[00:35:44] PROBLEM - Puppet failure on tools-trusty is CRITICAL 40.00% of data above the critical threshold [0.0] [01:20:45] RECOVERY - Puppet failure on tools-trusty is OK Less than 1.00% above the threshold [0.0] [05:02:42] Betacommand : I am trying to run an http server on tools labs. : http://wiki.languagetool.org/http-server [05:02:54] sorry. about that. [05:03:12] thought i was on a different tab. [07:31:03] ankita-ks: Have you taken a look at https://wikitech.wikimedia.org/wiki/Help:Tool_Labs#Web_server ? [07:31:12] yes [07:31:41] I have a very stupid question. When the service is up and running, what ip is it running on? [07:33:13] Have you tried simply sending an HTTP request from inside the instance, curl it or something? [07:33:54] on localhost, that is. [07:34:36] to what? the localhost? [07:35:39] Yeah, you could curl localhost to see if that gives any output. [07:36:19] nope. [07:36:22] does not [07:36:28] and i don't think it should [07:36:36] ankita-ks, this should work: $ curl https://wtfismyip.com/text [07:36:37] the job is not running on this instance [07:38:11] ankita-ks: https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Web seems to have some useful tips. [07:38:28] I already read that. [07:38:42] ankita-ks: For example, https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Web#Static_file_server [07:38:59] Right, but that's only for static resources [07:39:15] yes [07:39:20] I was going to suggest that this might be the publicly accessible URL: tools-static.wmflabs.org/toolname [07:39:23] but that doesn't make any sense. [07:39:26] no [07:39:41] The publicly accessible url displays what it should [07:40:00] but that is not what i want. I think the server is up and running but I have no way to test where. [07:40:48] ankita-ks: https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Web#Other_web_servers says that you can access your tool at https://tools.wmflabs.org/MyToolName [07:41:01] yeah [07:41:28] That doesn't seem to be working? You could test once with $ python -m SimpleHTTPServer 8000 to see if the basic setup works. [07:41:29] yeah:\ [07:41:34] So i can see the html pages there. But I also need the specific port number to ping [07:41:42] It's going to be port 80. [07:41:47] okay [07:41:48] Since it's a normal HTTP server. [07:41:52] let me check [07:41:59] But wait, you were running http://wiki.languagetool.org/http-server right? [07:42:43] ankita-ks: Their documentation says the default port it runs on is 8081. You could try both, 80 and then 8081. [07:43:13] okay [10:00:59] 6Labs, 10Labs-Infrastructure: missing database entries at categorylinks table on dewiki db - https://phabricator.wikimedia.org/T72711#1303478 (10Merl) Same problem again after two rollbacks: ``` $ mysql -hs5.labsdb -vvve "select page_id, page_latest, cl_to from dewiki_p.page left join dewiki_p.categorylinks o... [11:39:30] Hi, So if I have an http server running on tools-labs, how do I figure out what port it is listening on? [11:39:40] Betacommand , Coren , greg-g ^ [11:42:41] https://gist.github.com/ankita-kumari/57d76f47ba1d8711be16 [11:42:51] Here is a gist of what i did to get it running. [11:47:32] https://gist.github.com/ankita-kumari/6982718573541d4839a6 [11:47:42] Here is a gist of access logs for today. [11:48:01] This is what I have been trying to do : http://wiki.languagetool.org/http-server [12:13:17] ankita-ks: you already have a lighttpd running, you should stop that first with "webservice stop". Then the portgrabber adds the port number as last parameter to the command call [12:13:34] okay [12:13:41] ankita-ks: it should probably look more like this: https://dpaste.de/aLFx [12:14:27] ah ofcourse [12:14:30] trying this [12:14:39] make sure to assign it enough memory in the jstart call, the logfiles then are named "languagetool.err" and "languagetool.out" in your home directory [12:15:53] how do i assign memory? [12:16:12] jstart -N languagetool -q webgrid-generic -mem 4G ~/httpserver.sh [12:16:20] the -mem 4G part [12:18:05] ah okay [12:18:07] thanks [12:18:09] trying that [12:19:41] sitic : it's status is stuck at qw [12:19:47] and doesn't change to r [12:20:11] it might take a minute to start [12:26:48] sitic : no change yet [12:29:30] 10Tool-Labs-xTools: tools.wmflabs.org/xtools shouldn't give a HTTP 301 - https://phabricator.wikimedia.org/T100015#1303566 (10Glaisher) 3NEW [12:30:00] ankita-ks: yeah the jstart command was wrong [12:30:01] jstart -N languagetool -once -q webgrid-generic -l release=trusty -mem 4G ~/httpserver.sh [12:30:04] should work [12:30:26] So I'm getting a network timeout when trying to SSH into labs terminal [12:31:52] 10Tool-Labs-xTools: tools.wmflabs.org/xtools shouldn't give a HTTP 301 - https://phabricator.wikimedia.org/T100015#1303573 (10Glaisher) err, it should be a 301 to the proper place.. [12:32:09] 10Tool-Labs-xTools: tools.wmflabs.org/xtools shouldn't give a HTTP 301 to another redirect - https://phabricator.wikimedia.org/T100015#1303574 (10Glaisher) [12:35:55] Wakey wakey Coren [12:40:45] 10Tool-Labs-xTools: tools.wmflabs.org/xtools shouldn't give a HTTP 301 to another redirect - https://phabricator.wikimedia.org/T100015#1303580 (10Technical13) @Glaisher we haven't been too worried about fixing the redirects yet because we're in the process of migrating all of the tools to xtools.wmflabs.org/tool... [12:40:57] sitic : so it works now [12:41:18] But how do I figure out what port it is listening on? [12:41:27] 10Tool-Labs-xTools: tools.wmflabs.org/xtools shouldn't give a HTTP 301 to another redirect - https://phabricator.wikimedia.org/T100015#1303581 (10Technical13) p:5Triage>3Lowest [12:41:37] say, if i send a curl request, where do i send it [12:42:16] ankita-ks: http://tools.wmflabs.org/languageproofing [12:42:53] do you really need to have the port on wich the server runs on? [12:44:30] not really [12:44:33] and it worked. [12:44:36] thank you so much! [12:44:53] but i need to test one last thing (whether I really need the port) [12:46:11] :-) [12:49:24] sitic : say i did need a port number, what would it be? [12:49:44] the thing is when I send a curl request from command line, i get the appropriate response [12:49:54] but when i do the same from a ui, i don't [12:54:46] it's a new one every time you start it, and you can't determine on which server it starts … [12:55:35] ankita-ks: you could write a second bash script wich says the server and port, like: https://dpaste.de/x2AT [12:56:20] but I'm still not sure what the difference would be to instead just curling tools.wmflabs.org/languageproofing [12:58:39] sitic : https://github.com/ankita-kumari/languagetool/blob/master/languagetool-query/proxy.php [12:58:48] this where it requires a port number [12:59:12] I can't figure out how this would be different from a normal curl request. [12:59:31] But it just doesn't work. The changes are not reflected in the ui. [13:00:58] i think the problem is, that when you access http://tools.wmflabs.org/languageproofing/index.html the java server gets a request for /languageproofing/index.html, not /index.html [13:01:43] so it has to be aware that it doesn't live on the root directory, instead /languageproofing/ is it's root [13:03:17] I don't get it. [13:04:29] I am a little confused here. Do you mean when I send a request, the server gets a request for index.html? [13:05:30] Because I don't think that figures here in anyway. The index.html and porxy.php files are on my system. index.html gives the ui and proxy.html specifies where to send the request and such [13:05:45] sitic ^ [13:06:10] index.html was just an example [13:06:46] let's say you want to curl http://127.0.0.1/someurl [13:07:11] now that url is http://tools.wmflabs.org/languageprood/someurl [13:07:34] in the first case, the web server gets a request for /someurl [13:07:43] in the second for /languageproof/someurl [13:08:18] the server has to be aware that /languageproof/ is a prefix for all urls [13:08:53] so that it knows what to do (hope it's a bit clearer now) [13:09:07] yes it is [13:09:19] one doubt though [13:09:26] why is it different in two cases? [13:09:58] when i curl http://tools.wmflabs.org/languageproof/ [13:10:29] and when i send a request from ui. [13:11:41] ah curl http://tools.wmflabs.org/languageproof/ works (it also get's the correct response)? [13:11:48] yes [13:11:50] it does [13:12:10] to be specific [13:12:11] curl -vvv --data "language=en-US&text=a simple test" http://tools.wmflabs.org/languageproofing/ [13:12:22] gets be the exact xml response it should [13:13:02] then you can forget what I said [13:13:20] ah okay [13:13:24] but it only works from outside [13:13:46] when I curl from within labs I get a 400 [13:13:59] https://dpaste.de/KYoU [13:14:11] wait now it does [13:16:08] might just have been a bad moment. But then also I don't know why only curl works. you should just have to point to http://tools.wmflabs.org/languageproofing/ on port 80 in https://github.com/ankita-kumari/languagetool/blob/master/languagetool-query/proxy.php [13:18:01] yeah [13:18:02] :| [13:38:44] (03PS1) 10Sitic: Added setup.py to avoid python-cache issues with celery [labs/tools/crosswatch] - 10https://gerrit.wikimedia.org/r/212781 (https://phabricator.wikimedia.org/T97900) [13:39:02] (03CR) 10Sitic: [C: 032 V: 032] Added setup.py to avoid python-cache issues with celery [labs/tools/crosswatch] - 10https://gerrit.wikimedia.org/r/212781 (https://phabricator.wikimedia.org/T97900) (owner: 10Sitic) [14:50:00] 10Tool-Labs: Move toollabs to designate - https://phabricator.wikimedia.org/T100023#1303692 (10yuvipanda) 3NEW [14:50:51] 10Tool-Labs: (Some) Precise grid jobs seem to be failing - https://phabricator.wikimedia.org/T99946#1303699 (10yuvipanda) Most of them seem to be webgrid jobs for dplbot but tools.wmflabs.org/dplbot is up [14:51:42] 6Labs, 10Continuous-Integration-Infrastructure: Designate should support split horizon resolution to yield private IP of instances behind a public DNS entry - https://phabricator.wikimedia.org/T95288#1303702 (10yuvipanda) [14:52:46] hi, does toolslabs has backups of tools accounts? [14:59:36] Steinsplitter: atm no, why? [15:00:49] no google cache either :( [15:01:52] yuvipanda: thx, just curious [15:02:41] yuvipanda: ugh :( Is https://wikitech.wikimedia.org/wiki/Help:Tool_Labs/Backups all lies? [16:22:50] 10Tool-Labs: (Some) Precise grid jobs seem to be failing - https://phabricator.wikimedia.org/T99946#1303793 (10scfc) 5Open>3Resolved a:3scfc I was confused by the `*-webgrid-*`, but apparently it was just the host the jobs were scheduled on, the actual jobs were mostly scripts with the error message: ```... [16:51:23] 6Labs, 6operations, 3ToolLabs-Goals-Q4: Investigate kernel issues on labvirt** hosts - https://phabricator.wikimedia.org/T99738#1303823 (10yuvipanda) I feel like an idiot - I upgraded the kernel on labvirt100*6* - thankfully caught myself before doing a reboot. I'll let it be, and just upgrade 1005 as well. [16:51:43] so close to disaster [17:12:28] Change on 12wikitech.wikimedia.org a page Nova Resource:Tools/Access Request/Kjschiroo was created, changed by Kjschiroo link https://wikitech.wikimedia.org/wiki/Nova_Resource:Tools/Access_Request/Kjschiroo edit summary: Created page with "{{Tools Access Request |Justification=I am a graduate student with GroupLens at the University of Minnesota. I plan to use the Tools project to conduct research on Wikipedia...." [17:12:52] 6Labs, 6operations, 3ToolLabs-Goals-Q4: Investigate kernel issues on labvirt** hosts - https://phabricator.wikimedia.org/T99738#1303843 (10yuvipanda) Whoops - 1005 isn't coming back up. In shell with: ```ALERT! /dev/disk/by-uuid/861a4750-9243-4da7-b566-8c3cebfd6114 does not exist. Dropping to a shell!```... [17:27:43] 6Labs, 6operations: labvirt1005 doesn't boot up - https://phabricator.wikimedia.org/T100030#1303871 (10yuvipanda) 3NEW [18:11:36] Change on 12wikitech.wikimedia.org a page Nova Resource:Tools/Access Request/Kjschiroo was modified, changed by Tim Landscheidt link https://wikitech.wikimedia.org/w/index.php?diff=160279 edit summary: [18:55:27] Hi, can someone does something to fix i-0000086c.eqiad.wmflabs? Instance sate is ERROR and can not connect to it via ssh: channel 0: open failed: administratively prohibited: open failed [19:48:27] PROBLEM - ToolLabs Home Page on toollabs is CRITICAL - Socket timeout after 10 seconds [20:10:53] how can I view the my.cnf file? [20:11:15] I am tring to set up the SSH tunnel for MySQL Workbench [20:12:56] hello? [20:14:54] Superyetkin: it should be at /home//replica.my.cnf [20:14:54] PROBLEM - Puppet failure on tools-exec-07 is CRITICAL 100.00% of data above the critical threshold [0.0] [20:15:15] that is for replica [20:15:32] it does not include the mysql bind info (ports, etc.) [20:15:54] mysql is default ports etc though [20:15:56] Superyetkin, it should contain the info you need to connect [20:16:28] no, it has the username and password [20:16:36] I need the mysql bind port [20:16:52] I am tring with 127.0.0.1 to no avail [20:16:58] If it's not specified, it'll be the default 3306 [20:17:16] 127.0.0.1:3306? [20:17:31] Superyetkin: the address would be *.labsdb [20:17:36] it depends what you need to connect to though [20:18:03] no, I can connect fine with putty [20:18:16] just trying to adjust MySQL Workbench [20:18:38] I have set up an SSH tunnel on putty and am trying to use that in MW [20:18:49] Superyetkin: what database are you try to connect to? different databases are hosted on different servers and so on [20:19:00] trwiki.labsdb database [20:19:18] okay, use trwiki.labsdb instead of 127.0.0.1 [20:20:01] as hostname? [20:20:05] unknown host... [20:21:32] I believe I set up the SSH tunnel correctly on putty [20:21:54] but how can I use that to connect to MySQL Workbench? [20:22:06] hey panda [20:22:23] Superyetkin: you need to connect through ssh on MySQL workbench [20:22:42] man, that is what I am doing [20:22:53] but I need to use the tunnel I created on putty [20:23:22] I am on Win XP SP3 and using MySQL Workbench 5.2.24 [20:23:39] Superyetkin: MySQL Workbench -> New Connection -> Connection Method: Standard TCP/IP over SSH and then use that form to connect to tools and then trwiki.labsdb [20:23:54] due to recent cipher changes, I can no longer connect to the database via pure SSH config [20:24:27] JohnLewis: I already have a connection [20:25:06] and I need the localhost/127.0.0.1 thing to work [20:25:37] yuvipanda: are you there? [20:26:01] if MySQL Workbench is connecting to tools via ssh, trwiki.labsdb is the hostname to use. It works, I've just set up a test on my install [20:26:18] following the guide at https://phabricator.wikimedia.org/T99942#1302319 , I cannot still get MW to work [20:26:54] JohnLewis: that no longer works for my version [20:27:01] there have been cipher changes [20:27:23] I used to connect fine but it was broken a few days ago [20:27:54] bblack: do you have a word on this? [20:29:56] anyone there that can help out with this? [20:30:49] Hello, is there an issue with SSH at the moment? I'm getting an error that the "Server unexpectedly closed network connection" [20:31:26] 6Labs: Document PuTTY based tunneling workaround for accessing labsdb via older MySQLBench versions - https://phabricator.wikimedia.org/T99942#1304074 (10Superyetkin) valhallasw: That does not work. Are there any dependencies that need to be installed? [20:32:58] Hazard-SJ: has it been working for you in the past or is this a fresh attempt? [20:33:57] um… fussing with my irc client, brb [20:34:01] andrewbogott: It has worked in the past. The last time I used it was 5 days ago [20:35:29] hello? [20:36:01] ok, I’m back for real now [20:36:06] andrewbogott: It has worked in the past. The last time I used it was 5 days ago. [20:36:21] Can you tell me what system you’re connecting to? [20:36:33] tools-login [20:36:43] HEY [20:36:45] I can connect via WinSCP, but not via PuTTY [20:37:11] * Hazard-SJ tells Superyetkin hello [20:37:12] Hazard-SJ: are you using MySQL Workbench? [20:37:16] !log tools deleted and depooled tools-exec-07 [20:37:18] ok — I see that indeed the ciphers were updated. If you’re subscribed to labs-l I think there was an email with instructions. [20:37:23] No, Superyetkin [20:37:24] Logged the message, Master [20:37:28] Superyetkin: I’m guessing that you’re running into the same issue? [20:37:47] hmm, yes [20:37:55] the infamou cipher issue [20:38:21] this has really messed everything [20:38:49] follow up on this [20:38:50] https://phabricator.wikimedia.org/T99942 [20:39:10] recommendations listed there do not work at all [20:39:15] PROBLEM - Host tools-exec-07 is DOWN: CRITICAL - Host Unreachable (10.68.16.36) [20:40:28] Hazard-SJ: could you update putty to latest version and try again? [20:40:49] So you’re running XP, both of you? [20:41:13] I am on Win XP SP3 [20:41:23] MySQL Workbench version: 5.2.24 [20:42:05] mutante: any news about the MW integration? [20:44:04] Superyetkin: MW = mysql workbench? mediawiki? i am not sure [20:44:14] mysql workbench [20:44:16] mutante: Okay, I'll check on that [20:44:18] the reply from valhalla [20:44:32] about the paramiko version [20:44:34] the reply from valhalla does not help [20:44:41] I tried that [20:44:48] paramiko version will not help [20:44:49] Superyetkin: so you’re running the latest putty already? [20:45:01] (Sorry, the Ops are flying a bit blind on account of none of us run windows) [20:45:09] putty works for him , right [20:45:21] it's the mysql workbench [20:45:37] yes, putty and WinSCP work for me [20:45:44] MySQL Workbench, no [20:46:11] andrewbogott: it uses paramiko for ssh [20:46:38] paramiko update does not fix the issue [20:47:05] instead, an error saying "cannot open ssh tunnel" pops up [20:47:25] but it's a different error since the update? [20:47:29] mutante: do you have some awake time left to help with this? In theory Superyetkin should be tunneling via Putty but no one knows how to do that. [20:47:59] but I was told that is the right way to go... [20:48:04] mutante: As I understand it, tunneling is a (new) workaround because workbench doesn’t support the right version of ssh. [20:48:09] especially, after this [20:48:10] https://phabricator.wikimedia.org/T99942 [20:48:20] andrewbogott: did you see the ticket ? https://phabricator.wikimedia.org/T99942#1302377 [20:48:47] yes, I think that’s all I’m saying :) [20:48:56] Sorry, I only just got here, catching up w/everything. [20:48:56] mutante: That worked, thanks [20:49:11] Hazard-SJ: great :) [20:49:14] :) [20:49:23] what worked? [20:50:40] all over the net, it says the hostname should be 127.0.0.1 but that does not seem to be correct in my case... [20:50:51] may there be any other mysql hostname? [20:50:55] Updating PuTTY solved my problem, Superyetkin [20:51:18] hmm, I cannot upgrade MW... [20:51:37] Win XP SP3 does not support newer versions of it [20:52:20] mutante: do you know how paramiko works? [20:52:57] I thought paramiko workaround would work but got stuck on library stuff, I suppose [20:54:12] Superyetkin: I think your best bet is to find someone who’s using your same (or similar) setup — all any of us OPs can do is feel around in the dark. [20:54:50] why is it so hard to deploy a virtual Win XP box? [20:55:18] I suppose you are using VMs right? [20:55:42] It’s not that we can’t, just that we’re unlikely to be experts in half a dozen OSes that we never use :) [20:56:05] but, you CAN try [20:56:06] I don't get why XP is still actively used anymore tbh [20:56:26] JohnLewis, lol [20:56:45] Superyetkin: sorry, i'm not using Windows [20:57:04] you CAN create a VM, right? [20:57:10] that should not be too hard [20:58:08] Superyetkin: since you tried the paramiko update, did the error message change or stay the same? [20:58:13] that's what i was wondering [20:58:27] maybe it's just an additional/other problem now [20:58:29] the error mesage changed [20:58:44] after paramiko update, I got "cannot open SSH tunnel" [20:58:53] let's give both of the error messages to valhall [20:58:58] I suspect there is a missing library [20:59:01] valhallasw [20:59:25] that sounds likely and as if the update _did_ fix something [20:59:28] but on Windows, you cannot upgrade pyhon libraries [20:59:30] but there is just a second thing to fix now [20:59:51] 6Labs, 6operations, 3ToolLabs-Goals-Q4: Investigate kernel issues on labvirt** hosts - https://phabricator.wikimedia.org/T99738#1304100 (10Andrew) For the record: I've seen this once before. During an earlier labvirt crash I tried updating to a non 3.13 kernel (.15 I think?) and that box couldn't see its f... [20:59:55] no, I do not think it has "fixed" anything [21:00:03] it is just a different thing [21:00:11] well the error message you had before the update was related to the ciphers [21:00:24] and after the update it's just in general about opening a tunnel [21:00:25] right [21:00:31] ok, gotta to to sleep, too much jetlag. Sorry I’m useless. [21:00:46] so that would mean the paramiko update fixed the cipher issue [21:01:01] that does not help... [21:01:14] because, the paramiko update is not working... [21:01:16] that's the way to get closer to fixing it [21:01:28] it's possible to have more than 1 issue [21:02:08] we cannot even make sure the paramiko "update" went through [21:02:11] can you paste both error messages for me once? [21:02:23] or just on the ticket right away [21:02:42] cannot copy & paste from MW [21:02:51] well, it sounds like it's a simple copy [21:02:58] so not sure what "went through" would be [21:03:15] screenshot? [21:03:18] first error: no kex algorithm, etc. [21:03:22] you can just drag&drop an image into a phab ticket [21:03:30] seconf: cannot open SSH tunnel .... [21:03:56] ok, thanks [21:04:32] why can we not go with the putty tunnel approach? [21:04:49] 6Labs: Document PuTTY based tunneling workaround for accessing labsdb via older MySQLBench versions - https://phabricator.wikimedia.org/T99942#1304102 (10Dzahn) 14:01 < mutante> well the error message you had before the update was related to the ciphers 14:01 < mutante> and after the update it's just in general... [21:05:01] I have also tried that but got stuck with the mysql hostname, etc. [21:05:54] Superyetkin: http://homer.netsal.selu.edu/mediawiki/index.php/Creating_a_SSH_tunnel_to_MySql_using_PuTTY_-_Tyler_Hunt [21:05:59] that just worked for me [21:06:22] JohnLewis, tried that one, [21:06:26] DID NOT WORK AT ALL [21:06:53] really? works perfectly fine for me [21:07:14] JohnLewis, give me your connection settings [21:07:34] maybe I am missing something here... [21:08:31] Superyetkin: hostname: "tools-login.wmflabs:22" -- tunnel: source port: "3306" destination: "trwiki.labsdb:3306" [21:08:48] *tools-login.wmflabs.org:22 [21:09:30] tunnel??? [21:09:34] 6Labs: Document PuTTY based tunneling workaround for accessing labsdb via older MySQLBench versions - https://phabricator.wikimedia.org/T99942#1304105 (10Dzahn) So i would say we have found out what fixes the issue with us now using different key exchange protocols: upgrading paramiko as described by valhallasw.... [21:09:38] plus ensure to press add (I forgot first time). It'll appear as L3306. then in MW, just connect to 127.0.0.1:3306 [21:09:56] Superyetkin: Connection -> SSH -> Tunnels [21:11:26] 6Labs: Document PuTTY based tunneling workaround for accessing labsdb via older MySQLBench versions - https://phabricator.wikimedia.org/T99942#1304107 (10JohnLewis) http://homer.netsal.selu.edu/mediawiki/index.php/Creating_a_SSH_tunnel_to_MySql_using_PuTTY_-_Tyler_Hunt worked for me by having putty open 127.0.0.... [21:12:24] DOES NOT WORK... [21:13:07] Superyetkin: did you open putty and login then open MW and connect to 127.0.0.1:3306 using your replica username and password? [21:13:20] OF COURSE [21:13:40] the kex algorithm error still appears... [21:15:36] what to do? [21:18:58] ?????? [21:21:47] ?????? [21:23:05] Don't use XP? Wait for upstream to fix the issue? That's the only way to tunnel via PuTTY as far as I can see [21:24:41] JohnLewis: huh? [21:24:47] cannot upgrade my OS [21:25:01] Windoes is not FREE [21:26:22] can you create a WinXP VM and test it yourself? [21:26:33] indeed but there has to be a point when people go 'should I perhaps upgrade' which is usually when all the secure ciphers the version supports are considered insecure and the next OS released has reached its EOL. [21:27:37] I take Microsoft for granted regarding the EOL crap... [21:28:08] you cannot force people to upgrade [21:28:23] this is about budget (money). Understand? [21:28:43] True, I can not force people to. But there is the trade of. Do we support an OS that is about 15 years old or ensure the security of our systems? [21:28:45] new hardware, new OS: for free? NO [21:29:16] I do not expect you to support it [21:29:19] there are free operating systems [21:29:23] we use them [21:29:33] that's why we cant test it on windows [21:29:39] but you can at least try to reproduce the issue and fix it [21:30:34] I've reproduced the issue and I've proposed a fix for it that works. If it doesn't work for every users case, that's the game of laws. IIRC one XP user has already claimed we've fixed their cipher issue via methods already suggested. [21:30:56] if I gave you access through Teamviewer, would you be able to look at it?