:Diff/135572845. The links are working both before (archive.is) and after (archive.today), leading to the same archive page : If it ain't broke, don't fix it. Moreover, the bot is doing the opposite of another robot IrishBot (.avgas) : it:Special:Diff/133347863.
it:Special:Diff/136872396. Proposed archive page is an error message : "The Wayback Machine has not archived that URL. This page is unavailable for archiving. The server returned code: because page does not exist".
Proeksad, for whatever reason the "Cbignore" template was not configured as a setting for Russian Wikipedia. This setting has now been changed. Harej (talk) 20:26, 3 April 2024 (UTC)
The site has a "Are you human?" check box and that is probably the cause. I set the domain to Subscription for now. It will stop the bot from changing it to dead. It also means that bot won't be fixing dead links, for this domain. -- GreenC (talk) 15:01, 17 March 2024 (UTC)
Latest comment: 8 months ago4 comments3 people in discussion
Hi!
I'm translating InternetArchiveBot user interface into Hebrew, and I have a question.
The message urldatachangestate says "from <b>{{logfrom}}</b> to <b>{{logto}}</b>". I guess that "{{logfrom}}" and "{{logto}}" are something like "live", "dead", etc., but can you please explain more specifically what are the possible values?
And are they always in English, or can they be translated?
I'll update the documentation for translators after you reply.
Dead, Dying, Alive, Unknown, Subscription, Permadead, and Permalive are the statuses and yes, they are translatable. —CYBERPOWER(Chat)21:09, 3 April 2024 (UTC)
I've run into this a bit when going through the url=value CS1 pages. So, this bot was just run on https://en.wikipedia.org/w/index.php?title=Tony_Martin_(British_singer). If you look at the comparison between 9 January 2024 and 2 March 2024 (04:13), you'll see that one of the changes made was to the shortened Billboard link used by previous editors. I'm fixing it with the long links, but it seems IABot wants to change the symbols used to shorten URLs on Wikipedia into the code used in URLs? I've been fixing these for a while, but they aren't the only issues I come across in the CS1 pages, so it's the first time I've noticed which bot is doing this particular function.
Latest comment: 8 months ago3 comments2 people in discussion
Can you please authorize me to use IABot for Gagauz language to on the gagwiki (Gagauz Wikipedia)? I can currently use it for English (enwiki) and Russian (ruwiki), but not the Gagauz one.
When I try to use the bot on a gagwiki (Gagauz Wikipedia) page, I get "Permission error" and "The action you are trying to perform requires the analyzepage permission." and "This permission is obtainable with the following groups: basicuser, user, admin, root, bot".
Bot (innocently) allowing itself to look rude and arrogant/condescending/entitled
Latest comment: 8 months ago3 comments2 people in discussion
In case this has already been fixed, I apologize for being behind. I have no way of knowing whether it has, and/or have not found a place where I would have had. I guess something might be in the docs, but it has not been obvious or easy to find for me, sorry.
tldr: This could IMHO be fixed without any fuss and for goodwith a flick of the wrist by just adding a few words at the start of the first paragraph of the bot's message, making it begin with "Internet Archive Bot [Link] here." /tldr
I came across a place ([here https://en.wikipedia.org/wiki/Talk:Aerospace_engineering] and in fact many more) where there is a section, created by this bot, titled "External links modified", followed by an IMHO appropriate greeting, "Hello fellow Wikipedians", followed by a number of very appropriate factual statements, BUT THEN followed by,
"When you have finished reviewing my changes, you may follow the instructions..."
It seems to me that for a reader who, to this point of reading the section (and onwards), is not aware (as content may well be read from top to bottom rather commonly) that they are reading a message generated by a bot, being told rather bluntly that
"When you have finished reviewing my changes",
may appear to that reader to have been written by an author with a rather entitled personality and/or behavior, such as to assume that the reader "will" or "has to" review that authors's changes, as though the author were (feeling) entitled to the reader doing so.
It seems to me that this means running a risk of causing a casual reader to
be upset by what they may well perceive as "this kind of language and behaviour towards" [themselves and the "fellow Wikipedians"],
respond badly, such as
feeling treated condescendingly and/or
now feeling specifically disinclined to "review ... the changes"
thus producing a disservice to
the objective of having the changes reviewed by a person
peace, quiet and style on WP
"... you may follow the instructions ..."
It seems to me that this looks and feels like more of the same, and even more strongly so.
(I know the wording may sound innocent by itself, but it seems to me that it's the context that makes the difference.)
Remark: That part of the wording was not found on the page given above (seems something had been improved in the meantime) (but on a page I don't wish to link to.)
JFTR, that is for sure how I just felt when I had read that passage to that point without realizing I was reading something written by a Bot.
About followup (after a fix has been done) on older pages that still reflect the previous presence of the problem: Would it be historical misrepresentation or maybe just a nice idea to have the Bot occasionally fix (update) the wording it left there when it did, maybe in its free time :) ?
Thank you for the feedback. I'd like to note that the bot has largely stopped posting these messages, especially on English Wikipedia. Harej (talk) 21:31, 3 April 2024 (UTC)
Latest comment: 8 months ago4 comments3 people in discussion
Greetings. I read you plan to increase the limit on the single page tool. I need this to run on a page with about 800 links. When do you plan to increase the limit? SusanLesch (talk) 17:59, 20 March 2024 (UTC)
When there are non-ASCII characters in an email subject line, the entire subject should be encoded as UTF-8 so that it will display properly for the recipient. I received email from InternetArchiveBot about a submission for the Turkish Wikipedia with "Subject: Bot iÅŸiniz 18485 tamamlandı!" and about one for the Italian Wikipedia with "Subject: La tua attività di bot 16464 è stata completata!" The corresponding text in the body of the message displayed properly, with all the diacritical messages where they should be: La tua attività di bot 16464 è stata completata! I use gmail, so it's possible that gmail is doing something wrong.
Latest comment: 8 months ago5 comments3 people in discussion
The bot appears to mark https://ochem.eu/* pages as dead links. These are not dead: when I visit http://ochem.eu/article/99826, the page redirects and asks me to login, but I can login as a guest and get redirected back to the page I'm looking for. This elaborate double-redirection process may be blocking the site to crawlers and causing the false positives.
I would report this problem through the "report false positive" link, but that appears broken: it says I don't have the "reportfp" privilege, even though that should be available to all users.
Bernanke's Crossbow, usually when this happens it's because of geo-restrictions affecting our link checker. However, I visited that website with a VPN and the site would not load then either. So the website appears to at least be inaccessible to much of the Internet. Harej (talk) 21:51, 3 April 2024 (UTC)
Ah. In fact, I just discovered it's even weirder than that: until today, I've only ever visited the site in Firefox's InPrivate mode. I just tried it without InPrivate, and it fails to load then too (but works fine in InPrivate still). They must be doing something very strange with cookies.
Latest comment: 7 months ago5 comments3 people in discussion
Hello! Is there a way that InternetArchiveBot can use Moroccan Darja month names instead of English ones, on arywiki? If there's a configuration page where I can translate the month names, please let me know. Thanks! Ideophagous (talk) 09:53, 13 March 2024 (UTC)
en:User:Nemoralis - Hi sorry this appears to be my bot but it's a separate minor bot so I didn't see it when searching for this phrase. Right now this bot does not support translations. If you think it's very important I can add an if/then trap directly in the bot code just for azwiki to get around this limitation. Although I would prefer not, it could be done. -- GreenC (talk) 17:00, 15 May 2024 (UTC)
Small fixes in translatable messages
Latest comment: 7 months ago4 comments2 people in discussion
Latest comment: 7 months ago3 comments3 people in discussion
The bot is adding Wayback templates to cites which are already using cite web or cite news and filled archive-url=xxx on Chinese Wikipedia: [7]. Please fix it. Best regards, Tim Wu (talk) 07:22, 3 May 2024 (UTC)
[Bot is adding url-status=live to Cite web/cite news where offline is already set to "yes" or "1" which causes an error, because its contradictory. And the URLs are still offline so it makes literally no sense, they are not "live"]
Title changed to "HuffPost is now a part of Verizon Media"
Latest comment: 7 months ago3 comments2 people in discussion
The title of many HuffPost references has been changed to "HuffPost is now a part of Verizon Media" and the URL to https://consent.yahoo.com/v2/collectConsent?sessionId=1_cc-session_fafa9f49-54d8-4731-a897-b70175c6342b or to some other page at consent.yahoo.com. All the pages at consent.yahoo.com are now permanent dead links. Would it be possible to restore these HuffPost links to the original link, without losing the useful changes that may have been made to the pages since the link was replaced with the consent.yahoo.com link? Eastmain (talk) 15:22, 29 April 2024 (UTC)
@Eastmain: I'm not sure I'm entirely following. Could you clarify what you are requesting? If I'm understanding, HuffPost was acquired by Verizon, and they redirected their URLs to consent.yahoo.com? Or they were changed by other users? —CYBERPOWER(Chat)14:51, 15 May 2024 (UTC)
Sure. Feel free to open a pull request to make the change. If you need further help with anything just ask. I always welcome outside contributions to IABot. —CYBERPOWER(Chat)15:08, 15 May 2024 (UTC)
Latest comment: 7 months ago5 comments4 people in discussion
Today I noticed the bot has started translating parameters and parameter values in citation templates (example), inventing parameters which are undefined in our templates and invalid. Could somebody look at this? --Upwinxp (talk) 19:42, 14 May 2024 (UTC)
The bot reported in this edit that the DOI link has an error (it is obviously not true, DOI links are very stable). I thaught you night want to know about it and find the source of this mistake פעמי-עליון (talk) 19:58, 4 April 2024 (UTC)
פעמי-עליון, thank you for the report. The URL in question couldn't be found in our URL database (where links that are checked would be found), so I suppose this was a one-off situation. Please let me know if you see anything like this anywhere else. Harej (talk) 20:35, 10 April 2024 (UTC)
If they aren't open access, they should ideally be marked as such. Not only will it serve to inform readers that it's not readily accessible, but also the bot handles such cases differently and doesn't outright mark them in dead in certain situations. —CYBERPOWER(Chat)14:36, 15 May 2024 (UTC)
@Harej: do you think you can amend the redirect on Toolforge to remap old links from iabot.toolforge.org/iabot to iabot.wmcloud.org? —CYBERPOWER(Chat)14:46, 15 May 2024 (UTC)
@Vanbasten 23: IABot is reading from this page to get it's data for how to handle citation templates. You will need to flip those values around in the settings as the bot will always default to the first in the list. Once adjusted, the bot will change it's behavior automatically. —CYBERPOWER(Chat)14:57, 15 May 2024 (UTC)
Latest comment: 6 months ago5 comments3 people in discussion
Hi! Is there a way to add a flag on specific sources so they don't update the archive? There are some citations which are used to cite ongoing information, so the archives are always going to be wrong, if we could add something into the template to stop IABot archiving those specific sources that would be fantastic. Best Wishes, Lee Vilenski(talk • contribs)09:11, 2 May 2024 (UTC)
Latest comment: 6 months ago3 comments2 people in discussion
Have setup the Bot myself, but this comes:
PHP message: PHP Fatal error: Uncaught TypeError: mysqli_real_connect(): Argument #6 ($port) must be of type ?int, string given in /app/src/Core/DB.php:275Justman10000 (talk) 23:27, 4 May 2024 (UTC)
Latest comment: 6 months ago5 comments2 people in discussion
Hi! I found two translatable message that say "Sorry, but the language you have picked is not available yet." They invite the user to translate the interface at translatewiki.
Can they ever be shown in other languages? It looks like by their nature, they can only be shown in English, but maybe I'm misunderstanding something. Amir E. Aharoni (talk) 16:37, 11 May 2024 (UTC)
@Amire80:, no. They are hard-coded in English as there is no point in translating them. They will never be shown if the UI has a complete translation, and it will be irrelevant to those who can't understand English anyway as that is the base the translations come from. —CYBERPOWER(Chat)15:04, 15 May 2024 (UTC)
Great, thanks for the response. Since they can never be shown in their translated form, I'll remove them from the translation workflow. This is only a matter of configuration on translatewiki.net, and no action is needed in IABot code. Amir E. Aharoni (talk) 17:14, 15 May 2024 (UTC)
... I explored it a bit more, and I realized that languageunavailableheader and languageunavailablemessage are always shown only in English, but "incompletetranslationheader" and "incompletetranslationmessage" can be shown when the localization is incomplete, as their name implies. I'll do theconfiguration accordingly. Amir E. Aharoni (talk) 17:31, 15 May 2024 (UTC)
Latest comment: 6 months ago3 comments3 people in discussion
Hola. Aparte del ocurrente "verificar la verificabilidad", que imagino es una traducción no demasiado correcta, ¿de qué sirve enlazar en Internet Archive a digitalizaciones de libros de los que, por tener los derechos reservados, no es posible ver más que la portada y la contraportada? ¿Es para hacer posible comprobar que el libro existe? ¿Y por qué hay que comprobarlo? ¿Porque no nos fiamos del wikipedista que ha puesto la ficha bibliográfica incluido el ISBN? ¿No es suficiente para la comprobación el ISBN? Y como no nos fiamos del wikipedista hacemos perder el tiempo al lector invitándole a pinchar en un enlace que no le va a permitir comprobar nada más de lo que le dice el ISBN. ¿Es esa la idea? Saludos, --Enrique Cordero (talk) 09:53, 12 May 2024 (UTC)
Which book(s)? Normally with books at archive.org you can "search inside", view pages within the book - it is the same reason people add links to Google Books. It's useful for looking up information from the citation. -- GreenC (talk) 16:00, 15 May 2024 (UTC)
Latest comment: 6 months ago6 comments2 people in discussion
The translatable message permissionschange says:
Your {{locale}} user permissions have been changed by <a href="{{actionuserlink}}">{{actionuser}}</a>.
What is the {{locale}} parameter?
Also, the qqq documentation says "Email body with HTML and templates", but this is a bit generic. What is the email about exactly? I guess that it's about some permissions that were changed, but where and in what context? Amir E. Aharoni (talk) 20:30, 12 May 2024 (UTC)
locale describes the wiki. For example, if on the tool interface I gave you root, but only for enwiki, locale would be enwiki-English Wikipedia, or whatever the local translation of the Wikipedia name is. Does this help? —CYBERPOWER(Chat)15:10, 15 May 2024 (UTC)
Sorry, but this is an old edit. Many changes and improvements have been made to IABot since then. If this keeps happening, please re-report with a new example. —CYBERPOWER(Chat)19:01, 22 May 2024 (UTC)
Hi. I see your wiki has imported the CS1 Citation modules from enwiki. You can actually change your cite templates to use that and localize it very easily. However, this is not required. IABot now reads that Cite template and can use that knowledge to adapt its behavior to non-CS1 templates. It should behave correctly for English language templates. For the other templates, I recommend adding the local aliases to the Citation/CS1/Configuration module for the localized variants, and having them point to CS1 as well. This will make cross-wiki adaptations much easier. —CYBERPOWER(Chat)20:42, 22 May 2024 (UTC)
Your tool docs are featured as an example in the new Tool Docs Guide!
Latest comment: 5 months ago3 comments2 people in discussion
Hello InternetArchiveBot maintainers, contributors, and fans! I wanted to let you know that I highlighted the InternetArchiveBot documentation as a shining example in the new Tool Docs guide that I just published. Thank you for creating lovely tool documentation that can serve as an example to help others create and improve tool docs :-) This guide was created as part of the Doc Your Tool project for the upcoming 2024 Hackathon. If you're interested, please join that project to work on or talk about tool documentation during the hackathon! TBurmeister (WMF) (talk) 16:52, 16 April 2024 (UTC)
But archive.ph is the same service and the link with ph works fine. This is again a clear violation of the Dutch version of “if it ain't broke, don't fix it” guideline, just like the most recent time we spoke. Mondo (talk) 20:11, 1 February 2024 (UTC)
I replied in the Phab giving the technical reason why, it's done for functional reasons not cosmetic, archive.today is a special domain that is functionally more reliable then the other ones, and it's also the domain the owners of archive.today requested we use on Wikipedia as a safeguard against potential future outages. -- GreenC (talk) 14:42, 4 April 2024 (UTC)
They can request whatever they want, but at least on the Dutch Wikipedia, changes at the request of owners are seen as an unwanted change and even without their request it's seen as an unwanted change, so something still needs to be done about it. Mondo (talk) 14:57, 4 April 2024 (UTC)
MrLinkinPark333, the bot has been queued on those pages, but we can't guarantee the bot will fix those pages that are members of the maintenance category. It should fix the ones that are explicitly marked url-status=dead. Harej (talk) 20:29, 3 July 2024 (UTC)
Latest comment: 4 months ago4 comments3 people in discussion
Hi,
There's this message:
Once the search results load, which can take time, select the domains it found from the list that you want to modify and push "Submit" on the bottom of the page.
What does "that you want to modify" refer to? The domains or the list?
If it's the domains, could it perhaps be written like this:
Once the search results load, which can take time, select the domains it found and that you want to modify from the list and push "Submit" on the bottom of the page.
Latest comment: 4 months ago4 comments3 people in discussion
There's this message:
All citation templates are listed here from all wikis. Format it as if you are transcluding the template, and put new templates on a new line. Failure to follow correct formatting may break the bot.
Latest comment: 4 months ago4 comments3 people in discussion
Hello again :)
There's this message:
A non-critical DB has returned an error. This may have minor impacts on reliability.<br>Error {{errno}}: {{errormessage}}
It may be correct, but I wanted to make sure: Is it really a non-critical DB? Like, is there a critical DB and a non-critical DB?
I'm asking because the same sentence also mentions error, and it's much more common in software user interfaces that the errors are critical or non-critical, and not the DBs.
@Amire80:, the distinction is that replicas are non-critical. The main database that the UI works off is a critical database. It's a caution that the user interface might be slower because the databases that supply redundancy aren't working right. Harej (talk) 20:50, 3 July 2024 (UTC)
@Graham87: we checked the link database and can confirm the archive link is now correct, without the %20 encoded space. We weren't able to reproduce the original edit that caused the encoded space to be added to begin with, so we think this is a transient error. Please let us know if you see it happen again. Harej (talk) 21:04, 3 July 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hi and thank you for your bot. I am running it on a giant article that used to take maybe 5 minutes. Now after a couple minutes the screen goes blank and doesn't recover. I guess due to lack of patience I ended up with three different IABot edits. You need a better way to tell the user when the bot is done. P.S. Clearing my cache does not help. -SusanLesch (talk) 23:33, 7 June 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hello. On the justification "Rescuing 1 sources and tagging 0 as dead.", the bot changed a correct link to an incorrect link here: [8]. Please identify the issue that caused this so that it never happens again. I will go ahead and revert the bot's edit since it unequivocally ruined the original edit. Thanks for your hard work! Geographyinitiative (talk) 10:21, 11 June 2024 (UTC)
*{{cita libro
|apellido = Niemeyer F.
|nombre = Hans
|enlace-autor = Hans Niemeyer
|título = Hoyas hidrográficas de Chile, Tercera Región
|ubicación = Santiago de Chile
|editorial = [[Ministerio de Obras Públicas (Chile)]], Dirección General de Aguas
|año =
|edición =
|url = http://documentos.dga.cl/CUH2886v3.pdf
|ref = harv
|fechaacceso = 25 de julio de 2019
|urlarchivo = https://web.archive.org/web/20181111130729/http://documentos.dga.cl/CUH2886v3.pdf
|fechaarchivo = 11 de noviembre de 2018
}}
There are 12 such documents of Hans Niemeyer F., one for every of the (1980) Regions of Chile.
Most of the rivers, lakes, salars, geisers and many other objects of Chile have one of these documents as bibliography, I guess, tehy appear so 1500 times in the Spanish Wikipedia.
All these sources in the servers of "Ministerio de Obras Públicas (Chile)" have been deleted and archive.org also deleted these documents, I suppose on behalf of the ministry.
Now, a "DGA-0065.rar" (23.52 MB) file appeared with all these documents under
changing (working) https to insecure http and linking dead url as live
Latest comment: 4 months ago5 comments3 people in discussion
See here - the bot claims that the original link is still live (which it isn't), and changes the archive link (archived via trove, not the internet archive) from https to http. Nigel Ish (talk) 09:38, 4 July 2024 (UTC)
Latest comment: 4 months ago5 comments3 people in discussion
Hi!
Something slightly different this time.
I've completed the translation of the bot into Hebrew on translatewiki. Along the way, I sent a few more message corrections. There are now six pull requests at GitHub. It would be nice to review them.
Checked the pull requests, but some of them had merge conflicts. Please make sure nothing was lost as I merged them in. —CYBERPOWER(Chat)17:38, 1 August 2024 (UTC)
Thank you! It all looks good.
I fixed the spelling in the one message where you asked for a fix, and I added one more white space fix. Once these two are done, that will be it (at least for now). Amir E. Aharoni (talk) 19:50, 1 August 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hi everyone. I'd like to enable back the bot for the Asturian Wikipedia. According to the current bot status for astwiki, it was disabled in October 2023 with the reason: "dead link template still not working right". I checked the template (as well as other related ones), and they are mostly the same as on Spanish Wikipedia (eswiki), where the bot is indeed active. I'd like to ask some intel on what config or template needs review for the bot to be enabled back in astwiki.
Thanks in advance! YoaR (talk) 15:19, 10 June 2024 (UTC)
Internet Archive snapshots of Google Books pages: valid or not?
Latest comment: 4 months ago3 comments3 people in discussion
It is my understanding that Google Books is globally deprecated as an archive and all links to it are treated as permanently dead by the bot. However, I was able to manually find an Internet Archive snapshot of a Google books link tagged as dead. Do you encourage or discourage adding such snapshots to the bot's archive database? Huntthetroll (talk) 21:46, 14 June 2024 (UTC)
By default, the entire books.google.com domain is set "Subscription" ie. skip processing. However if you found a certain URL that works, you can log into iabot.org and set the archive URL for that one URL, and change it to "Dead" status (or "Live" depending what it is). GreenC (talk) 20:58, 28 July 2024 (UTC)
The bot is correctly identifying them as alive in the article. There was a transient issue a while ago that caused this behavior, which should long be fixed. —CYBERPOWER(Chat)17:57, 1 August 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
The bot made this edit in ukwiki. It cleared the content of a tag "<ref group="к" name=":0">" that wasn't redundant, which caused an error (FAQ claims that "The bot often makes maintenance edits to articles in the course of its work. This includes removing redundant citations from articles."). I assume it's because there's a same name tag that is in another group. MonX94 (talk) 11:49, 23 June 2024 (UTC)
Latest comment: 4 months ago4 comments2 people in discussion
In this edit the bot caused harv/sfn no-target errors for "U.S. Department of State / Office of the Historian (1950b)", and "Mitterrand (1990)", and a multiple-target error for "Wörner (1991)". I have undone it, but would appreciate it if the bot were taught not to screw up references. I can be contacted on en-wiki. DuncanHill (talk) 12:12, 28 June 2024 (UTC)
There's something amiss with those references. They both reference identical content but have different attributions and authorship. The bot simply merged the two identical cites. I would take a closer look at the references. —CYBERPOWER(Chat)18:00, 1 August 2024 (UTC)
Please could InternetArchiveBot be enabled on these wikis? It already has bot user rights on all three, from previously guarding against link rot, (but with moving to WikiTide and back something got turned off or broken). Rob Kam (talk) 22:43, 3 July 2024 (UTC)
It historically was having issues with OAuth with valid keys producing invalid responses. But it's fixed now and working again.—CYBERPOWER(Chat)12:09, 13 August 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hi,
I tried to use Manage individual URLs and it returned "http error 403, Access to iabot.wmcloud.org was denied, You don't have authorisation to view this page." Can you fix that? Anoop Bhatia (talk) 04:18, 17 July 2024 (UTC)
That doesn't sound like what you got was actually IABot denying you access but something on wmcloud itself. If it's still happening, let me know, but from what I see, you should have access to that tool. —CYBERPOWER(Chat)18:39, 1 August 2024 (UTC)
Bot reports working links and doesn't provide error messages
Latest comment: 4 months ago3 comments2 people in discussion
On the Dutch Wikipedia, the bot is adding talk pages at high speed with 'false positives', i.e. reporting of links that aren't actually broken. Furthermore, no error message is given. See for instance this talk page - but the same holds for (nearly) all recently created talk pages.
Those are messages stating that the links failed to get saved into the Wayback Machine. Someone had apparently activated the bot's feature that leaves those messages behind to make users aware that the URLs on the article can't be captured. —CYBERPOWER(Chat)18:51, 1 August 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Please don't do this (first link) where the web archive template went within the other link's display text. It causes a Wiki-link in External-link error, and is a nearly eradicated tracked syntax error on enwiki. Moving Webarchive outside like this is error free and preferred. Thank you. Zinnober9 (talk) 20:55, 17 July 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hello - I'm trying to access the Modify URL tool, but it appears that I don't have sufficient permissions. How can these permissions be obtained? Thanks! Breaktheicees (talk) 22:00, 19 July 2024 (UTC)
Normally this accessed by experienced users only. Users with at least 3 months of experience and 1000 edits are automatically given access.
IABot generally stalls out in some cases when it needs to communicate with the Wayback Machine and there are service issue preventing such. IABot will usually recover on it's own in those case. As of today, the queue is current with the oldest job being yesterday. —CYBERPOWER(Chat)19:15, 1 August 2024 (UTC)
Formats. Because IABot just can't assume the format by looking at it, it needs a list of possible formats the wiki uses to check against. If none of them matches, it defers to the designated default format, in most cases being ISO. —CYBERPOWER(Chat)19:16, 1 August 2024 (UTC)
High load warning with 504 error and nothing processed
Latest comment: 4 months ago7 comments4 people in discussion
It has been showing High load warning since few days ago. As of this comment, it's stating that "the current estimated lag is 1843 minute(s) and 59 second(s)" and has been constantly increasing minutes by minutes. Running a single page on "Analyze a page" doesn't nothing, regardless of page sizing, it would just throw 504 error after some time and unlike previously where 504 error was thrown and the article was actually archived successfully, this time round nothing really happens even waiting after a day or two. Kindly look into this issue. Thanks and regards, Paper9oll (talk) 16:34, 24 July 2024 (UTC)
This has been the case since at least July 19. The stated time is currently reduced to "387 minute(s) and 19 second(s)", but that's still more than 6 hours. However, a single-page archival I started on July 19 has either been cancelled or is still pending six days later. Is there an issue that blew up the queue? Can the queue be manually cleared once to resume normal operations? IceWelder (talk) 17:13, 25 July 2024 (UTC)
It seems to be getting worse. I've tried to run it on a page since yesterday and the lag counter just keeps counting up.
Service issues with the Wayback Machine will cause requests to back up and stall out. IABot usually recovers on its own as it has been these past few weeks. However, when a sufficient number of requests do indeed back up, it will eventually just bring the whole website down. Nothing that can be done but wait for the service to recover and let the bot catch back up. —CYBERPOWER(Chat)19:18, 1 August 2024 (UTC)
@Cyberpower678 Thanks for the follow up. I also noticed that, as of this comment, the warning is showing "estimated lag is 754 minute(s) and 11 second(s)". However, the request was completed and published without any noticeable delay. I was expecting the publishing to occur after "754 minute(s) and 11 second(s)". I'm a bit confused—was there actually a lag, or did I misinterpret the warning message? Paper9oll (talk) 12:54, 2 August 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
A good deal of such links were attributed as dead (like this). Now all of them replaced by links to riamediabank.ru [9]. Will the bot remove those 'deadlink' templates automatically or is a special launch required? Romano1981 (talk) 11:16, 28 July 2024 (UTC)
IABot is not configured to remove dead links. Their very presence will actually make the bot think the links are dead and you can expect that it will try to add archive URLs. However a new version of IABot is in the works where it can remove dead link tags with much greater reliability than now. —CYBERPOWER(Chat)19:19, 1 August 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hi. Its an old edit, I dont know if anyone alerted you about it or not but its not right, it makes size of image parameter not work. Compare with previous version of article. Bikar (talk) 15:15, 31 July 2024 (UTC)
Latest comment: 4 months ago3 comments2 people in discussion
Hi all, I ran IABot on Carbonyl iron against some dead links, hoping it would find something on Wayback. The busy message was displayed as discussed above. I left it running, forgot about it and fixed the dead links myself. With this edit the bot then deleted my changes (which I reverted), and has continued to make changes although I haven't asked it to. I can't imagine anything more pointless than an edit war with a bot, so please, how do you ask a bot to leave a page alone? Cheers, MinorProphet (talk) 09:34, 5 August 2024 (UTC)
With the ongoing backlog, the lag time estimated on the warning is how long it is expected to take before it can complete the request. You ran it twice it seems, and it finally caught up to your request. It wasn't really edit warring with you, but simply finishing what you asked it to do. I'm working on the delays.—CYBERPOWER(Chat)12:47, 13 August 2024 (UTC)
Latest comment: 4 months ago5 comments2 people in discussion
I ran IABot on w:Minneapolis when your bot had a huge backlog, maybe more than 900 minutes, maybe it said more than 1000. I came back much later and ran it again when there was no backlog, result was no change to the page. Tonight I looked at the article history -- your bot had made a number of edits it had no business making, confusing the entire day's history. I am at my wits end trying to figure out what happened. But all signs point to IABot? -SusanLesch (talk) 02:06, 14 August 2024 (UTC)
It looks like the 4 edits prior to the IABot entry were all reverted by your bot. Help!
Not exactly reverted. What happened was the revision live at the time you submitted the request was used and processed. Once the backlog finally cleared, it was able to finish processing your request and pushed it live. All you need to do is revert the bot edit and run it again. The backlog issue should be fixed now. —CYBERPOWER(Chat)07:37, 14 August 2024 (UTC)
Not that simple. This happened in the middle of continuous editing and I have manually redone everything your bot undid. Anyway glad to know it is fixed. Thank you. -SusanLesch (talk) 12:46, 14 August 2024 (UTC)
Latest comment: 2 months ago6 comments3 people in discussion
Hey there, I've tried to run the bot several times on en:Howl's Moving Castle (film), but it never ends up adding any archives and simply ends the attempt after a few seconds. I'm not sure what the cause of this is, and I'd appreciate any help. Let me know if you have any questions! —TechnoSquirrel69 (sigh) 18:56, 19 June 2024 (UTC)
It runs fine for me. Bear in mind archive URLs are generally only added when the bot declares a link as dead which means it would have needed o fail 3 consecutive checks each spaced 3 days apart. Archive URLs don't show up immediately after a link goes down. —CYBERPOWER(Chat)17:54, 1 August 2024 (UTC)
@Cyberpower678: This seems to happen even when the option to archive non-dead links is enabled. If it helps at all, here's a screenshot of the results when I tried to run it on the page again.
Latest comment: 2 months ago3 comments2 people in discussion
Hi there, This bot adds all the right properties to the references on the page; there is only one small error: for the url status (live or dead), instead of 'dodeurl=nee', it says 'dodeurl=no '. So: 'no' should be 'nee', and there is no need for an added space. I don't seem to have the rights to change this myself. :-) Thank you in advance, Laurier (talk) 10:33, 12 August 2024 (UTC)
Very interesting, a no is creeping in there. IABot tries to be backwards compatible with older versions of CS1 modules, but I do recommend eventually updating to the latest version enwiki employs. It's much more flexible and IABot works better with it. I'll take a look into the issue when I get a moment. —CYBERPOWER(Chat)12:56, 13 August 2024 (UTC)
In order to get the bot to run on your wiki, you will need to install the OAuth extension into your MediaWiki installation. I will create the respective accounts after and IABot will need a bot flag, and I will need permission to use the OAuth Consumer Registration tool. —CYBERPOWER(Chat)13:01, 13 August 2024 (UTC)
Permission error
The action you are trying to perform requires the definearchivetemplates permission.
This permission is obtainable with the following groups: root
What templates are you trying to add? Most used templates are already defined and need only be applied in the local configuration for the wiki itself. —CYBERPOWER(Chat)09:18, 15 August 2024 (UTC)
Whatever it is to get similar to en.Wikipedia on repaired links. For example, archived from the original (PDF) on date, retrieved date. I've now imported Template:Webarchive to the wikis. Maybe that's all that was required? --Rob Kam (talk) 09:51, 15 August 2024 (UTC)
I guess I should make you admin of your wikis. Interesting you don't seem to be a sysop on your own wikis. Is that a Miraheze policy thing? —CYBERPOWER(Chat)12:37, 15 August 2024 (UTC)
It shouldn't be a problem. On my wikis I'm "Member of: Automoderated users, Bureaucrats, Interface administrators, Administrators". Rob Kam (talk) 15:29, 15 August 2024 (UTC)
Bot replaces templated urls in citations incorrectly
Latest comment: 2 months ago4 comments2 people in discussion
If a citation URL is the output of a template, the bot tries to convert the template invocation into a URL which results, of course, in an incorrect URL. An example of this behaviour is here, which I have undone. Could the bot be programmed to avoid doing this sort of change? Thanks. --RobertG (talk) 09:30, 15 August 2024 (UTC)
Hmmm, the bot should know better, and should know how to handle these cases. I'll make sure the next major update of IABot will have this working again. —CYBERPOWER(Chat)09:33, 15 August 2024 (UTC)
Check to make sure you have the tool set to the correct wiki. You can find the wiki you are actively on in the dropdown on the top right of the page.—CYBERPOWER(Chat)06:42, 16 August 2024 (UTC)
Done—CYBERPOWER (Chat) 17:46, 1 August 2024 (UTC)
Fixed—CYBERPOWER (Chat) 18:35, 1 August 2024 (UTC)
