freenode/#devuan/ Thursday, 2019-03-07

* roo^y added extension 'Grammar and Spell Checker - LanguageTool' to the Firefox in Refracta 9, as it makes him look slightly less of a moron :D02:54
roo^yi may have over-looked 'Grammarly'. I've taken for granted pre-added addons for browsers03:00
roo^y*Actually, Under FF Preferences, Languages, Choose your preferred language for displaying pages, the box is ticked for 'Check your spelling as you type'. Maybe I should've restarted FF to update it, to see if the current version has a bug before resorting to a addon :P03:06
roo^y(I looked in preferences after a web search taught me FF has it's own built in spelling check)03:08
gnarfaceroo^y: i've had an issue where some sites' form fields will somehow unset the spell check box, but if i right-click on the field itself and re-enable it there manually it persists for the rest of the session03:40
gnarface(i still haven't bothered to try to figure out whether that is a bug in firefox or a bug on the site because i have the depressing suspicion it is both)03:40
roo^yI see. I reported an error on a payment page, only for it to be NoScript again, blocking a service03:45
roo^yon the flip side, my lack of using 'an' instead of 'a', is being corrected with the new extension :P03:47
BeerbelottHow does one recycle pseudo-terminals attached t odead processes?12:41
Beerbelott(apart from reboot)12:42
BeerbelottLike that:            pts/0        2019-03-01 13:15             26476 id=ts/0  term=0 exit=012:42
guido_gtry: ps -f -t pts/013:11
msiismsince yesterday, i'm having a strange problem with freshclam. when i boot my system, it starts running and then takes up 100% of one of my 2 CPUs and doesn't stop doing that. i've stopped the process and re-run it with verbose output, which is giving me no clues, however: (this is also all that gets written to freshclam's log file). what could be the problem?13:14
gnarfacesounds like normal behavior for an anti-virus program to me, but i don't really know for sure13:15
gnarfacemaybe try turning off some features to see what happens13:16
msiismgnarface: well, i've been running lcamav for years and have never seen that.13:16
gnarfaceoh, well something could have gone wrong too13:16
msiismmy system is also up to date, so this can't be it, i guess.13:16
gnarfaceis there any disk i/o while this is happening?13:17
msiismnot that i can see.13:18
msiismit seems to just stop.13:18
msiismwell, "stop"13:18
msiismthe last time i had it go on for up to 18 minutes before i killed the process.13:18
sixwheeledbeastdmesg or syslog show anything?13:20
msiismoh right, let me check.13:20
gnarfacedaemon.log might have stuff too13:20
sixwheeledbeastI have had programs do similar if there is no space13:21
msiismthe partition this is on show as being filled 66%, so that should be ok. ram usage is at 34%.13:21
gnarfacecould it be permissions related?13:23
msiismhm... searching syslog and daemon.log for anything related to clamav doesn't reveal anything.13:23
msiismgnarface: well, i didn't change anything.13:23
gnarfacenothing at all?13:24
msiismi mean, it's not like i'm working with clamav intensiveley. it just sist there in case i feel like checking some one else's usb key.13:24
sixwheeledbeastMaybe it's detected a file it is struggling to process.13:26
xinomilofreshclam or clamd ?13:26
gnarfacewould it be unpacking archives?13:26
gnarfaceto scan file contents?13:26
Beerbelottguido_g: ps says TTY could not be found. The man says tty need to be specified either as /dev/ttyS1, ttyS1, S113:27
msiismxinomilo: freshclam13:28
sixwheeledbeastAre you sure you have the latest version if the logfile says you are not?13:29
msiismgnarface: i don't know.13:29
gnarfacethat's a good point, you say you're up-to-date, but what mirror URL are you using in your sources.list?  the auto.mirror one got deprecated at some point13:30
msiismsixwheeledbeast: well, not having the latest version is the normal case on devuan (as on debian). maybe that's beginning to cause a problem here.13:30
xinomilothis outdated version is normal in debian/devuan13:30
gnarfaceoh ok13:30
msiismgnarface: i'm using, which i should probably change, but not because it's out of date.13:31
guido_gBeerbelott: works here, try it with existing pts13:32
guido_gBeerbelott: if there is nothing on the pty, it should be freed by the kernel13:32
xinomilomsiism: from the link i don't see manual run of freshclam stopping at somepoint, does it stay stuck in "Downloading daily-25380.cdiff [100%]" ?13:33
xinomilomaybe permissions on /var/lib/clamav or some network failure?13:33
sixwheeledbeast ?13:34
msiismxinomilo: yes, this where it stays.13:34
msiismsixwheeledbeast: thanks.13:35
sixwheeledbeast"There is definitely some bug in today's ClamAV update. I have the same problem on all servers running ClamAV (different distributions, different versions)."13:35
xinomilothere's a workaround in #15 message in bug report.. try that13:36
msiismand then "It does finish eventually and shouldn't repeat." :)13:36
Beerbelottguido_g: It works on existing pts/013:36
BeerbelottI already know it's dead as it's reported by who -d13:36
msiismxinomilo: yes, seeing that. i'll try that, i guess.13:37
sixwheeledbeastJust leave it running13:37
BeerbelottThe question is... why is it not recycled13:37
msiismsixwheeledbeast: ok, then i'll try the workarund when it doesn't finish within the next hours.13:37
xinomiloyou should see "daily.cld is up to date (version: 25381, sigs: 1504578, f-level: 63, builder: raynman)" after success..13:38
msiismxinomilo: ok.13:38
gnarfaceoh so it's doing unpackaged inline updates?  so something DID change....13:38
gnarfacei wonder if it would be possible to temporarily roll it back13:38
errandir1Beerbelott: It is not recycled because there are still processes using it. Dead processes are still processes and they still hold counters & locks.13:39
Beerbelotterrandir1: There is no process reported by ps w/ the PID reported by `who -d`13:39
sixwheeledbeastFreshclam gets the new "virus" files inline13:40
Beerbelotterrandir1: Is there a way to trace that dead process?13:40
errandir1use lsof to see which process is using /dev/pts/013:43
msiismok, freshclam is done.13:57
msiismtook about an hour, as it seems.13:57
Beerbelotterrandir1: Ohhh... stale docker container tty allocations...13:57
BeerbelottThx a ot :)13:58
Beerbelottlsof... that's a basic thing I forgot :-13:58
xrogaanbe wary with chromium:
xrogaan0 day security breach touching the FileReader interface. Might allow a third party access to the local file system (IIUC)15:03
rafalcppxrogaan: where do you see this description of this bug?15:16
xrogaan>  Google is aware of reports that an exploit for CVE-2019-5786 exists in the wild.15:19
rafalcppseurity tracker on debian is rather useless15:20
fsmithredis it a feature?15:20
rafalcppxrogaan: tl;dr radom web page can read files from your disk?15:20
xrogaanrafalcpp: not really, it's purpose is to display across debian's versions. Fixed in sid.15:21
xrogaanI assume it is so, yes.15:21
FlibberTGibbetnice to see that Devuan gets a mention from the guy who wrote "FreeBSD Mastery: Jails" :)15:22
xrogaanescape the sandbox kind of situation.15:23
FlibberTGibbetapparently he likes Devuan becuase it runs sanely inside a FreeBSD Jail when he needs to install something Linux-centric.15:33
xrogaanisn't ets a backdoored silly "secure" layer?15:42
heaven-backdoored? D:16:29
heaven-did someone create a backdoored crypto layer that nobody uses?16:29
xrogaanI don't know, I heard something silly about banks needing to inspect packets for security and so something less secure got created.16:57
DocScrutinizer05>>a backdoored crypto layer that nobody uses<< thanks to articles like the above linked one18:55

Generated by 2.17.0 by Marius Gedminas - find it at!