[01:53:24] I submitted my wiki request 3 days ago. At how many days should I contact support [01:55:43] If it were me, I'd wait about a week, but as for official Miraheze policy, I'm unsure [02:05:16] Which one is it? I may be able to take a look at it if I haven't already [02:05:52] deaddrops.miraheze.org [02:06:20] looks like #32588 [02:07:13] It is, taking a look rn [02:09:10] The one thing that I'm concerned about in the request is that the website associated with it contains real addresses [02:13:09] So the USBs are implanted on, like the side of the road or under a bridge or a in a bench. [02:13:31] It's like geocaching if you've heard of that but with usbs [02:14:02] I do acknowledge it is saucy so that's really up to the admin team whether to permit such a wiki [02:15:04] the real addresses are these walls and public structures, not a hit list of peoples houses lol [02:16:10] Under UK law, geocaching is legal as long as you have the proper permission to do so; I assume the same applies to sticking USBs in walls [02:16:43] Another issue that results from this is that there would be no way to tell if a USB featured on the wiki is legal or not [02:17:47] Would a wiki on locations of graffiti be illegal? [02:19:07] Hmm...I'm actually not sure on whether to approve your wiki or not [02:19:15] I'll ask another wiki creator about it [02:32:49] That would have fewer issues, potentially [02:34:04] I'm assuming USB contents are unknown to site hosts? So there is a risk that the USBs themselves could contain illegal material? [02:42:08] The database doesn't show the contents of the usbs. So they can be anything. Couldn't a geocache also have illegal material? [02:42:22] Also true [02:43:51] UK law is the sticking point here, our resident expert and treasurer (among many titles) @Owen may have some insights when they are online. [02:58:42] don't think so [03:02:05] What's going on with the database rn? [03:04:34] I can't find anything about scheduled maintenance but when I try to edit an article it says an admin locked the database to read-only mode [03:16:48] AFAIK generally means it crashed, it automatically sets it to read-only when it reboots [03:45:47] That makes sense [03:46:06] So it’s just a matter of time then or does someone has to fix someone on our wiki specifically [03:48:34] Yeah I'm getting a "Database locked" error [04:00:39] @Site Reliability Engineers need to unlock the database manually, so would help them if you tell what wiki it is at least [04:04:15] same here [04:05:17] wiki is in my nickname [04:08:12] Is there a way to know which wikis undergo unannounced maintenance? [04:15:20] ahh [04:15:26] KS3 ICT Wiki here [04:23:51] none of the SREs are here [04:25:04] Well, Void is here [04:40:15] there is no way to not let database lock itself after reboot? [04:45:45] Not really, usually it is due to the database going OOM (Out of memory) [04:46:20] I believe the read-only on restart is done for safety, to avoid corruption etc. [04:47:38] fair [05:00:50] Safe failure [05:44:15] Unlocked [05:44:38] yay! [05:44:40] Thank you [05:55:59] Pretty much yes [10:59:23] this doesn’t seem to apply to mobile for some reason, do you know why? [11:03:42] if you have a mobile setup that uses another skin then the element itself may not be the same [11:20:55] goodbye [11:34:08] Hello [11:34:54] Now I have the wiki [11:35:04] I would get that role with the users checked [11:35:12] That role in green [11:38:07] its just the default web one if thats what you mean [11:44:07] [1/2] type .auth for the bot to link you Discord and Miraheze accounts together [11:44:07] [2/2] it might don't work at first and needs several tries [12:03:21] .auth [12:22:43] Was the database maintence disclosed ahead of time? Because I don't remember it being announced. [12:25:43] Just read above, seems it was unplanned, understood. [12:27:58] what does one have to do with a user who says they're younger than 13? [12:28:51] pretty sure there was a blurb on it in one of Meta's policies [12:29:40] /auth [12:30:32] an unannounced maintenence is basically an automatic database lock when the server restarts, which is usually after running out of memory [12:31:05] Yeah that's what I figured. Do I just wait for it to fix or? [12:32:45] head to https://reports.miraheze.org/, use the + on the upper right, send a data protection report [12:33:29] is it ongoing? it is typically fixed when sre is aware and able to lift the lock, if it's in progress sre needs a ping [12:37:38] Don't know exactly what you mean when you say ongoing, but I haven't been able to save an edit this morning. [12:37:57] and you are still getting a database lock error? if so, please link the wiki [12:38:08] actually even if not just to have a look [12:38:44] [1/4] [12:38:44] [2/4] > Warning: The database has been locked for maintenance, so you will not be able to publish your edits right now. You may wish to copy and paste your text into a text file and save it for later. [12:38:44] [3/4] > [12:38:45] [4/4] > The system administrator who locked it offered this explanation: The primary database server is running in read-only mode. [12:40:35] cc @Site Reliability Engineers to check out ^ [12:40:49] from here you can only wait I'm afraid [12:41:06] Understood, thank you for your help. [12:42:29] np, apologies for the inconvenience [12:46:22] fixed [12:47:52] Thank you! [15:04:43] hi! [15:21:45] @Reception123 [[Special:RequestImportDumpQueue/370]] is not actually complete [15:21:45] https://meta.miraheze.org/wiki/Special:RequestImportDumpQueue/370 [15:21:46] [15:24:16] https://sebbog.miraheze.org/wiki/Special:AllPages?from=Infobox&to=&namespace=10 [15:55:20] My new wiki has been approved, but I don't seem to be able to make any edits. Am I missing something obvious? (I'm logged in.) [15:57:49] probably missing bureaucrat rights due to a bug, what't your wiki? [15:58:01] ^ @raidarr [15:58:30] also, you can't edit at all or you can't access wiki settings? [15:58:47] need to know the wiki to see [15:59:54] no need to post the url itself, just the subdomain would do [16:01:34] [1/2] Will eventually contain adult content for educational purposes [16:01:34] [2/2] ||HypnoKink|| [16:02:43] I was able to create a "talk" page from the default starterpage, but I can't edit or create any other pages [16:08:08] right after creation? haven't done anythong w/ settings yet? [16:09:01] I haven't. [16:10:08] ? [16:10:10] When nothing was working, I tried to turn on advanced mode, but it turned itself back off again 🤷‍♂️ [16:12:22] nothing is stopping users from making edits on the wiki at the wiki level at least, I'm inclined to think something local is causing an issue [16:12:47] permission assignments are correct and I'm not really sure what 'advanced mode' is [16:13:02] not a standard part of mw functionality that I know of [16:13:35] clearing your miraheze cache data may be of use, otherwise I'm not sure what it would be [16:14:58] No need to bother the stewards, your request will be fulfilled in 1-3 working days, do not rush. [16:15:59] [1/2] Thanks. I'll try that. [16:15:59] [2/2] There's not some obvious step I'm missing for creating the first page, or editing the default Main Page is there? It should be as straightforward as any other wiki site? [16:16:48] the main caveat is that a miraheze wiki is a near-default mediawiki installation and things like visualeditor or special buttons to set up pages need to be specifically enabled, otherwise it should be perfectly normal especially to edit [16:17:36] Something suddenly changed and I can edit! Perhaps the whole page wasn't loading or something? I'm good now! [16:18:40] I'd say self correcting problems are the best but it makes one wonder how they happened in the first place :p [16:18:48] @raidarr [16:19:02] are you available, i want to ask a question [16:19:43] ask anytime, I'm usually in the vicinity for most of the day [16:19:49] today at least [16:21:10] can you give a little description about importing or exporting the page? [16:21:46] I don't have much knowledge about import and export, if you give a definition it will help to improve my knowledge 🙂 [16:22:19] Special:Export -> puts desired pages into an XML file; Special:Import -> puts contents of the xml file into pages on the wiki [16:22:55] Export usually works fine but import can time out with anything big, so a large import may require help through Meta's Special:RequestImportDump [16:23:37] hmm okey, thank you very much raidarr 🙂 [16:23:53] thank you help (+add) lol 😄 [16:23:54] np [16:24:10] ow yeaaah no problem [16:24:12] hahaha [16:24:19] 🤣 [16:24:33] [1/2] you write down a category to get all those pages into an xml file which includes the last 1 thousand revisions or the latest revision and what user made that change or you could write down each page manually to put into the xml [16:24:34] [2/2] to import you just upload the xml with the pages inside and you write down the interwiki prefix of the wiki you exported from [16:24:56] anyway this is back to chat, let's be serious lol [16:26:07] thank you help very much [16:26:53] the last 1 thousand revisions or the current is optional with this tick [16:27:29] 👍🕋 [16:27:56] Ay, I put the Kaaba by mistake. sorry [16:28:35] Include templates includes all templates present inside of the pages so lets say you write down: Page1 and Page1 uses Template:1 then you export both into the xml [16:29:05] 👍 [16:29:48] Not sure on context here but you typically want all revisions for attribution purposes... [16:30:26] miraheze would not be able to handle the full history [16:30:52] Import or export? [16:31:02] first time hearing about last revion thing tbh, on template import [16:31:37] importing the full history of a page [16:31:38] I misread, my bad [16:32:08] Yes it can, and it must or you can unintentionally violate copyright if the license requires it. [16:32:37] you use that instead. [16:33:19] i imported a 7mb file and it says complete but it did not import [16:33:54] Special:RequestImportDump exists for this purpose iirc [16:34:05] larger imports do not play well but they can still be done in that way [16:34:10] [[Special:RequestImportDumpQueue/370]] [16:34:10] https://meta.miraheze.org/wiki/Special:RequestImportDumpQueue/370 [16:34:10] [16:34:23] says complete but not [16:34:32] its not complete [16:34:36] then that is something to bring to @Reception123 for troubleshooting [16:34:42] an unintended error of the system [16:35:20] 2 days ago [16:35:21] he's been out a bit lately as have I, so I do apologize for any delays on the matter [16:35:53] well, really, all the stewards and a portion of sre have, been a slow/awkward time [16:36:33] unfortunately I am in no position to address importdump errors but that is definately unusual and not the rule [16:37:09] what rule? [16:37:21] imports for the most part work fine [16:37:31] I am not sure why they would not in this case [16:38:03] Up to 250mb imports have been done with RequestImportDump [16:38:17] well a 7mb one did not work [16:38:28] so again, a bug in handling that case specifically [16:38:42] unfortunately miraheze is very very stretched in troubleshooting errors at the moment [16:39:37] @Hey Türkiye just coming back to this, I recommend using the full history for this reason. Again not 100% sure on context though. [16:41:44] okay [16:43:55] hoping this 30mb+ xml works then [16:44:06] What are you not 100% sure about? I don't understand. [16:44:12] now its 40mb+ [16:44:24] 40+mb of what exactly [16:44:28] importing all of Category:Date mathematics templates from wikipedia [16:44:33] oh dear [16:44:54] 80mb+ now [16:45:13] you said 250mb+ works so this should work i guess [16:45:14] one has to wonder about the need for all that [16:45:44] no need for 1 thousand revisions in my opinion [16:46:09] my wonder is why all that much template data is required [16:46:17] sounds like more wikipedia dependency bloat [16:46:55] which corresponds with one of the worst things to troubleshoot because wikipedia's bloat is eldrich and poorly understood by most [16:47:04] okay [16:47:11] wont use them then i guess [16:47:18] no clue what you are talking about [16:47:20] it's your perogative if you feel you need them, just my opinion [16:47:31] I do not make that call nor would I wish to [16:48:14] revision history is desired in any case because it is a simple and complete way to provide the attribution required by the liscence of wiki content including wikipedia [16:49:36] it is technically acceptable to link to the original source, but that mechanism is flawed, potentially breakable and in terms of steps needed to complete usually more difficult than the built-in attribution of import/export [16:51:41] What you are trying to import or do. Most require the full revision for license attribution purposes. [16:52:41] iirc doesn't wikipedia settle with a linkoff, or am I mistaken? [16:53:12] again, extra steps to make that happen consistently and not a good replacement for full attribution, but legal 'enough' [16:54:25] i really dont understand [16:54:40] "most require full revision"? [16:57:34] Think of it this way -- If you had a bunch of authors on a paper and the license for using the paper requires you to cite all of them, that's what including all revisions accomplishes. [16:58:00] Miraheze Trust & Safety have held that in most cases all revisions must be exported to be compatible with Wikimedia's licensing [17:09:12] Technically that is correct if you copy and paste, you can link it in the edit summary, but for imports you could have a permanent attribution template on the page, but that also doesn't look as nice... [17:14:20] i do link to the most recent revision [17:16:16] just if importing worked and was faster than one day [17:17:32] [[Special:RequestImportDumpQueue/370]] [17:17:32] https://meta.miraheze.org/wiki/Special:RequestImportDumpQueue/370 [17:17:32] [17:18:06] Import in its current concept will never work for large stuff [17:18:16] Because it's a runtime process [17:18:27] ? [17:18:49] You'll never unless we overhaul import be able to do it from the web for anything large [17:19:12] It just won't work with it processing everything at runtime [17:19:22] okay [17:19:57] i know someone who cant import over 8 megabytes (not miraheze) and one template is 20mb+? [17:20:04] with full revision [17:20:29] Ye not that big of a shock [17:22:16] will use current revision so i dont have to get the list again for wikis without anything like Special:RequestImportDump [17:23:21] also no one removed the "Complete" status for the import [17:23:45] like i get that you cant probably import the file i sent but not even remove the status? [17:25:15] you guys probably just click import or whatever and say its complete [17:28:58] Nothing technical is a simple as one button [18:11:16] yeah it takes 2 days to remove a status [18:12:51] Let's not have the sarcasm [18:45:00] Will redo now. [18:53:36] Please message deleted 2 administrator? [18:53:45] NotAracham stop please [18:53:48] @Hey Türkiye, please use the #bots channel for bot request stuff. [18:53:51] okey okey [18:54:45] shit... refuse channel use [18:57:59] @NotAracham What do you think about inactivity on wikis (i.e. dead wiki, no one is editing that wiki, only 1 user)? [18:58:15] "Question" [19:01:32] I'm not quite sure what you're asking. I don't have an opinion on that beyond the [[Dormancy Policy]] [19:01:32] https://meta.miraheze.org/wiki/Dormancy_Policy [19:01:32] [19:26:21] How long on average does it take for an email to be replied to? [19:30:52] Who you email [20:06:56] For those interested, as a temporary measure db121 will no longer go read only after restart so there won't be any more periods of no editing [21:13:05] @Reception123 "temporary measure" [21:19:43] Don't think it's temporary :) [21:32:11] Temporary, until the heat death of the universe? 😛 [21:33:56] Temporary until we decide otherwise [22:10:47] [1/3] That doesn't fix the underlying issue of it OOMing. It should be temporary as a measure to make it not readonly until it actually can be worked on to improve things to not OOM in the first place. It isn't the best to ignore an issue by making it just not visible. Something causes an OOM, and it should really be investigated as to what does as the "pe [22:10:48] [2/3] rmanent" solution to avoid it. Even if it won't be readonly on start again, I still think "temporary" should be accurate until the underlying issue can be fixed or at least definitively identified. I understand it isn't easy and not saying you should, just from my pov, taking measures just to ignore the issue could only potentially cause more issue [22:10:48] [3/3] s than it solves if there is a real issue other than just the size of the database causing it in the first place that is. That's how I've always thought of things anyway. [22:26:15] Two things: First, I know. Second, read only on start is a non-typical non-default behavior. From my point of view and understanding of mariadb, it doesn't make a whole lot of sense out side of replication servers. [22:27:51] That's fair, and yeah, it doesn't make much sense outside of replication servers, as for that would cause a lot more issues, here it wouldn't cause any issues to have it out of read only. [22:29:40] Really need to study tuning mysql at some point. Currently got an idea on how to recover thumbs listings in swift though, so that's exciting