[00:06:05] gute: you could copy the wmf config [00:10:14] gute: https://github.com/wikimedia/operations-puppet/search?utf8=%E2%9C%93&q=SSL_ciphersuite&type=Code [00:26:02] jeremyb oh but I do not use apache [00:26:17] huh? [00:26:46] hold om [00:26:48] on [00:28:23] gute: so read all of both pages of the results I linked and cherry pick out the nginx bits?? [00:28:53] we don't use apache much either. mostly nginx [00:34:30] jeremyb Thank you, though my head is spinning now. do you really need nginx? cant i get it to work anyway? [00:38:14] > OuKB: I'm having trouble finding info on the subject. Is this right? http://nginx.com/resources/admin-guide/nginx-ssl-termination/ [00:38:48] I thought you were already intending to use nginx… [00:39:10] yes, but it seems so troublesome now. [00:39:22] it's not [00:39:35] are you sure? [00:39:35] if you want parsoid specific answers I can't give them [00:39:39] yes [00:39:54] I run nginx myself [00:40:35] well ok, how long do you reccon it would take to set up? [00:42:41] me? 5 mins. you may need to read some docs that I can skip :) [00:43:26] I did give you our conf to get started with though. (see protoproxy in particular) [00:46:33] well, to tell the truth I don't even know what I am looking at. so much stuff that's new to me. Protoproxy? Puppet? Well, It feels like I took water over my head. [00:50:50] ok, well I have to run. maybe ttyl [00:55:01] ok. thank you for everything! [01:14:06] gute: nginx is indeed very easy to set up [01:14:54] it also supports spdy, and of course you can also use it to host other content [01:15:41] that said, you could also modify the code slightly to support https right in node -- for that see http://stackoverflow.com/questions/5998694/how-to-create-an-https-server-in-node-js and http://nodejs.org/api/https.html [01:16:39] you'll have to modify api/ParsoidService.js, near 'listen' at the end [01:26:20] well I have some wikis that are private and some that are not, all of which I'd like to use visualeditor [01:27:02] that's fine [01:27:17] Parsoid itself just forwards the cookie (if supplied) to MW [01:27:33] so if the user is not authenticated, Parsoid won't see any of the content either [01:28:10] are you currently using Apache for MW? [01:29:34] ok. on my synology nas (on which I have my wikis) there seems to already be a nginx installed. I have never used it, so if there is a quicker and easier way, I'd be glad. [01:32:13] I wish there was a tutorial on this. [01:38:49] its four in the morning here. got to get some sleep. ttyl [01:43:55] gute: it's a few lines only [01:43:55] gute: http://nginx.com/resources/admin-guide/reverse-proxy/ [01:43:55] and http://nginx.org/en/docs/http/configuring_https_servers.html [01:43:55] but good night in any case! [01:43:55] please, could you then help me. I would go to sleep much happier if I got this working. [01:43:55] gute: did the init script work btw? [01:43:55] no. I am waiting for a response from synology. [01:43:56] oha [01:43:56] I fear that they might be less than responsive about tinkering [01:44:34] gute: so the first thing I'd set up is the proxy [01:44:50] you can then add https in a second step [01:44:50] ok [01:44:56] see the first link [01:45:16] this? http://stackoverflow.com/questions/5998694/how-to-create-an-https-server-in-node-js [01:45:31] no, http://nginx.com/resources/admin-g [01:45:32] uide/reverse-proxy/ [01:45:45] sorry for the line break [01:45:59] http://nginx.com/resources/admin-guide/reverse-proxy/ [01:46:51] ok. what file do i edit? [01:47:32] on debian it's /etc/nginx/sites-enabled/default [01:47:39] not sure about openwrt [01:48:02] find -name nginx* [01:48:02] ./etc/logrotate.d/nginx [01:48:03] ./etc/init/nginx.conf [01:48:03] ./etc/nginx [01:48:03] ./etc/nginx/nginx.conf [01:48:03] ./run/nginx.pid [01:48:07] ./usr/share/nginx [01:48:09] ./usr/bin/nginx [01:48:11] ./var/lib/nginx [01:48:13] ./var/log/nginx [01:48:15] ./volume1/@optware/etc/nginx [01:48:17] ./volume1/@optware/var/nginx [01:48:19] ./volume1/@optware/share/nginx [01:48:21] ./volume1/@optware/share/www/nginx [01:48:23] ./etc.defaults/logrotate.d/nginx [01:48:25] ./etc.defaults/init/nginx.conf [01:48:27] ./etc.defaults/nginx [01:48:29] ./etc.defaults/nginx/nginx.conf [01:48:31] ./var.defaults/lib/nginx [01:48:33] ./var.defaults/log/nginx [01:48:37] ./opt/etc/nginx [01:48:39] ./opt/var/nginx [01:48:41] ./opt/share/nginx [01:48:43] ./opt/share/www/nginx [01:48:45] ./volumeUSB2/usbshare2-2/tardis/rootbackup/run/nginx.pid [01:48:46] hey, do you pipe the find straight into IRC? [01:48:47] ./volumeUSB2/usbshare2-2/tardis/rootbackup/var/log/nginx [01:48:49] ./volumeUSB2/usbshare2-2/tardis/rootbackup/var/lib/nginx [01:48:51] ./volumeUSB2/usbshare2-2/tardis/rootbackup/usr/bin/nginx [01:49:09] better to use a pastebin [01:49:11] no copy paste [01:49:18] (next time) [01:49:26] like gist.github.com [01:49:43] ls [01:49:47] gute: I'd try /etc/nginx/nginx.conf [01:50:53] ok. have it opened [01:52:11] see if there's a server {} section in there [01:52:22] yep [01:52:22] with a listen line etc [01:52:29] okay [01:52:29] yep [01:52:46] server { [01:52:46] listen localhost:412; [01:52:46] server_name _; [01:52:46] location ~ \.php$ { [01:52:46] fastcgi_index index.php; [01:52:47] fastcgi_pass unix:/run/php-fpm/php-fpm.sock; [01:52:49] fastcgi_param SCRIPT_FILENAME $fastcgi_script_name; [01:52:51] include fastcgi_params; [01:52:53] } [01:52:55] location ~ /\.ht { [01:52:57] deny all; [01:52:59] } [01:53:23] now add something like http://nginx.com/resources/admin-guide/reverse-proxy/ [01:53:31] this pasting is probably dangerous as hell, but right now I just want it to work [01:54:00] basically another location block, with a proxy_pass line [01:54:23] location /some/path/ { [01:54:23] proxy_pass http://www.example.com/link/; [01:54:23] } [01:54:38] location /parsoid { proxy_pass http://127.0.0.1:8000/; } [01:54:48] something like that [01:55:00] assuming parsoid is listening on port 8000 [01:55:28] then restart nginx with /etc/init.d/nginx restart [01:55:38] and test it by requesting /parsoid [01:55:53] actually, try /parsoid/ [01:56:07] with the trailing slash, in both the config & the request [01:57:40] ah. so the /parsoid was not the actual location to parsoid. [01:58:34] it's the path that you'll expose it as through nginx [01:59:20] -ash: /etc/init.d/nginx: not found [01:59:36] ok [01:59:39] ah, /etc/init/nginx [02:00:01] /etc/init/nginx restart [02:00:01] -ash: /etc/init/nginx: not found [02:01:49] nginx -h [02:01:51] nginx version: nginx/1.4.4 [02:01:52] Usage: nginx [-?hvVtq] [-s signal] [-c filename] [-p prefix] [-g directives] [02:01:52] Options: [02:01:52] -?,-h : this help [02:01:52] -v : show version and exit [02:01:52] -V : show version and configure options then exit [02:01:54] -t : test configuration and exit [02:01:56] -q : suppress non-error messages during configuration testing [02:01:58] -s signal : send signal to a master process: stop, quit, reopen, reload [02:02:00] -p prefix : set prefix path (default: /etc/nginx/) [02:02:02] -c filename : set configuration file (default: /etc/nginx/nginx.conf) [02:02:04] -g directives : set global directives out of configuration file [02:02:13] maybe /etc/init/nginx.conf ? [02:02:19] that was in your listing [02:02:47] gute do NOT paste [02:02:59] k thx [02:03:21] how about nginx -s reload [02:03:42] c: i know. [02:05:09] nginx -s stop [02:05:15] nginx -s reopen [02:07:46] damnit. I just restart. [02:10:00] that should work too [02:12:18] so then I just try http://some.dns/parsoid/ ? [02:13:54] gute: yes [02:14:01] provided that your parsoid is running [02:14:15] you might have to start it manually if that's not working yet [02:15:15] no. did not work [02:16:54] but http://some.dns:8000/ does show the parsoid page? [02:18:52] yep [02:20:03] and your location line has /parsoid/ ? [02:20:12] if so, check the logs [02:20:22] /var/log/nginx [02:22:39] 2014/09/09 04:09:40 [emerg] 3562#0: still could not bind() [02:22:39] 2014/09/09 05:04:39 [notice] 25651#0: signal process started [02:22:39] 2014/09/09 05:04:39 [notice] 25653#0: signal process started [02:22:39] 2014/09/09 05:04:39 [error] 25653#0: open() "/run/nginx.pid" failed (2: No such$ [02:22:40] 2014/09/09 05:04:57 [notice] 25853#0: signal process started [02:22:41] 2014/09/09 05:08:10 [notice] 27291#0: signal process started [02:22:43] 2014/09/09 05:08:10 [error] 27291#0: open() "/run/nginx.pid" failed (2: No such$ [02:22:45] 2014/09/09 05:11:14 [emerg] 5718#0: bind() to 127.0.0.1:412 failed (99: Cannot $ [02:22:47] 2014/09/09 05:11:14 [notice] 5761#0: signal process started [02:22:49] 2014/09/09 05:11:14 [error] 5761#0: open() "/run/nginx.pid" failed (2: No such $ [02:22:51] 2014/09/09 05:17:50 [notice] 25051#0: signal process started [02:23:24] gute: use pastebin, please! [02:23:29] !pastebin [02:23:29] To avoid overflowing the channel with inane amounts of text, use https://dpaste.org/ or other awesome pastebin sites to share code, errors, and other large texts. [02:24:23] ok. sorry [02:24:32] gute: can you see anything from nginx at all? [02:24:41] http://some.dns/ for example? [02:24:46] how can i check? [02:26:35] yeah. sure. is that from nginx [02:26:37] ? [02:27:40] not sure what you mean. never used nginx before. [02:28:08] it's a web server [02:28:38] so .. THE webserver on my nas? [02:28:50] sounds like it, yes [02:29:01] ok. yeah. it works [02:29:33] okay, then only the location / proxy bit seems to be malfunctioning for some reason [02:31:18] https://dpaste.de/kGjU [02:31:45] gute: try 127.0.0.1 [02:31:48] that's localhost [02:32:08] proxy_pass http://127.0.0.1:8000/; [02:33:58] https://dpaste.de/0SEs [02:34:01] better?= [02:34:09] does not work either. [02:34:37] looks good [02:34:42] parsoid is still running? [02:35:08] and this is inside the server {} block? [02:35:19] next to other location blocks? [02:35:52] https://dpaste.de/U61f [02:36:59] wow, it's a pastebin! \o/ [02:37:32] =) sorry. noooobidoo. [02:38:43] maybe move the parsoid block up the way I did in https://dpaste.de/ziQi ? [02:39:03] what kind of response did you get at /parsoid/? [02:40:52] https://dpaste.de/cj4T [02:42:12] getting a 404 for that snippet [02:42:58] https://dpaste.de/tzOT [02:43:34] same [02:44:04] it says "Sidan du s�ker hittades inte." It's swedish for "The page you requested was not found" [02:44:35] http://hagur.synology.me/parsoid/ [02:45:48] oh man, right, running parsoid on a synology nas, crazy talk ;) [02:45:58] hehe [02:46:08] =) [02:46:13] the issue is more configuring nginx to proxy https [02:46:35] http://hagur.synology.me:8000/ [02:47:04] * gwicke points production VE to the new node [02:47:18] haha [02:47:44] * gwicke reconsiders after waiting for http://hagur.synology.me:8000/enwiki/Main_Page [02:48:16] but hey, it loads ;) [02:48:40] gute: could you paste your entire nginx config file? [02:48:45] (to pastebin) [02:48:58] sure. [02:49:44] https://dpaste.de/fQhN [02:51:17] worker_processes 1; <--- is this an issue? [02:51:21] gute: that looks fine to me [02:52:15] hmm... gah! I'm going mad! [02:52:31] *hypnotoad* [02:52:51] hehe [02:53:00] those embedded systems are a bit tricky [02:53:18] often stuff is compiled out, or init systems just work very differently, or.. [02:53:54] alright... yeah, a million things can go wrong. [02:54:03] your web server on that box seems to listen on port 80 too, but the config only sets up 412 [02:54:18] and https isn't on 412 either [02:54:33] to me that suggests that some other web server is also running [02:54:35] no.. theres a dsm interface too [02:54:43] or there is some port forwarding [02:55:02] could be [02:55:32] that's pretty much out of my area, owning no synology myself etc [02:55:52] you could try to use google for 'synology web server' or the like [02:55:57] well it was a good shot in the dark. [02:56:01] 'synology nginx' [02:56:14] yes, sorry it didn't work [02:56:31] also, GO TO BED! ;) [02:56:54] =) [02:57:39] I have installed a bootstrap, ipkg. it has nginx ... [02:57:49] http://ipkg.nslu2-linux.org/feeds/optware/syno-i686/cross/stable/ [02:58:09] there were some errors in the log [02:58:18] maybe google for those + 'synology' [02:58:37] I'm not sure about installing that, since there already is a nginx. [02:58:59] ah, so you didn't install it? [02:59:26] i did at first, but then uninstalled it. then we got going. [03:00:02] okay; I'm out of ideas really [03:00:09] me 2 [03:00:22] thanks a lot though! [03:00:31] often easier to tackle such stuff after some sleep [03:00:35] (wink, wink) [03:00:41] you are welcome! [03:00:44] yes. youre right. [03:00:56] I'm going to bed now. [03:01:25] (I will just go and hit my nas with a sledgehammer first) [03:01:32] good night everyone! [03:01:41] again, thanx! [03:01:47] noooo ;) [03:01:51] goodnight! [03:01:56] ;) [04:01:37] hi can someone help me with a mediawiki issue? [04:01:49] ask, and we'll try to help :) [04:02:24] I upgraded to the latest version and my icons are all broken, including logo which dissapeared [04:06:55] what version did you upgrade to? 1.23.3? is your wiki public? [04:07:37] 1.23.3 and yes: http://www.grailoftruth.com/wikitruth/index.php?title=Main_Page [04:09:32] so, it looks like http://www.grailoftruth.com/wikitruth/skins/common/images/cross1A.png is redirecting to http://grailoftruth.com/wikitruth/skins/common/images/cross1A.png (no www) which is 404'ing [04:09:51] but that url works for me...hmm [04:11:34] yea its wierd, when i right click on the broken image and copy the link into the browser it shows the image, but the image wont display on the main page for some reason...none of the images do [04:13:54] I can't figure out why the images don't work [04:13:59] when I look at my browser's network console it's showing them redirecting [04:15:17] I have no idea either? [04:18:20] Do you have some sort of referrer based blocking in your apache config? [04:18:20] what is that? [04:18:20] When I hit the image URLs directly they work but they 404 when linked from the website [04:18:21] they worked fine in the last version of mediawiki [04:19:51] The domain is changing so it's possible that you have some snippet of apache config that is set to serve a 404 if someone tries to link to your images directly from another domain. That would explain the direct urls working but the images embedded in www. not working. [04:20:29] Seems semi-confirmed because I get images at http://grailoftruth.com/wikitruth/index.php?title=Main_Page [04:20:38] but not http://www.grailoftruth.com/wikitruth/index.php?title=Main_Page [04:21:03] IN both cases the image urls are to http://grailoftruth.com/ [04:21:49] I have some security software installed in wordpress under the root, could that be doing it? (although it worked fine with the last version of mediawiki) [04:22:40] It might. The config wold be something like described here -- http://www.cyberciti.biz/faq/apache-mod_rewrite-hot-linking-images-leeching-howto/ [04:23:13] ok let me take a look... [04:25:41] Maybe in the prior install you didn't have the cross domain image linking problem? Like I said images are working for me from http://grailoftruth.com/wikitruth/index.php?title=Main_Page but http://grailoftruth.com/wikitruth/ redirects to http://www.grailoftruth.com/wikitruth/index.php?title=Main_Page and then the images 404. [04:33:18] I checked th wordpress security and I dont think its that. The apache thing...is that in htaccess? [04:34:01] Yeah it would be in a htaccess file or the main virtual host config files [04:36:56] there are several of these in htaccess: RewriteEngine On [04:38:16] tommy_: Anything that mentions HTTP_REFERER ? [04:38:56] Or just Referer if in a SetEnvIf statement [04:40:20] Yes, there are two of these: RewriteCond %{HTTP_REFERER} !^http(s)?://grailoftruth\.com [NC,OR] [04:41:45] hah. That's probably it. add "(www\.)?" so It looks like !^http(s)?://(www\.)?grailoftruth\.com [04:42:15] That will let you load the images from got and www.got [04:43:43] that worked! THANK YOU!!!! [04:44:02] You're welcome. I'm sort of proud of myself for that guess [04:44:17] I have no idea what that is for...I hope its not something that compromises my security elsewhere? [04:44:17] * bd808 pats self on back [04:44:37] yup,,,your good. I'm impressed. [04:45:01] We just loosened the rule a tiny bit. It will still stop ransomdomain.com from deep linking your images [04:45:16] s/ransom/random/ [04:46:00] ok...I'd much rather have my images work than worry about someone hotlinking to them, especially for such a small website [04:46:23] thanks again, I have been trying to figure this out for days! [04:46:40] The purpose of a rule like that is basically to conserve your bandwidth. It keeps another website from using you as a content delivery service. [04:47:29] Ah ok...good to know. I will keep an eye on that [04:47:53] It was a pretty random issue. I would guess that something else in your WM config changed (or changed in what MW expects) that is causing the redirect to your own www subdomain. [04:49:04] ok...well everything else in mediawiki seems to be working? so i guess its ok unless another error pops up? [04:49:26] Yeah. Get back to making your content now :) [04:49:41] lol. ok....thanks again!!!!! :) [05:43:43] hello [05:44:13] i have added these commands to my localsetting [05:44:13] $wgSMTP = array( 'host' => 'mail.parsintelligent.com', 'IDHost' => 'mail.parsintelligent.com', 'port' => 587, 'username' => 'karami@parsintelligent.com', 'password' => 'Pars98765', 'auth' => true ); $wgShowExceptionDetails = true; [05:44:26] but my wiki can not send email ? [05:59:58] could anybody help me? [06:46:33] could you help me? [06:51:59] Carmela> No. [06:58:27] : could you help? [06:58:39] i have added the commands in my localsetting.php [06:59:06] $wgSMTP = array( 'host' => 'mail.parsintelligent.com', 'IDHost' => 'mail.parsintelligent.com', 'port' => 587, 'username' => 'karami@parsintelligent.com', 'password' => 'XXX', 'auth' => true ); $wgShowExceptionDetails = true; [06:59:24] but i could not get email and show me error [06:59:31] i think i should install pear email [06:59:46] i run wiki on our website [06:59:56] and i do not know who could i do this [07:10:59] @slap-on [07:11:00] I will be slapping annoying people since now [07:11:04] sasan: hi [07:11:15] hi [07:11:42] could you help me? [07:11:44] do you see anything in logs on your mail server? [07:12:13] no nothing [07:12:17] can you even access from that webserver? did you try doing that? [07:12:40] i just add those commands to m y local settings [07:12:41] what is port 587 is it SSL or plain? [07:12:45] and it show fatal error [07:12:50] plain [07:12:51] what you see in fatal error [07:12:52] no ssl [07:12:59] let me copy to you [07:13:00] ok? [07:13:10] you can use pastebin if it;s huge [07:13:38] pastebin? [07:13:42] what is it [07:13:58] http://tools.wmflabs.org/paste [07:14:13] !paste is http://tools.wmflabs.org/paste [07:14:14] Key was added [07:14:43] what should i do there? [07:15:33] paste the error [07:16:03] a min [07:25:05] : sorry [07:25:09] i lost connection [07:25:16] i have added the error there [07:25:28] have you seen the error [07:25:29] ? [07:26:06] let me check [07:26:33] ok [07:27:13] yes I see it [07:27:39] yes indeed you are missing PEAR mail package [07:27:42] so you need to install it [07:28:49] how can i install it? [07:28:52] could you help me [07:28:57] i run my wiki on website [07:28:58] what OS you have on server [07:29:07] i run it on my website [07:29:18] ok, what OS you have on server where your website run [07:30:02] parsintelligent.com [07:30:41] I guess you will have to contact your provider then [07:31:30] you mean ins server should be installed pear mail package? [07:33:08] in my cpanel [07:33:15] i have too many options for email [07:33:21] which one do you mean?? [07:33:31] how could i find pear mail? [07:34:44] : please help me solve this problem [07:35:36] I don't really know how cpanel works I only know how to maintain servers which you have direct access to [07:35:53] you should contact the company which hosts your website [09:33:18] How do I login as ADMIN after I enabled the LDAP Authentication plugin? [09:33:51] ManDay: by using the password for user "ADMIN"? [09:34:24] or do you mean you want to log into a non-ldap admin account? i don't think that's possible. [09:35:02] you should just assign some ldap user(s) to the appropriate groups. i think it's possibel to map ldap groups to mediawiki groups, but i doN#t know how [09:35:21] perhaps you can just assign the mediawiki group directly [09:35:22] !promote [09:35:23] There is no such key, you probably want to try: !autopromote, !createandpromote, !emailconfirmed, [09:35:25] bah [09:35:33] !createandpromote [09:35:33] To recreate the admin user, run "php maintenance/createAndPromote.php" on the command line. [09:35:50] DanielK_WMDE: I'll try with the LDAP groups [09:35:59] The LDAP Authentication man page says, though: [09:36:10] A list of groups that won't automatically have their members // removed, but will have them added. The sysop, bureaucrat, and bot // groups are always considered locally managed. [09:36:44] I'm unsure as to what "locally managed" implies. Does that mean that even if I put a user into a LDAP "sysop" group, he will not be sysop? [09:37:04] ManDay: in that case, log in with the appropriate user once, so a user record is created in mediawiki. then promote that user to admin (aka sysop) status [09:37:34] i havn't used ldap, i can only guess that "locally managed" means that ldap groups are not used, but the normal mediawiki group mechanism is [09:37:37] DanielK_WMDE: Ok, I'll try that cli script [09:38:13] you need one member in the "bureaucrat" group. that user can then assign groups to other users [09:38:49] !userrights [09:38:49] For information on customizing user access, see < http://www.mediawiki.org/wiki/Help:User_rights >. For common examples of restricting access using both rights and extensions, see < http://www.mediawiki.org/wiki/Manual:Preventing_access >. [09:38:56] ManDay: --^ [09:39:04] Thanks, I'll try php maintanence/createAndPromote.php --bureaucrat --force ManDay [09:39:18] that should work, i guess [09:39:45] you can then use Special:Userrights to add yourself to the admin group, etc [09:39:54] I think it worked! [09:40:01] Thank you! [09:45:05] ManDay: have fun :) [10:11:04] If I see a mediawiki the style of which I like, can I readily download their skin? [11:36:51] Hi has someone any Idea if there is a extetion requiert to use the new_grid (as the use it at http://wiki.cyanogenmod.org/w/Devices#vendor=;) [12:54:54] Hello [12:59:16] To whom it may concern, when we upload pictures, we alwayr recieve an error message Directory „mwstore://local-backend/local-public/5/58“ could not be creted. Does anyone know how to fix this problem? [13:01:33] found out how it works thanks anyway [13:03:39] how did you solve it? [13:05:12] mendes_: can the web server create files in the images/ directory? [13:05:26] make sure it can, then it should work [13:06:20] the public user have all rights [13:16:54] Is there anything I have to do if I want to move the wiki from the server's root path into /wiki (because I learned putting it in the server'S root is considered bad practice)? [13:18:10] Buy a family pack of patience and go through https://www.mediawiki.org/wiki/Manual:Short_URL ? [13:18:23] hah [13:21:06] MatmaRex: you need to adjust $wgScriptPath and/or $wgArticlePath [13:21:17] ga, sorry. that was for ManDay ---^ [13:21:30] thanks [13:21:55] There is no ArticlePath in my LocalSettings.php atm - what should I make it? [13:22:00] ManDay: best practice is to move the actual installation to /w/ and map the virtual path for the articles to /wiki/ [13:22:18] yes. i'm currently setting up a short-url rewrite [13:22:26] new to nginx aswell, so I'm reading up [13:22:30] then you will neewwgArticlePath anyway [13:22:41] huh? [13:23:00] sorry - then you will need wgArticlePath anyway [13:23:13] so, yes, create it. [13:40:15] Hello, I want to add search terms to a mediawiki page, for example I want that a page which is about the google calendar is found with the keyword google kalender. [13:40:29] What is a good way to add something like meta keywords to a mediawiki page? [13:42:28] Can a user be called XYZ if the name they enter on the login page is ABC? [13:42:57] johannes__: Create the Google Kalender and page with "#REDIRECT [[Google Calendar]]" [13:42:58] (Specifically I have an LDAP login with user abbreviations such as kl2 but I want users to show up with their complete name in the Wiki) [13:43:55] i restricted the user access to my wiki [13:44:12] till iadd them to a group, but now, i want to open their access just in a page [13:44:16] how can i do that? [13:46:26] could you help me? [13:59:45] please answer me [14:03:31] So I have a question about URLs in page naming. [14:06:36] We all know the issue about readability of escaped characters in urls. eg: 'foo (bar)' converts to 'foo%20%28bar%29' and 'foo "bar"' converts to 'foo%20%22bar%22' [14:07:20] But MediaWiki goes a step further and cleans its URLs of encoded characters like '%', replacing them with '.' [14:07:33] And we end up with a silly mess like this address: [14:07:35] https://en.wikipedia.org/wiki/Pledge_of_Allegiance#Addition_of_.22under_God.22 [14:07:43] Why can't this URL be simply [14:07:51] https://en.wikipedia.org/wiki/Pledge_of_Allegiance#Addition_of_.under_God. [14:08:22] by replacing '"' with '.' ... or go a step further and strip out the encodable characters entirely [14:08:39] I present to you the possible, within our grasp [14:08:48] https://en.wikipedia.org/wiki/Pledge_of_Allegiance#Addition_of_under_God [14:08:56] but alas [14:08:58] https://en.wikipedia.org/wiki/Pledge_of_Allegiance#Addition_of_.22under_God.22 [14:10:23] it's a double-whammy of encoding nightmares [14:15:22] You might could use a renaming scheme. [14:16:06] I'm certainly not overly intelligent about... Well, most anything. But I would think it should be theoretically possible to write a renaming scheme similar to what you're using for short URLs. [14:48:23] Racoon: Actually, it looks like Wikipedia is using a Lua module to dynamically generate anchors that don't include the special characters. [16:33:12] hi [16:33:48] is there any know bug regarding wlm file naming? [16:38:20] Juandev: you may want to ask the WLM folks… [16:38:34] or check WLM list archives [16:38:39] jeremyb: well, I wonder where [16:39:42] hello, i restrict the access to my wiki [16:40:01] but i want user to see one specific page [16:40:09] what should i do? [16:40:10] umm [16:40:36] karami: you can create a ns for that and give speciall permision for that ns [16:41:05] how could i generate ns? [16:42:01] you need to better define your problem [16:42:18] please see my wiki [16:42:23] probably should use a whutlist and not NS [16:42:25] parsintelligent.com/wiki [16:42:34] no body can access to wiki [16:42:42] jeremyb: yup, thats right [16:42:52] I think they only want one specifc user to be able to access this page, though. [16:42:52] i want to get access to one page only [16:42:57] https://www.mediawiki.org/wiki/Manual:$wgWhitelistRead [16:43:19] Traditional Whitelisting won't do that, I didn't think. [16:43:25] moliver: did I mention better define? [16:43:46] jeremyb: You did. :D So, I thought I'd help out. [16:43:54] :) [16:44:43] you mean i should add [16:44:53] $wgWhitelistRead = array( "specific page n ame") [16:45:08] karami: That would work, but it would make that page available to -anyone-. [16:45:23] yes, i want that:) [16:45:23] karami: Even anonymous users. [16:45:26] let me test [16:45:28] :) [16:45:34] I stand corrected, then. =p [16:50:22] : i add this [16:50:38] $wgWhitelistRead = array( "خبرنامه های کلی") [16:50:47] but my wiki is failed;( [16:50:59] it is not loaded [16:53:07] : please help [16:56:05] karami: That should be it, but I'm not sure if there's anything special to do with non Latin characters. =/ [16:56:29] karami: Wait, did you terminate the line with a ;? [16:56:42] karami: I.e. $wgWhitelistRead = array( "خبرنامه های کلی"); [17:41:24] I have an extension that defines a parser function whose output includes a CSS file. Unfortunately, the CSS doesn't get included on the wiki page unless the page has just been "purged". [17:41:59] I assume it's due to some ResourceLoader/MediaWiki cache behavior. Is there any way to get around that (short of disabling the cache)? [17:44:20] jeremyb: dataset for performance dashboard [17:45:36] Yaron: Should work if you don't add to wgOut, but to the parser's ParserOutput object [17:55:01] $parserOutput->addModuleStyles [17:58:18] I've just used bitnami to get one click installation of media wiki on OS X. It seems to have worked, my browser goes to http://localhost:8080/mediawiki/Main_Page ok, but where is this page on my computer? [18:00:08] I've got several hundred PDFs in a folder, and I want to create a web of maybe a dozen pages -- each page will link to several of these PDFs [18:00:22] So I'm trying to figure out where I should put the folder of PDFs [18:02:08] pi-: You can use [[Special:Upload]] on the wiki to upload the images. [18:02:14] Carmelo: Meet Carmela. [18:02:19] :) [18:03:35] pi-: You can also use the script maintenance/importImages.php to import a directory of files [18:04:02] (mediawiki is a mite confused about images and other media on the whole) [18:04:47] oh so I can't just find the root folder, drag in files using finder, and link them from the pages? [18:05:23] pi-: No, because MediaWiki needs to track the images in its database [18:05:47] I think you can drag files into the upload page, though. [18:05:57] Only into the actual file input, but still [18:06:03] mmk, how about non-image files (like PDF)? [18:06:12] Oh, yeah, it's fine [18:06:20] MediaWiki calls everything an image that isn't text, basically [18:06:28] why ot [18:06:30] ^not [18:06:38] What? [18:06:46] Seems reasonable :) [18:08:10] I can't see how to implement either of the solutions you gave. There must be some Basic readme document I've missed...? [18:08:40] FoxxT: sorry, I missed your response before. Thanks, I'll try that! [18:10:49] pi-: Go to http://localhost:8080/mediawiki/Special:Upload [18:12:27] Is there any way I could've found that without knowing that URL? I look in http://localhost:8080/mediawiki/Special:SpecialPages but it doesn't seem to be in there... [18:15:40] It won't let me upload a PDF: "Permitted file types: png, gif, jpg, jpeg" [18:18:47] FoxxT: it worked! [18:18:59] Yaron: \o/ [18:23:16] pi-: In the base dir of your MW intallation find LocalSettings.php [18:23:16] pi-: Add a line: $wgFileExtensions[] = 'pdf'; [19:58:01] hey all, one of my wikis just recently (no pun intended) started showing a blank page and recording the errors shown at https://gist.github.com/Calygos/8fa649c5d96f95d7e885 when accessing the Recent Changes page. I've only found one reference to this error but it mentioned a version error that I doubt is the case here. [19:58:03] Any ideas? [20:02:17] jcl: what about error messges for other pages? [20:02:43] https://www.mediawiki.org/wiki/Manual:How_to_debug [20:02:48] nothing outside the normal stuff (missing favicon.ico, etc.) [20:03:11] i did enable the wiki debug logging but it wasn't very useful in this situation [20:04:31] i have 5 wikis all on the same version of everything and this one wiki (the smallest of the 5, in fact) is the only one showing the problem [20:05:24] since it just started recently (not sure exactly when), and nothing had changed in the server config, i'm guessing it's some kind of content issue, but i am not sure how to confirm that. [20:07:46] actually i know exactly when it started based on the error logs, Friday afternoon. :P [20:11:51] actually i'm not sure about that. the first error message i see about it is from last friday but the last recentchanges table entry in the database is from august 24 [20:19:29] MatmaRex: Thanks. [20:20:19] anytime [20:46:25] Thanks FoxxT! [20:55:07] superm401: do you know why the anchor gets stripped? https://translatewiki.net/wiki/MediaWiki:Helplogin-url [20:55:46] Nemo_bis, IIRC, that is supposed to be a page name or URL. [20:56:19] superm401: which it is [20:56:47] Nemo_bis, by anchor you mean the fragment (#How_do_I_register)? [20:57:00] Maybe that's not supported. Probably just no one thought to support it. [20:57:25] Heya. Anyone know if there's a way to setup namespaces that are searched by default? I.e., it only searches the project namespace by default, but I want it to search the category namespace as well. [20:58:19] ^ ping manybubbles [20:58:39] moliver: yeah - there is a global - let me find it [20:59:14] manybubbles: Many thanks. I thought there might be, but my Googlefoo is weak. [20:59:38] moliver: its $wgNamespacesToBeSearchedDefault [20:59:50] you can see the docs in DefaultSettings.php if you search for it [21:00:01] It's an array, I would guess? [21:00:02] looks like you do $wgNamespacesToBeSearchedDefault[NS_PROJECT] = true; [21:00:05] yeah [21:00:10] manybubbles: Ah, kk. Thanks! [21:00:14] an array where the keys are the namespace [21:12:02] manybubbles: are you into mediawiki core by any chance ? i have a question related to blocking [21:12:44] the question is: is it possible to give a user the right to block for an hour only ? [21:13:26] I've just done the one-click media-wiki OSX installation -- Where is my wiki located on my filesystem? [21:13:37] csteipp: will probably know [21:14:01] matanya: I don't think there's any way to set a time limit on a security right, unfrotunately. [21:14:13] i thought so too [21:14:31] but i would like it to be confirmed [21:15:54] Ah found it /Applications/mediawiki-1.23.3-0/apps/mediawiki/htdocs [21:21:33] I didn't know we had a one click option [21:23:38] matanya: there's a bug somewhere for expiring user groups [21:24:00] could be done as an extension reasonably easily [21:32:15] jeremyb: bitnami [21:32:51] https://bitnami.com/stack/mediawiki [21:35:22] How do I create a new wiki page? [21:36:46] pi-: i'm afraid no one here knows about that distribution, much less is able to support it [21:37:14] generally you can create new pages by searching for the page title, then clicking the link to create it that will be shown [21:38:05] I don't think it's a distribution specific question -- Yes, super, Thanks! [21:38:38] legoktm: that was not the question. i was : i can i give patrollers the right to block for an hour only [21:38:56] matanya: nope. [21:41:44] oh, you mean block duration not right duration [21:41:52] matanya [21:41:53] yes [21:42:09] thank legoktm [21:42:12] s [21:42:41] that's a bit more interesting than my first interpretation [21:53:17] jeremyb: it was a request on he.wiki [21:53:32] i said it was impossible, but came here to verify [21:53:58] matanya: It could be written, but it's not currently possible. [21:54:12] matanya: Also, I don't think it's a good idea to have yet more special kinds of rights. [21:54:34] James_F: i agree to both points [21:54:41] *with [21:54:42] OK. :-) [21:55:13] James_F: you see, we do agree from time to time :D [21:55:34] matanya: Often! :-) [22:06:30] Is there some standard way to present quoted text? [22:07:02] On my page I first give a link to a PDF then underneath I quote a paragraph out of it [22:07:07] "like this" [22:07:15] Tx [22:08:12] Is there any way to have it indented? [22:09:09] :like this [22:09:16] If I give a leading space it won't wordwrap [22:09:28] and then ::like this [22:09:38] Fantastic! [22:09:56] and it is good to read the docs/manual [22:10:10] Why did I put off using a wiki to organise my research for 3 months?! [22:10:38] until you start signing your mails with ~~~~ [22:10:58] then you know you took it too far [22:11:50] matanya: could you link me to doc/manual? [22:11:55] or look for the edit button on sites which are not wiki(p|m)dia when you see a mistake and wonder where it is [22:12:24] pi-: https://www.mediawiki.org/wiki/Help:Editing_pages [22:13:10] and https://www.mediawiki.org/wiki/Help:Formatting [22:15:40] super [22:16:41] I think it was something as simple as not wanting to have to install three different things that kept me from doing this earlier. Now I found bitnami that set it all up in a single click. [22:17:45] I wonder whether a thesis could be submitted as a wiki [22:18:04] if you can, let us know :) [22:18:29] ewww, no. dont quote/indent like that. use
[22:19:35] and https://en.wikipedia.org/wiki/Category:Quotation_templates [22:30:17] matanya: Nope, you either have the right to block or not... but it's an interesting idea for a feature [22:30:39] thanks csteipp