[00:03:42] SQL: I'll look into iy [00:03:44] *it [00:07:21] what's the time in UTS it started running? [00:08:45] worker-02 has this weird error: [00:08:52] https://www.irccloud.com/pastebin/QDrzKa1T/ [00:09:13] worker-01's journal is flooded by diamond errors (which I just restarted) [00:24:45] and in destructor tell it's gone [00:24:48] Because I've already looked for the obvious answers. And why would anyone assume that someone who is able to make a program work in a default case hadn't already read the manpage? Kismet does not work out of the box. You actually have to know how to configure it. But, I gave a none standard case where it does not work...for which I'm looking for a problem. [00:24:53] what about insert mode [00:24:55] zhuyifei1999_: somewhere around 20:00 UTC [00:25:25] then it's probably worker-01. logs are lost :( [00:25:30] :( [00:25:45] If run on the toolforge, that query crashes (out of memory) [00:26:02] Which is odd - I know i've pulled a list of sysops before [00:26:10] <|aaron8> No more so than "A is a subset fo B" [00:26:27] <|aaron8> oh yes i see that mojava is avialbe [00:26:34] SQL: then that makes sense [00:26:36] <|aaron8> doesn't that kill my hashes? [00:26:36] <|aaron8> Ideally all of the lifts are isolated to the problem-domain specific combinators [00:43:12] jiji606, as long as it's related to linux. ;) [00:43:22] ( •_•) [02:55:59] I mean, seriously, trying to *edit* video is tough enough on a beefy desktop I have locally. Add in network bullshit, and an overworked server or 40... [02:59:48] Hey zhuyifei1999_, didn't quite get the paws breaking potential of https://gerrit.wikimedia.org/r/434356 [03:00:34] dev-requirements.txt would be deleted [03:00:59] so `/srv/paws/bin/pip install --no-cache-dir -r /srv/paws/pwb/dev-requirements.txt` will error out [03:02:09] chicocvenancio: ^ [03:02:10] Makes sense, we can remove that from paws [03:02:19] ok [03:03:00] we should probably remove that part that deals with requirements.txt as well [03:03:31] requirements.txt is also deleted in that patch and that sed will fail [03:08:16] I'll have to check the deal with Pillow [03:08:57] But there should be a cleaner solution to that with the patch [03:26:06] diverdude, i've even used xdotool and wmctrl to automate that from a make file run from vim before. whatever makes things simple. i'm coding something for avr and using the arduino as a demoboard the arduino ide, core, usually aren't involved at all. [03:26:13] ohh boy. [03:26:18] i've heard people got rich with crypto [03:26:23] Got it, I think. [03:26:30] hmm tells me that I need to --force-reinstalls, and that will break some packages. It should just fix/upgrade the packages that would be broken right? [03:26:34] N with 0 vs without are very different. [03:30:17] Butterfly^: Well, he's burned bridges here, for sure. [03:30:21] ppf: well I'm pretty sure you've got the issue I'm thinking of. I wonder if it's an ACPI problem. There's a potential workaround for that if so. Read http://iam.tj/prototype/enhancements/Windows-acpi_osi.html [03:43:08] othias: As far as I remember it is (condition) ? (true) : (false) [03:43:14] [@kevinmarks] ↩️ come along to IndieWebCamp New York - people there will talk about the kind of tech culture you are asking for, and they can probably help you fix your blog too https://indieweb.org/2018/NYC (http://twtr.io/1iLdJ5Y6a5b) [05:34:26] I broke my asm into different files and boy did it feel good when I got done. [05:34:32] boa tarde [05:37:41] zim: those are the types of the arguments to the function [05:37:46] might have been improved since [05:53:13] Well, except that if the list is [a, b, c, d, e] and (foo c), then I only need (all bar [a, b, d, e]). In that case, I don't car about (bar c) [05:53:17] tried PRIME, does not seem to do anything.. [05:59:39] dStruct: try starting by forming a support question that can be answered without referring you to a dentist. [05:59:44] In my case I think I've mostly used it to allow users to talk to their arduino devices. [06:09:34] writing the value of the variable to cout [06:09:37] password8: I know right [07:00:38] johnjay: You'll probably have a better time finding help in ##c++-general [07:00:46] Like if I misspell something. [07:00:54] usually i use 1 tab, to tab over the one big where clause. by tabs i mean spaces [07:01:01] asymptotically I'll give you 3 shitcoins to peruse my selfie collection ;) [07:01:08] CircadianRebel: oi [07:01:15] can sourceforge be safe u think [07:03:43] darkmeson: I think you mean those pesky vampires [07:03:50] if I did, I'd be making it [07:03:58] seems a bit overkill though [07:15:33] okay, yeah I was assuming I would use the same nick across all servers [07:15:38] semen is high protein though [07:15:41] smallville7123, sudo dnf whatprovides */elf.h --disablerepo=rawhide-modular [07:19:15] samwilson: the k8s workers seems to use a different exum4 config (looks like production?) and for some reason listens on * [07:20:01] it may or may not work (if it works it may break any time), so /me said nothing :) [07:54:36] darkmeson: Nah, it's pointless. [08:26:07] so it might make sense to just avoid it all together [08:26:15] s&p all the way imo [08:26:20] I/O maybe but i think it says something else, maybe typo [08:26:27] I don't want arm unless windows goes too [08:26:32] in that regard, aac is one of the better choices [09:56:35] hungry [09:56:42] Can you use uCurrent Gold with you Oscilloscope instead of Multimeter? [09:56:50] dadada: whatever you say, but I don't see the point of it [10:04:02] (03PS1) 10D3r1ck01: Add Pywikibot repo for indexing/search [labs/codesearch] - 10https://gerrit.wikimedia.org/r/463719 [10:16:56] jaroslaw: custom argument parsing. [10:17:00] I studied in France, very different from what I knew... [10:17:10] Lilium: hi, I'm afraid you have to ask wikipedia folks that, as we are not affiliated with that bot or process at all [10:17:15] Oh, I thought he was trying to do a dual boot. [10:24:03] [itchyjunk]: a common use topology on some set is as a prototypical notion of a 'space' [10:24:08] eia1x: by "tried fixing" you mean using the --fix-missing option? [11:24:10] rq-4 [11:24:17] sibiria: falstad seems to be very cool. Thanks for this link. Is there a way to show "how much REAL time there is between the button press and the logic change" ? [11:24:21] empoz_: i3 (source: i3-wm): metapackage (i3 window manager, screen locker, menu, statusbar). In component universe, is optional. Version 4.14.1-1 (bionic), package size 2 kB, installed size 21 kB [11:24:30] So it seems like that could go on forever. [11:24:37] when it first came out [11:51:00] no, just dc chopping and dc non stop [11:51:07] https://pastebin.com/YT29ZMJv [11:51:11] no_gravity: trying to avoid ex(?) mode for that, thank you (: [11:52:45] make it stop [11:52:53] lieven: if there is “twitch plays pokemon” and “twitch installs arch linux” why couldn't there be “twitch proves the riemann hypothesis” after all [11:53:02] there is no reason to ever have untagged traffic on a trunk [11:53:10] And then they go to Coq and show how Coq does 1/0 = 0 and because Coq is math, that means math agrees that 1/0 = 0. [11:53:16] but was video editing that bad in former times [11:53:24] guiverc: that's it - I knew there were some :) [12:53:42] just use C [12:53:48] it is not [12:55:17] Compu-Celebi: debian believes linux is like a car that you are free to put together. they provide the pieces https://itvision.altervista.org/files/linux-car-kit.jpg [12:55:21] works from a script [13:33:11] Just trying to configure a proxy and security groups, having an issue getting it set up right, when I noticed that my instance doesn't appear to have an IP, but every other one I checked did. Is that an issue? https://tools.wmflabs.org/openstack-browser/server/hashtags-prod.hashtags.eqiad.wmflabs [13:33:50] In the Horizon instance list the instance should have an IP of 172.16.1.222 [13:47:27] Samwalton9, might be a bug in openstack-browser not recognising the eqiad1-r IPs [13:47:54] Ah, that could be it. Just noticed some other 172 IP projects have the same thing [13:48:35] yeah they're still pretty rare [13:49:17] it might be that you can't get them from the nova API directly anymore? not sure [13:54:21] Samwalton9, yeah instead of there being a 'public' key in server.addresses there is 'lan-flat-cloudinstances2b' [13:54:53] that'll screw up the bit of templates/server.html that looks in server.addresses [13:55:04] still it's not a gerrit repository so I'll just open a task [13:57:08] https://phabricator.wikimedia.org/T205869 [13:58:02] Alright :) [13:58:38] Side note, I think I just locked myself out of SSH access to my instance while trying to set up the security groups but I can't figure out what I did... [14:00:28] Samwalton9, are you still permitting SSH in to port 22? [14:01:10] Believe so, I've got a TCP - 22 - 0.0.0.0/0 rule [14:01:15] yes [14:01:24] but it doesn't respond? [14:01:39] I'm getting permission denied when trying to connect [14:01:49] well that's not going to be a network problem is it [14:01:51] I can connect to another project no problem [14:02:53] do you have a particular SSH config for the other project? [14:03:15] I thought trump surrendered again [14:03:22] yay! i just compiled lmsensors from github, changed (for ubuntu distro) only the PREFIX in Makefile (from /usr/local --> /usr), and did sudo make install [14:03:28] both ? [14:03:36] I have the same config for both, and could connect fine earlier, having not changed it [14:04:36] it is possible to lock yourself out of an instance by mucking around with security config in there [14:04:48] was puppet working on the instance? [14:06:27] if so you could stick a "passwords::root::extra_keys" hiera config in and get in as root and sort it out [14:06:42] hmm. I'd only been changing things in Horizon, and primarily just the security groups and proxy, but I'm pretty new to this so I wouldn't put it past me to have screwed something up. [14:06:58] if not you could ask ops to try getting in [14:07:12] somehow I doubt you would've managed to cause yourself a permission denied error [14:07:50] by fiddling with horizon [14:08:00] there are some specific things you could've done I guess but... unlikely [14:08:21] Samwalton9, can you paste the whole output of ssh -vvv hashtags-prod.hashtags.eqiad.wmflabs ? [14:11:09] Krenair, https://pastebin.com/wsUFpNLz [14:21:30] Samwalton9, wait are you samwalton or samwalton9 in LDAP? [14:25:03] !log mw-extension-ids deleting project as per https://wikitech.wikimedia.org/wiki/News/Cloud_VPS_2018_Purge#OBSOLETE_mw-extension-ids [14:25:05] Logged the message at https://wikitech.wikimedia.org/wiki/Nova_Resource:Mw-extension-ids/SAL [14:27:24] andrewbogott, any chance you can try logging into hashtags-prod.hashtags.eqiad.wmflabs ? [14:27:46] sure, trying [14:27:52] just to check whether logging in works for ops at least [14:28:40] so far it's hanging for me, with personal and root key both. A security group thing maybe? [14:28:43] uh [14:28:49] that's interesting [14:29:04] I double checked, I can log into at least one other VM [14:29:11] so my network isn't totally messed up :) [14:29:24] ok, after a long wait I got Permission denied (publickey). for my personal key [14:29:28] still waiting on the root key... [14:29:32] krenair@bastion-01:~$ ssh hashtags-prod.hashtags.eqiad.wmflabs [14:29:32] Permission denied (publickey). [14:29:41] and now I'm in [14:29:45] Krenair, I think I have both, but for this samwalton9 [14:29:57] double-checking with root key... [14:30:07] andrewbogott, remind me, does the ops group get you past the project membership check? [14:30:16] it doesn't [14:30:19] but a root key does [14:30:23] right [14:30:34] and you're not in that project [14:30:41] yeah [14:30:41] so this is expected behaviour so far [14:30:50] Except it's so slow :) [14:30:57] ew [14:30:59] nothing obviously broken on that VM [14:31:04] I just tried again and it's become slow? [14:33:06] (this is an eqiad1 VM btw) [14:33:32] this project is just called 'hashtags' right? [14:33:52] yes [14:34:00] something is weird [14:34:01] $ OS_TENANT_NAME=hashtags openstack server list [14:34:06] [14:34:14] did you set OS_REGION_NAME ? [14:34:21] Oh, good point [14:34:24] eqiad1 :) [14:34:35] ok there it is [14:40:58] Krenair: I'm not seeing that same crazy delay for other things in eqiad1 [14:41:14] Samwalton9, how much work have you done on this VM so far? [14:41:58] It's tempting to suggest getting a new VM [14:42:18] worth trying a reboot first at least [14:42:25] It wouldn't be a big deal to get a new one, I have more-or-less everything in a docker container I can just spin up again. [14:45:28] yeah [14:45:41] could also just try a small fresh one to check SSH access problems against [14:46:58] Just created a new instance in the same project and getting the same issue [14:48:15] Also, just double checked ssh from another computer and same permission denied [14:49:54] but other projects work? [14:50:03] which other projects? [14:50:09] Yep, connected to twl with no problem [14:50:27] VM can't connect to the outside.. so no DNS, can't reach LDAP server, etc.. [14:51:23] hm [14:51:33] would explain why it's slow to deny access I suppose [14:51:56] Samwalton9, you said you fiddled with security groups [14:52:10] In eqiad1 there are outbound rules as well as inbound [14:52:15] so probably an egress rule is broken [14:52:31] it still permits outbound access to everything right? [14:52:46] only if there are rules allowing it [14:53:06] Samwalton9: I'm going to add myself to that project so I can see what's going on [14:53:52] it doesn't seem to be the VM's iptables that's causing it (flushed rules, stopped docker... no go -- started docker again, rules are all back) [14:54:44] it can't ping the gateway (172.16.0.1) [14:55:23] well that's broken [14:55:32] I can ping that IP from eqiad1.bastion.wmflabs.org FWIW [14:55:47] so I imagine it's an outbound security rule on that instance? [14:56:05] My having broken something connection-wise is quite likely, though I think I was only modifying security groups. Having a 'reset default to default' option would be nice, FWIW :D [14:56:16] 'default security group to default' I mean [14:56:25] #upstream [14:56:39] the network being used for docker is weird [14:57:35] ah nevermind [14:57:41] ok, sorry, I need to duck out to work on another thing now, sorry. I'd recommend adding Krenair to that project so he can check out the egress rules [14:58:03] It's possible that the default egress rules set up during migration are wrong, although if so I'd expect to have seen that elsewhere [14:58:08] Thanks for the help andrewbogott :) [14:58:54] my cn is Alex Monk [14:59:36] Added you to the project [15:02:34] ok, I'm back :) There are no egress rules AT ALL for that project, which will definitely break things. [15:02:48] Samwalton9: I'm hoping you deleted them and it's not that my migration broke them… any idea? [15:03:11] Much more likely that I deleted the wrong thing :) [15:03:26] was that project migrated? [15:03:53] or is this a green field eqiad1-r project? [15:04:32] I don't remember [15:04:50] Looks like it's been around for ages [15:05:04] I think this was created directly on eqiad1-r; I had to wait a little while after requesting creation so that eqiad1-r was available [15:05:18] hm, ok [15:05:22] https://phabricator.wikimedia.org/T204059 [15:05:33] okay [15:05:37] Try logging in now Samwalton9 [15:05:47] I just added egress rules, TCP/UDP/ICMP to 0.0.0.0/0 [15:05:54] I'm in :) [15:05:55] to the default security group [15:05:57] it seems to work for me [15:06:05] Did everyone's firefox stop having a 'reload' button today, or is that just me? [15:06:23] I've got a reload button [15:06:25] Krenair: those new rules look reasonable to me [15:06:53] Krenair: so you've got <- -> right in a row, right? [15:07:00] To the left of the nav bar? [15:07:07] Thanks all for unbreaking :) [15:07:16] <- -> [15:07:27] well, what the heck [15:07:42] FF62 [15:08:12] me too [15:08:23] andrewbogott, is it worth checking eqiad1 default security groups on other projects? [15:08:36] yes [15:08:48] okay well [15:08:52] I don't think I can do that as novaobserver [15:09:10] Samwalton9, feel free to remove me from the project if you don't want me anymore :) [15:09:27] Random spot check (project 'design') has any protocol any port 0.0.0.0/0 [15:09:30] so that seems good [15:10:03] yeah, second check has that too [15:10:25] so it has good defaults but it's still seems like too much rope IMO [15:10:52] Krenair, well... do you have 5 mins to help with the issue I was working on when I broke this? Namely that I have a tool running, local connections working perfectly, but I'm struggling to get access via hashtags.wmflabs.org working. I'm pretty new to this and I'd appreciate being pointed in the right direction. [15:11:03] Samwalton9, heh ok [15:11:10] I can help with webproxy stuff, I don't know anything about docker [15:11:14] Thanks :D [15:11:34] I have a web proxy set up (hashtags.wmflabs.org -> http://172.16.1.222:80) [15:12:13] ok [15:12:35] looks like docker is listening on port 80 [15:12:41] your IP is 172.16.1.222 [15:12:46] so... so far so good [15:13:22] hm [15:13:26] but the name doesn't resolve [15:14:01] uh oh [15:14:04] andrewbogott [15:14:22] Yeah as far as I can figure out I either don't have my security groups set up properly, or the docker/nginx/gunicorn parts aren't correct. [15:14:28] alex@alex-laptop:~$ dig hashtags.wmflabs.org @labs-ns0.wikimedia.org +short [15:14:28] alex@alex-laptop:~$ dig hashtags.wmflabs.org @labs-ns1.wikimedia.org +short [15:14:28] alex@alex-laptop:~$ dig hashtags.wmflabs.org @labs-ns2.wikimedia.org +short [15:14:28] 208.80.155.156 [15:14:29] alex@alex-laptop:~$ dig hashtags.wmflabs.org @labs-ns3.wikimedia.org +short [15:14:30] 208.80.155.156 [15:14:40] the webproxy thing has only put the A record in eqiad1 DNS? [15:14:40] Samwalton9: how recently did you set up that proxy? [15:14:55] I set it up a while ago [15:14:57] Like, last week [15:15:02] ok, looking [15:15:04] Actually I might have recreated it earlier today [15:15:30] So a couple of hours ago at least [15:16:00] Created At [15:16:01] 2018-09-20T16:34:58.000000 [15:16:45] hmm [15:17:07] horizon shows it existing in the wmflabs.org. zone regardless of whether I select eqiad or eqiad1-r as the region [15:17:29] yeah, DNS should be identical between the two regions [15:17:33] "should" [15:18:16] for the dns service, eqiad has http://labservices1001.wikimedia.org:9001, eqiad1-r has http://cloudservices1003.wikimedia.org:9001 [15:18:40] this is going to turn out to be something to do with ipv6 firewalls I bet [15:19:03] so they both know about the zone but the record hasn't made it through to pdns in eqiad? [15:19:13] seems like [15:26:41] (03CR) 10Jean-Frédéric: [C: 032] Handle empty report [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461593 (https://phabricator.wikimedia.org/T204581) (owner: 10Lokal Profil) [15:27:11] Is that fixable? [15:27:35] Samwalton9: I'm working on it but the problem is not obvious yet [15:27:42] Ah, thank you :) [15:29:14] Krenair: actually, it looks like a database getting overloaded. So nothing closely connected with the designate setup [15:29:46] the eqiad designate database getting overloaded? [15:29:49] (03CR) 10jerkins-bot: [V: 04-1] Handle empty report [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461593 (https://phabricator.wikimedia.org/T204581) (owner: 10Lokal Profil) [15:29:53] I guess the entire m* cluster it's on? [15:30:18] TRILLION [15:30:23] lcd.print("Push the buttons"); // print a simple message [15:37:11] Krenair: there are connection limits on the databases, I suspect that now that we have two regions we're hitting limits since we have twice as many services running... [15:37:17] Anyway, discussing with a DBA now [15:37:32] ok [15:37:55] No guarantee that this is the problem but it's at least /a/ problem [15:38:34] yay computers [15:40:50] brb [15:49:02] (03CR) 10Jean-Frédéric: [C: 032] "recheck" [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461593 (https://phabricator.wikimedia.org/T204581) (owner: 10Lokal Profil) [15:52:09] (03CR) 10jerkins-bot: [V: 04-1] Handle empty report [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461593 (https://phabricator.wikimedia.org/T204581) (owner: 10Lokal Profil) [15:52:12] (03Merged) 10jenkins-bot: Handle empty report [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461593 (https://phabricator.wikimedia.org/T204581) (owner: 10Lokal Profil) [15:53:44] (03CR) 10jenkins-bot: Handle empty report [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461593 (https://phabricator.wikimedia.org/T204581) (owner: 10Lokal Profil) [17:23:21] Maybe, although people keep reminding me that I don't exist in the physical world so that might be tricky. [17:23:27] consider acquiring hardware by a vendor which support open source next time. [17:30:24] isnt x299 the worst hedt platform that intel created? [17:30:32] BCMM yes [17:47:54] iCookie: do i get to keep the mercury? [17:47:59] YADW: in the blame list "19.599s dev-sda5.device" is worrying; suggests a drive problem [17:48:03] :1:30: error: parse error on input ‘if’ [17:48:10] and executable1 worked fine [17:48:16] 「Crime rate in Germany at lowest level in 30 years, but antisemitism is on the rise | The Independent」 [17:57:36] (03PS1) 10Urbanecm: Add "do not fix" comment in tox.ini [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463812 [17:57:54] (03CR) 10Urbanecm: [C: 032] Add "do not fix" comment in tox.ini [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463812 (owner: 10Urbanecm) [17:58:19] (03Merged) 10jenkins-bot: Add "do not fix" comment in tox.ini [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463812 (owner: 10Urbanecm) [18:05:57] (03PS1) 10Urbanecm: Typo in README.md [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463814 [18:06:15] (03CR) 10Urbanecm: [C: 032] Typo in README.md [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463814 (owner: 10Urbanecm) [18:06:33] (03Merged) 10jenkins-bot: Typo in README.md [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463814 (owner: 10Urbanecm) [18:08:57] (03PS1) 10Urbanecm: Add a "do not fix" comment to tox.ini for "#indentation contains tabs" [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463815 [18:10:22] (03CR) 10Urbanecm: [C: 032] Add a "do not fix" comment to tox.ini for "#indentation contains tabs" [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463815 (owner: 10Urbanecm) [18:10:43] (03Merged) 10jenkins-bot: Add a "do not fix" comment to tox.ini for "#indentation contains tabs" [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463815 (owner: 10Urbanecm) [18:12:41] (03PS1) 10Urbanecm: Remove fixed rules [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463817 [18:12:55] (03CR) 10Urbanecm: [C: 032] Remove fixed rules [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463817 (owner: 10Urbanecm) [18:13:05] (03PS1) 10Urbanecm: Add "do not fix" comment to "expected 2 blank lines, found 1" [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463818 [18:13:13] (03Merged) 10jenkins-bot: Remove fixed rules [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463817 (owner: 10Urbanecm) [18:13:21] (03CR) 10Urbanecm: [C: 032] Add "do not fix" comment to "expected 2 blank lines, found 1" [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463818 (owner: 10Urbanecm) [18:13:40] (03Merged) 10jenkins-bot: Add "do not fix" comment to "expected 2 blank lines, found 1" [labs/tools/wikinity] - 10https://gerrit.wikimedia.org/r/463818 (owner: 10Urbanecm) [18:14:17] They do indeed have some similarities, but I am not sure whether it would make sense to take both of these courses. The study advisors cannot answer such questions, since these are courses from different universities and they do not really know about math, but only about rules and formalities. [18:14:25] overlooking something, type, etc [18:14:45] and runs these as commands at appropriate lines [18:14:52] they are software programs [18:19:24] (03CR) 10Lokal Profil: "recheck" [labs/tools/heritage] - 10https://gerrit.wikimedia.org/r/461594 (https://phabricator.wikimedia.org/T204580) (owner: 10Lokal Profil) [18:26:59] !help multiple users are reporting that toolforge ssh is extremely slow right now [18:26:59] WikidataFacts: If you don't get a response in 15-30 minutes, please create a phabricator task -- https://phabricator.wikimedia.org/maniphest/task/edit/form/1/?projects=wmcs-team [18:27:46] hm, seems to be better already as I write this… reports were 20-5 minutes ago [18:30:07] ok, please let us know if the issue returns. I just checked it and it seems okay now [18:30:15] ok, thanks [18:37:15] is there a tag that does it automatically [18:37:18] very specific [18:38:07] gtirloni: for future reference, usually when the toolforge bastion is slow it means people are running jobs on the bastion even though they're not supposed to [18:38:45] ok [18:39:09] I saw some processes in `htop` in the “D” state (blocked on disk), but they went away soon after I sent my first message [18:39:22] and htop truncates the username so I can’t tell you the full tool name :/ [18:39:47] yeah, anything involving disk I/O is going to be painfully slow on bastion because of NFS [18:44:10] (03PS1) 10Urbanecm: Add "do not fix" comment to "expected 2 lines" error [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463826 [18:44:21] (03CR) 10Urbanecm: [C: 032] Add "do not fix" comment to "expected 2 lines" error [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463826 (owner: 10Urbanecm) [18:44:37] (03Merged) 10jenkins-bot: Add "do not fix" comment to "expected 2 lines" error [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463826 (owner: 10Urbanecm) [18:46:47] So either the size of the structure is bounded in some arbitrary way (making all the operations O(1)), or the range of the lookup function is somehow restricted by holding values of a fixed finite type [18:46:56] ...... 5000€ o.0 !!! [18:48:09] (03PS1) 10Urbanecm: Remove fixed rule [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463828 [18:48:22] (03CR) 10Urbanecm: [C: 032] Remove fixed rule [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463828 (owner: 10Urbanecm) [18:48:44] (03Merged) 10jenkins-bot: Remove fixed rule [labs/tools/map-of-monuments] - 10https://gerrit.wikimedia.org/r/463828 (owner: 10Urbanecm) [18:55:58] Elec_a: snap packages are auto-updated the snap process [18:56:02] a____ptr, while it can certainly depend on the class and how it is taught, you also took it before you spent countless hours studying math [18:56:10] I think you'd actually hit every rate limit [18:56:16] I use an ISP called PlusNet. [18:56:22] maximum of convex function is still unique [19:10:41] does the language get really different with other compilers? [19:10:51] ∈ ε [19:10:58] saifu. see? [19:11:08] average bitrate [19:15:50] (03PS1) 10Urbanecm: Add instructions on how to setup local development environment [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463831 [19:15:52] (03PS1) 10Urbanecm: Do not hardcode path to API in JavaScript assets [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463832 (https://phabricator.wikimedia.org/T195848) [19:16:16] (03CR) 10Urbanecm: [C: 032] Add instructions on how to setup local development environment [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463831 (owner: 10Urbanecm) [19:16:40] (03Merged) 10jenkins-bot: Add instructions on how to setup local development environment [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463831 (owner: 10Urbanecm) [19:20:05] (03PS1) 10Urbanecm: Better instructions regarding OAuth creds [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463834 [19:20:48] (03CR) 10Urbanecm: [C: 032] Better instructions regarding OAuth creds [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463834 (owner: 10Urbanecm) [19:21:06] (03CR) 10Urbanecm: [C: 032] Do not hardcode path to API in JavaScript assets [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463832 (https://phabricator.wikimedia.org/T195848) (owner: 10Urbanecm) [19:21:45] (03Merged) 10jenkins-bot: Better instructions regarding OAuth creds [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463834 (owner: 10Urbanecm) [19:21:46] (03Merged) 10jenkins-bot: Do not hardcode path to API in JavaScript assets [labs/tools/commons-mass-description] - 10https://gerrit.wikimedia.org/r/463832 (https://phabricator.wikimedia.org/T195848) (owner: 10Urbanecm) [20:30:31] Samwalton9: I can't reproduce that proxy issue; I think it might have happened because there was a DB issue at the time. I'm going to delete and recreate it and see if that fixes things. [20:34:44] ok, should be fixed [20:34:56] looks like ns0 has it [20:35:01] nothing from ns1 though [20:50:33] andrewbogott, http://hashtags.wmflabs.org/ is alive! [20:50:38] * Samwalton9 does a little dance [20:50:48] Thank you :) [20:50:55] great! [20:51:45] Hi, im wondering how do i create a group in ldap? (im using mw vagrant) [20:53:54] bd808 might know? [20:54:47] n2: but a nice screen and good controls [20:54:50] paladox: in a local LDAP? Just by issuing the appropriate commands to the directory (like by applying an LDIF) [20:55:00] bd808 yep [20:55:17] ldapmodify keeps asking for a password [20:55:49] oh [20:55:51] oh, yeah it needs a password. The password should be in the hieradata for the role I think [20:55:54] i think i found the password [20:56:55] heh. I was very creative with the default passwords -- https://github.com/wikimedia/mediawiki-vagrant/blob/master/puppet/hieradata/common.yaml#L410-L412 [20:58:04] heh [21:01:30] I can only imagine how many warnings a default Active Directory install would have :) plus, can't use last 32 passwords [21:04:20] hmm the new group is not showing in gerrit. [21:04:49] CrystalMath: you lose IP rights, of course it has consequenes. [21:04:53] only $650? [21:05:29] paladox, do LDAP groups show up by default in gerrit? [21:05:38] Krenair they should [21:05:55] they don't in prod do they? [21:06:04] if configured correctly in the config [21:06:05] i had to hack mine not to use prod [21:06:06] Krenair they do [21:06:07] just based on visability i guess [21:06:41] so you're saying that if logged in as an admin, https://gerrit.wikimedia.org/r/#/admin/groups/ would list gerrit groups? [21:06:57]