IRC Logs for #crux-devel Monday, 2015-01-05

*** Romster has joined #crux-devel01:42
*** mavrick61 has quit IRC03:29
*** mavrick61 has joined #crux-devel03:30
*** Claus_ has joined #crux-devel03:59
*** Feksclaus has quit IRC04:02
*** Claus_ has quit IRC04:52
prologicndn06:35
prologicerr06:35
prologicheh :)06:35
diverseprologic: we got a problem on our hands :(06:37
diversealso what's "ndn"?06:40
prologichmm?06:40
prologicndn == typoe :)06:41
diverseprologic: read what happen on #crux06:41
prologicI didn't see anything happen?06:41
prologicwhen?06:41
diversejust recently06:41
prologichmm06:41
diverseusers are being affected from doing a ports -u which cause them to loss ~200 ports and important ones too06:42
prologicoh06:43
prologicis that a serious problem?06:43
prologicdid someone actually remove ports from opt malciously?06:43
diverseapparently, because I had to provide someone a Pkgfile to a port that just gone missing06:43
diverseI have no idea, I checked the git history and nothing really stands out as "malicious"06:44
diverse2 users where affected06:45
diverse*were06:45
diverseI don't see where the 200 ports gone missing06:47
prologicanyway I'll check it out later06:48
prologicbbl06:48
prologicfamily  :)06:48
diversewhat perfect timing, no one else seems to be around atm06:49
diversesarcasm intended06:50
RomsterteK_, jaeger jue looks like opt rsync script aborted at libxvid. can you see why it did that and run the hook again? or maybe i'll just update something and push it so it runs it.07:10
diversethat sounds similar to what happened to leo on #crux07:11
Romsteri'll just update one of mine and git push07:12
Romstershould fix it.07:12
Romsteri fixed it guys with another git push07:26
prologicahh07:39
prologicthat's what I thought07:39
prologicrsync out of sync with git tree07:39
diverseprologic: why didn't you tell me this sooner -_-07:40
prologiccouldn't be sure :)07:40
prologicI have no access to crux.nu's server07:40
prologicother than git access07:40
Romstersame07:48
Romsterbut seeing the commit logs of opt its not hard to see it not all removed there.07:49
diverseI saw the commit logs, I just wasn't sure what was going on "behind-the-scenes"07:49
Romsterplus we have and sync issues before.07:50
Romsters/and/had07:50
prologicsome please try to resolve daisy.shortcircuit.net.au ?08:16
prologicsomeone*08:16
frinnstdaisy.shortcircuit.net.au has address 150.101.176.19408:21
frinnstdaisy.shortcircuit.net.au has IPv6 address 2001:44b8:2101:6801::208:21
Romstersame result here.08:25
frinnstromster that kmod-32 thingy, btw09:12
frinnstit should work as is. If not we need to find out why it doesnt09:12
Romsterprobably my setup09:14
Romsterdistcc/ccache but i can't see why that matters when its a gcc error09:15
prologicnice09:28
prologicthanks guys09:28
Romsterprologic, i find new dns names update fast it's the existing ones that can take up to 3 days in some dns name servers.09:29
frinnstwell configure should detect if it works or not09:29
Romsteri didn't look that deep at it yet09:29
jueRomster: you've read my comment about kmod? please think twice before you commit such things09:37
Romsteryes i was in a rush to build wine and i did test on more than one machine09:40
Romsterof course i got all my machines setup the same09:40
Romsterso same results...09:40
Romsternow i am sick of hearing it i get it all god damn ready09:40
Romsteri believe the git hook rsync script needs to check if it's still running and exit in a sane state before it runs again on two short time interval commits. or if that wasn't what caused opt.git to lose half it's ports... then that needs looking at.09:42
juefrinnst, jaeger: any objections to build openssl as of FS#1096?11:28
*** mavrick61 has quit IRC11:37
*** Lukc has quit IRC11:37
*** Lukc has joined #crux-devel11:38
*** mavrick61 has joined #crux-devel11:38
*** nrxtx has quit IRC12:24
*** nrxtx has joined #crux-devel12:34
*** nrxtx has quit IRC12:34
*** nrxtx has joined #crux-devel12:34
*** Feksclaus has joined #crux-devel13:58
jaegerjue: none from me13:58
frinnstnot at all13:59
frinnstI thought we had implemented it already14:00
jueno, not yet, will commit it later today14:02
frinnstI should probably start to reply to bugs more. When i first saw the ticket i just though "yeah, thats a good idea"14:03
teK_jfyi: that's the two curves deemed "insecure" by several crypto experts ;)14:05
jueteK_: oops, do you have some refs for that?14:08
teK_http://crypto.stackexchange.com/questions/10263/should-we-trust-the-nist-recommended-ecc-parameters14:09
teK_and, if you like, from djb: http://cr.yp.to/talks/2013.05.31/slides-dan+tanja-20130531-4x3.pdf14:10
juethat's strange, because at least debian and RH compiling openssl that way14:10
teK_my understanding was that there's no official alternative yet14:10
teK_"Das NIST traut ja den eigenen Kurven nicht mehr."14:11
teK_http://www.heise.de/security/artikel/Nach-Snowden-Wenig-Schlaf-fuer-Kryptoforscher-2392236.html14:11
jueteK_: so, what now? do you suggest to _not_ build openssl as of FS#1096?14:16
juesorry, have to run now, back in about 2 hours14:17
teK_no, I say include the acceleration14:17
teK_my guess is that it will some time for NIST to aannounce new curves..14:17
teK_I wanted to say: we are encouraging using suspicious crypto even m ore ;)14:18
teK_I was quite surprised to hear that TOR uses these, too14:18
teK_I switched from ecdsa to ed25519 for all my SSH keys (because of this and because I'm a djb-fanboy 8))14:19
teK_bye, jeu14:19
teK_jue14:19
teK_The "sales pitch" for 25519 is more: It's not NIST, so it's not NSA. Not speed  <<< well put :p14:22
jaegerheh14:23
teK_though it still is faster than ecdsa.14:25
teK_http://www.alexrad.me/discourse/why-rosyna-cant-take-a-movie-screenshot.html14:57
frinnstBinutils 2.25 released15:04
frinnstsomething to play with at home :)15:05
*** leo-unglaub has joined #crux-devel15:50
jueteK_: thanks17:32
teK_sure17:35
mechaniputerThe portdb webpage still only shows about half of the ports from opt.17:59
*** leo-unglaub has quit IRC18:05
*** leo-unglaub has joined #crux-devel18:17
teK_should be fixed in ~10 minutes18:30
teK_(for opt it is already fixed)18:31
teK_fixed already (jaeger, is that the correct word order?)18:31
jaegereither works in this case18:32
teK_kk th18:36
teK_x18:36
*** c0x has joined #crux-devel18:39
*** Workster has quit IRC20:10
*** Workster has joined #crux-devel20:29
*** leo-unglaub has quit IRC23:34

Generated by irclog2html.py 2.11.0 by Marius Gedminas - find it at mg.pov.lt!