00:02.03 | *** part/#devuan Oldmoss (~Oldmoss@46.165.242.166) |
00:06.27 | *** join/#devuan Striktarn (~Striktarn@unaffiliated/striktarn) |
00:12.16 | *** join/#devuan barbara (~barbara@2601:196:8500:8819:ba86:87ff:fed6:7f66) |
00:12.29 | barbara | im getting wifi drops with RTL8723BE is there a special firmware i need to use other than firmware-realtek? |
00:13.39 | *** join/#devuan Chanku (~Chanku@2601:704:2:a700:21b:77ff:fea3:4a99) |
00:15.06 | *** join/#devuan Killer0byte (~Killer0by@net-188-218-67-144.cust.vodafonedsl.it) |
00:16.13 | *** join/#devuan Besnik_b (~Besnik@ppp-94-66-221-218.home.otenet.gr) |
00:23.44 | *** join/#devuan infobot (~infobot@rikers.org) |
00:23.44 | *** topic/#devuan is Recent (2017-05-25): Jessie 1.0.0 stable release http://ur1.ca/qxaa5 || This is the Devuan https://devuan.org/ discussion channel (logged at https://botbot.me/freenode/devuan - with useful 'search') | Please take off-topic conversation to #debianfork | /msg chanserv info #devuan | !listkeys #devuan <foo> | Devuan Forum: https://dev1galaxy.org/ |
01:21.35 | *** join/#devuan Chanku (~Chanku@2601:704:2:a700:21b:77ff:fea3:4a99) |
01:38.27 | *** join/#devuan rdav (~rdav@61.181.148.122.sta.dodo.net.au) |
01:47.24 | *** join/#devuan Humpelstilzchen (erik@x4e32a70f.dyn.telefonica.de) |
01:49.04 | *** join/#devuan Drugo (~Drugo@62-11-1-95.dialup.tiscali.it) |
02:00.16 | *** join/#devuan Hoshpak (~Hoshpak@p200300C90BC5C500ED5346C8EB83FF37.dip0.t-ipconnect.de) |
02:18.43 | *** join/#devuan GoatAvenger (~GoatAveng@unaffiliated/goatavenger) |
02:42.29 | *** join/#devuan aristene (~aristene@177-179-211-40.user.veloxzone.com.br) |
03:03.55 | *** join/#devuan theclash (~luca@57.red-88-12-180.dynamicip.rima-tde.net) |
03:14.27 | DocScrutinizer05 | barbara: you generally need the exact firmware for the particular chip |
03:15.32 | DocScrutinizer05 | realtek sounds way too familar to believe they had hust one firmware and that fits the RTL8723BE |
03:15.38 | DocScrutinizer05 | just* |
03:16.23 | DocScrutinizer05 | I guess firmware-realtek must be a package with quite a number of firmwares in it |
03:17.11 | DocScrutinizer05 | otoh it's not really plausible the chip would accept a wrong firmware |
03:21.36 | gnarface | it's more likely there is a bug in the firmware |
03:21.51 | DocScrutinizer05 | yeo |
03:21.54 | gnarface | or it's outright missing from that package version |
03:21.58 | DocScrutinizer05 | https://www.ixquick.com/do/search?q=RTL8723BE+is+there+a+special+firmware+i+need+to+use+other+than+firmware-realtek |
03:22.56 | gnarface | there were some common ones floating around i know from laptops only a few years old that were missing from a firmware package in jessie but present in the version in ascii, but i could have sworn that was the intel one... |
03:25.02 | gnarface | with realtek stuff though, often they have partial functionality available without firmware |
03:25.16 | *** join/#devuan rdav (~rdav@61.181.148.122.sta.dodo.net.au) |
03:44.01 | *** join/#devuan zyliwax (~zyliwax@unaffiliated/zyliwax) |
03:47.28 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
04:00.29 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
04:08.50 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
04:15.17 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
04:20.35 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
04:26.39 | *** join/#devuan guru (~guru@2601:196:8700:c91:3252:cbff:fec5:6f94) |
04:26.46 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
04:40.41 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
04:45.51 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
04:46.23 | *** join/#devuan poo-poo (~wigums@unaffiliated/wigums) |
04:57.05 | *** join/#devuan pav5088 (~pav5088@101.165.10.120) |
05:15.24 | *** join/#devuan tarbz2 (~Thunderbi@186.23.155.100) |
05:31.37 | *** join/#devuan mchasard (~mchasard@48.233.8.109.rev.sfr.net) |
05:44.39 | *** join/#devuan debdog (~debdog@ip-37-201-6-149.hsi13.unitymediagroup.de) |
06:05.54 | bozonius | gnarface: any ideas why my logwatch stopped sending mail? I've tried tracing the program directly, and it appears to be generating the report, but it does not mail it tome |
06:05.58 | bozonius | to me. |
06:06.08 | bozonius | It was working until about 5 or 6 days ago |
06:14.16 | *** join/#devuan Infiltrator (~Infiltrat@unaffiliated/infiltrator) |
06:24.38 | gnarface | bozonius: did you perhaps remove or change the configuration of the local mailserver (exim4 by default) |
06:25.49 | gnarface | bozonius: running out of space on the partition holding the mail spool could be a problem too. you checked your own user and root's mail, right? did you look to see if it makes it into the mail spool or the mail logs? |
06:28.58 | gnarface | bozonius: i can think of too many things that could sabotage it to narrow it down with so little information about your system. for example if you had your own user aliased to root to receive root's mail and the logwatch report was actually going to root, then you destroy that user to recreate it with a change to the spelling of the username, i doubt the mail system is smart enough to track that change automatically |
06:29.23 | *** join/#devuan peetaur (~peter@p57AAA0E1.dip0.t-ipconnect.de) |
06:30.05 | *** join/#devuan mchasard (~mchasard@48.233.8.109.rev.sfr.net) |
06:31.27 | gnarface | bozonius: if it gets as far as trying to actually mail it, i'd expect some mention of either success or failure in the logs unless it's a full partition issue and the logs are on the same partition as the mail spool (which is common actually) |
06:32.44 | gnarface | bozonius: if the /var partition is not full though, and there's no complaint in the mail logs, i'd expect something must be sabotaging logwatch's ability to even attempt to mail the generated report - like perhaps the report generation itself dies partway through maybe due to a permissions issue, like failure to save it to a temp directory before mailing or something... |
06:32.51 | bozonius | I installed dovecot on Jun 29; I never installed postfix |
06:33.03 | gnarface | hmm |
06:33.30 | bozonius | my /var is at 22% full |
06:33.36 | bozonius | this is a pretty new install |
06:33.40 | bozonius | of devuan |
06:34.04 | gnarface | hmmm |
06:34.18 | bozonius | exim4 is running |
06:34.38 | gnarface | could have something to do with the dovecot install changing some system users or permissions? i'm not using that or postfix, not familiar with them |
06:34.58 | bozonius | well, if that were the case though, why did logwatch work until Jul 11? |
06:35.17 | bozonius | (about 2 weeks w/o incident) |
06:35.23 | gnarface | something restarted it? |
06:35.46 | bozonius | I do have my own user aliased to pick up root's mail -- that has always worked. In fact, I ran a test just now, and it is still working |
06:35.53 | gnarface | ok |
06:36.06 | gnarface | and there's no stale mail in the mail spool for the root user or the mail user? |
06:36.13 | bozonius | from my regular user, "echo test mail from user | mail root" and "user" gets the mail |
06:36.27 | bozonius | how can I tell? |
06:36.50 | gnarface | look in /var/spool/mail |
06:37.12 | bozonius | just fyi, I don't recall changing my regular login user name... though I DID do that on a differnt system |
06:37.27 | bozonius | but not on this one |
06:37.53 | bozonius | I even tried removing and re-installing logwatch, to no avail |
06:38.25 | bozonius | only 1 file in /var/spool/mail and that is my regular user |
06:38.37 | bozonius | the regular user can read its mail |
06:38.45 | gnarface | and it doesn't have any unreceived system messages sitting in it? |
06:38.50 | bozonius | using, say, mail(1) |
06:39.10 | gnarface | also check /var/log/mail.* and /var/log/exim4/* |
06:39.21 | bozonius | well, there are a lot of messages sitting in it... I have my thunderbird configured to keep old messages for a month or so |
06:39.26 | gnarface | any successful or failed delivery should be mentioned in one of those mail logs |
06:39.32 | bozonius | ok |
06:39.51 | gnarface | by unreceived messages sitting in the mail spool, i just mean verify that there aren't any you haven't seen |
06:39.54 | bozonius | but if there are failures, it stands to reason they might not make it to the inbox! |
06:40.18 | bozonius | oh, I read them every day, pretty much. I try to keep on top of this stuff |
06:40.26 | gnarface | yea but if the failures aren't even logged, then something is failing before the mail even gets sent |
06:40.37 | bozonius | (right) |
06:40.54 | gnarface | at which point i'd be looking for syntax errors in the cron jobs |
06:40.54 | bozonius | let me look at those log files... |
06:41.52 | gnarface | syntax errors in the cron jobs should show up in uh... /var/log/syslog or /var/log/daemon.log i think |
06:45.55 | bozonius | but we don't need to go so far as cron even. If I run "logwatch --output mail" from the command line, no mail is sent |
06:46.08 | bozonius | (that's how it is invoked in the cron script) |
06:46.13 | gnarface | though there are certainly exceptions, excluding hardware failure this type of thing usually does keep working on it's own until a human interferes. so go back in your mind to the days leading up to the first failure. try to think of anything you might have changed, even if it wasn't apparent it would be related |
06:46.31 | bozonius | I've actually done that. |
06:46.53 | gnarface | dovecot sounds like the smoking gun to me but i agree that the timing is a bit of a stretch |
06:46.56 | bozonius | In fact, on the 12th, there was no mail from logwatch and I did look around and try to recall what I might have changed |
06:47.47 | bozonius | well, the only issue I can see with dovecot is a failure on a dot lock file, but the problem is that that error has been showing up since i installed dovecot. Dovecot is an imap server anyway; we'd want to look at postfix or exim I think. |
06:48.02 | gnarface | it's possible at least in theory for something to break that doesn't show up until some normal log rotation restarts a daemon that only then incidentally picks up the broken config |
06:48.43 | bozonius | do you know if exim4 has been updated recently? |
06:49.08 | bozonius | nvm |
06:49.20 | bozonius | mine was last updated Jun 27, long before any of this |
06:49.49 | gnarface | dovecot is an imap server but might concievably share some users/filesystem permissions |
06:50.06 | gnarface | it has to be able to write to the same maildir as exim to do it's job |
06:50.18 | gnarface | actually i'm not sure what dovecot is |
06:50.25 | gnarface | i'm using courier-imap from the dovecot project without dovecot |
06:51.49 | gnarface | oh |
06:51.53 | gnarface | hostname changes? |
06:51.59 | bozonius | ah! |
06:52.02 | bozonius | that could be. |
06:52.13 | bozonius | I may have added the domain name |
06:52.17 | bozonius | huh |
06:52.20 | bozonius | ok |
06:53.04 | gnarface | maybe in adding another ip or domain, you inadvertently changed the one that the host sees as it's "default" or canonical name? |
06:53.15 | gnarface | that could choke a mail config for sure |
06:53.24 | bozonius | oops |
06:53.43 | bozonius | dang |
06:53.50 | bozonius | "/etc/hostname" |
06:53.55 | gnarface | also the hostname gets inserted literally into dozens of mail related configs. you might have to update it by hand or dpkg-reconfigure those packages owning those configs at least |
06:53.59 | gnarface | hah |
06:54.01 | bozonius | damn leading slashes |
06:54.04 | gnarface | or just /etc/hostname |
06:54.20 | bozonius | but the basename, the hostname itself, did not change |
06:54.24 | bozonius | would not have changed |
06:54.38 | gnarface | <PROTECTED> |
06:54.43 | bozonius | sure |
06:55.13 | gnarface | if you're relying on dhcp it might matter the order it starts up |
06:55.22 | bozonius | No dhcp here |
06:55.36 | bozonius | and /etc/hosts was last updated Jun 30... |
06:56.03 | gnarface | how about dns servers? |
06:56.04 | bozonius | I suppose I could completely re-installed exim4 |
06:56.13 | bozonius | like dnsmasq |
06:56.20 | gnarface | you probably don't have to do that. try dpkg-reconfigure exim4-config |
06:57.08 | bozonius | ok, but as far as dnsmasq, none. I have my devuan config'd to use a local dns server. and there are no known problems there |
06:57.24 | bozonius | I can find all my other machines, and they can find my devuan machine |
07:00.57 | bozonius | not sure what kind of config for exim4. It's mostly local mail; dovecot serves other machines on my network for imap from this machine. and any smtp (sending mail) from thunderbird is handled by t-bird's own config |
07:00.57 | gnarface | did you mention a stale lockfile issue with dovecot? |
07:01.12 | bozonius | no, no stale lockfile. |
07:02.07 | bozonius | My gut is telling me this is a logwatch issue. |
07:02.18 | gnarface | exim4 is only asking about SMTP here. if you're not receiving or sending mail from/to remote machiens over SMTP, then it's local mail |
07:02.29 | bozonius | I was able to get a report from it sent to my inbox a while ago, but I can't remember what I did... |
07:02.30 | gnarface | exim4 doesn't touch imap |
07:03.27 | *** join/#devuan cioran89 (~cioran89@p5DDF091A.dip0.t-ipconnect.de) |
07:03.58 | gnarface | one thing i've had go wrong on me is ipv6 related - with ipv6 disabled some update re-added the default ipv6 loopback device to the exim4 config on one of my machines (may have been my fault for not paying attention) |
07:04.35 | gnarface | exim4 will fail to do a damned thing if it's configured to look for ::1 on a system that has ipv6 disabled |
07:04.49 | bozonius | well here it is: If I run "logwatch --range yesterday --output mail" from the cmd line, it sends the report to regular user's inbox |
07:04.51 | gnarface | (simple to fix by removing it while running dpkg-reconfigure exim4-config though) |
07:05.07 | gnarface | it'll only run yesterday's report? |
07:05.43 | gnarface | that's odd... is it possible this could be could be caused by clock drift? |
07:05.50 | gnarface | or a timezone shift? |
07:06.04 | bozonius | but logwatch is not invoked that way in the cron script: "logwatch --output mail" |
07:06.20 | bozonius | (but maybe "yesterday" is the default) |
07:06.38 | gnarface | but you said calling it without "--range yesterday" didn't work? |
07:06.42 | bozonius | right |
07:06.45 | bozonius | odd? |
07:06.51 | gnarface | yea |
07:06.57 | bozonius | --range today doesn't work either |
07:07.21 | gnarface | do you have tzdata installed, is your timezone correct, and is the system time correct? |
07:07.25 | bozonius | but why would that affect whether it sends mail or not... |
07:07.32 | gnarface | i don't know, honestly |
07:07.53 | gnarface | easier to verify than find out why |
07:08.02 | gnarface | easier to fix if true |
07:08.02 | bozonius | so far as I know, yes. But as far as drift... i did notice a little while ago, hours ago actually, that my time was slightly off |
07:08.49 | bozonius | and also, I did not have my ntp pointed to my local timeserver, so I corrected that. |
07:08.51 | gnarface | if it drifts backwards and something sees files that are timestamped from the future, i've known that to cause certain types of cron jobs to fail |
07:08.58 | bozonius | but this has been happening for days |
07:09.10 | bozonius | ah good point |
07:09.22 | gnarface | that definitely hoses logrotation actually, now that i think of it |
07:09.49 | bozonius | I think it was off by 6 seconds |
07:10.04 | bozonius | -6 seconds |
07:10.07 | gnarface | hmmm. that doesn't seem like enough to be a big deal. |
07:10.10 | bozonius | yeah |
07:10.12 | gnarface | but i'm not sure |
07:10.36 | gnarface | if logwatch is part of a cron script with other stuff, esp. log rotation, maybe check to see what's happening before it |
07:10.52 | bozonius | nope. logrotate is separate from logwatch |
07:11.25 | bozonius | in fact, the only thing the logwatch script does is check to see if the script even exists then executes it |
07:11.39 | bozonius | there are some comment lines... lol |
07:11.48 | bozonius | nothing juicy though |
07:12.25 | bozonius | you know what... it's a perl script |
07:12.34 | bozonius | the actual program that gets called |
07:12.43 | bozonius | and I think I installed a perl module! |
07:13.16 | bozonius | that was to support another script I run via cron also, but I wonder if maybe... |
07:14.25 | gnarface | ah, cpan, the bane of perl reliability |
07:14.30 | bozonius | yeah |
07:14.35 | bozonius | Statistics::LineFit |
07:14.47 | bozonius | my other script won't run without it. |
07:14.53 | gnarface | alot of the important stuff is actually packaged stably in the debian repos. always check for a package of it before going to cpan |
07:14.56 | bozonius | I mean, it could, but not effectively |
07:15.19 | bozonius | I did. I could not find that module. But if YOU can find it, I'llsubstitute it |
07:15.29 | bozonius | I'll substitute it |
07:16.10 | gnarface | nah i got nothing, but does the timing on this module add line up? that does seem like it could be sabotaging report generation for sure. that would explain why it appears to start generating the report but fails to ever send it |
07:17.21 | bozonius | otoh, my other script has been running since long before this stuff started; looks like the scripts were installed on this machine Jun 2 |
07:17.22 | gnarface | perhaps there's an alternative you could use |
07:17.29 | gnarface | hmmm |
07:17.39 | bozonius | I mean, Jul 2 |
07:18.05 | *** join/#devuan Irrwahn (UNKNOWN@p5799376F.dip0.t-ipconnect.de) |
07:18.11 | gnarface | it could have maybe not run into issues until it had to generate a report where the data calls for the use of that module? like a really long line or something? |
07:18.11 | bozonius | maybe, but it would mean doing some recoding etc. That could take some time |
07:18.34 | gnarface | that's another thing that fits in with the failure to generate a report theory |
07:18.52 | gnarface | if somehow the problem is only excited by certain types of input data |
07:18.55 | bozonius | but it these are 2 entirely separate cron jobs, script sets |
07:19.27 | gnarface | sure but the logwatch one maybe has coding for optional modules that weren't tested? |
07:19.39 | bozonius | could be |
07:19.55 | gnarface | it might be worth it to make sure it *can't* use Statistics::LineFit |
07:20.05 | bozonius | the LinFit module I added from cpan wuld haev to be explicitly referenced though |
07:20.28 | gnarface | i think it's at least plausible it might be called from a fork but only "if present" |
07:20.39 | bozonius | right. |
07:20.45 | gnarface | when you added that module from cpan you didn't change perl versions did you? |
07:21.03 | bozonius | but there'd have to be a "use" statement somewhere deep in the bowels of logwatch or its own modules |
07:21.41 | bozonius | I don't think so, no. Again, I have only had this VM with devuan for about 2 or 3 weeks now. I installed it as soon as I notice 1.0 was out |
07:22.04 | gnarface | oh it's a VM |
07:22.15 | gnarface | that puts a spotlight back on ipv6 support... |
07:22.27 | bozonius | I have ipv6 disabled |
07:22.53 | gnarface | try this |
07:22.54 | bozonius | I've never enabled it in my VMs. I realize I probably should, but that's a different tale |
07:23.18 | gnarface | try this: cat /etc/exim4/update-exim4.conf.conf|grep '::' |
07:23.49 | gnarface | does the line with dc_local_interfaces show up? |
07:24.04 | bozonius | I get: dc_local_interfaces='127.0.0.1 ; ::1' |
07:24.11 | gnarface | bingo |
07:24.25 | bozonius | bingo |
07:24.28 | bozonius | bingo? |
07:24.36 | gnarface | exim4 is trying to listen on both ipv4 and ipv6 localhost |
07:24.44 | gnarface | that won't work in a VM that doesn't have ipv6 enabled |
07:24.50 | gnarface | it wouldn't ever have worked |
07:24.55 | bozonius | but it WAS working |
07:24.58 | gnarface | something must have updated your config |
07:25.06 | bozonius | my VM config? |
07:25.11 | gnarface | your exim4 config |
07:25.14 | bozonius | oh |
07:25.26 | bozonius | well, I did not intentionally reconfig anything myself |
07:25.31 | gnarface | unless you used to have ipv6 enabled in the VM previously? |
07:25.51 | bozonius | not likely at all |
07:25.55 | bozonius | hold on |
07:26.02 | gnarface | well it's easy to fix |
07:26.45 | bozonius | actually, VBox does not manage that AFAIK |
07:26.49 | bozonius | it's the guest OS |
07:26.50 | gnarface | you can re-run dpkg-reconfigure exim4-config or you can just change that file and re-run update-exim4.conf as root (then restart exim4) |
07:27.17 | gnarface | if you only have it set as dc_local_interfaces='127.0.0.1' |
07:27.26 | gnarface | then exim4 knows only to enable ipv4 networking |
07:27.45 | bozonius | even though I can somehow get a mailed report if I pass the yesterday option? |
07:27.53 | gnarface | i'm using a different virtualization solution, but i have to set that in my guests too |
07:28.05 | gnarface | oh yea, that's right the yesterday option |
07:28.13 | gnarface | that's weird as hell |
07:28.25 | gnarface | but i also can't imagine this could have worked otherwise |
07:28.38 | gnarface | hmmm |
07:28.57 | gnarface | and you didn't update to the backports perl version or anything weird like that, did you? |
07:29.09 | bozonius | not intentionally, as I recall |
07:29.11 | gnarface | i'm running out of ideas here |
07:29.17 | bozonius | it's OK |
07:29.33 | bozonius | I'm wondering if a good old-fashioned Microsoft reboot might help here... |
07:29.44 | gnarface | get any other modules from CPAN perhaps? maybe just accidentally upgraded stock ones? |
07:30.02 | bozonius | Don't think so, unless LineFit might have hauled in a few itself |
07:30.20 | gnarface | it would help if you could see some error message about where/how exactly the logwatch report generation fails |
07:30.56 | bozonius | yeah, that's my next step. Go deep into the bowels of the script and see precisely what it is doing when it tries to send mail |
07:31.10 | bozonius | and that means you won't see me for a while... LOL |
07:32.17 | bozonius | it probably wouldn't hurt to see if your ipv6 idea could be it |
07:32.27 | bozonius | wth |
07:38.20 | gnarface | maybe you simply added something that writes log files in a format it doesnt' like, or with permissions it chokes on |
07:38.37 | gnarface | log file *names* in a format it doesn't like, i meant to say |
07:40.51 | gnarface | well anything is possible |
07:41.20 | gnarface | if it's generating web reports it could even be a character set problem |
07:44.04 | bozonius | I am viewing the debug output from logwatch now... |
07:44.25 | bozonius | looks like the "mailer" it uses is "sendmail," but sendmail is a link to exim4 |
07:44.56 | gnarface | that's not unexpected |
07:45.22 | gnarface | exim4 explicitly includes command-line binary compatibility with sendmail for this purpose. lots of stuff still calls it that way |
07:45.35 | bozonius | well, I just wanted to make sure we're not looking at something more peculiar. |
07:45.50 | bozonius | other tools do the same; sendmail being so popular |
07:45.52 | gnarface | of course, it's important that it's working right and you should verify, but no, that's not peculiar. that's the stock setup anywhere exim4 is in use. |
07:46.13 | gnarface | i'm pretty sure php can't even send mail without it |
07:47.13 | gnarface | somewhere in the exim4 docs are a set of simple example commands you can run against it to test the mail queueing and delivery functionality |
07:48.32 | gnarface | on the clock issue - you verified the guest and host clocks are both matching, right? i'm not sure it's possible for them not to match, but if they didn't, that could be an issue too |
07:49.03 | gnarface | this is grasping at straws though at this point |
07:49.27 | bozonius | they should be matching now |
07:50.07 | bozonius | see, one of my VMs is a firewall (ipfire) and it is configured to be the local ntp timeserver. It, in turn, gets its time signal from the internet somewhere |
07:50.31 | bozonius | even if the fw's idea of the correct time is off somehow, at least all the local machines will be in sync. |
07:51.08 | gnarface | ah, ok |
07:52.16 | bozonius | good idea to do it this way, I think. Plus then I'm not being a mooch on Internet NTP bandwidth |
07:52.48 | bozonius | I'd like to think there is a NTP server mgr out there who smiles upon considerate NTP users... |
07:53.45 | bozonius | I'm going to rerun the test with higher debug value; I'm not seeing stuff about mailer |
07:54.25 | gnarface | remember it could have nothing to do with the mail; it might not be even finishing the report generation |
07:54.51 | gnarface | or it may... finish it and fail to save the tmp file ? |
07:55.01 | gnarface | say, you haven't recently moved /tmp to tmpfs have you? |
07:55.07 | gnarface | is your /tmp on tmpfs? |
07:55.13 | bozonius | it removes the tmp directory automatically |
07:55.26 | bozonius | the temp dir is set to use /var/cache/logwatch |
07:55.47 | gnarface | hmmm |
07:55.49 | gnarface | is it tmpfs? |
07:55.58 | bozonius | actually, /var/cache/logwatch.<randomjunk> |
07:56.24 | gnarface | but |
07:56.27 | gnarface | is your /tmp tmpfs? |
07:56.30 | bozonius | oh |
07:56.34 | gnarface | and is it out of space |
07:56.35 | gnarface | ? |
07:57.18 | bozonius | my /tmp is a LVM volume formatted ext4 and mounted to /tmp; usage is at 1% |
07:57.30 | bozonius | but good to point that out |
07:57.36 | gnarface | oh, ok. just making sure. the common default tmpfs /tmp setting of 16MB isn't enough for all types of servers i've seen |
07:57.47 | bozonius | I've gotten bitten by devoured disk space recently on a different system... |
07:58.02 | gnarface | ls |
07:58.04 | *** join/#devuan Pali (~pali@Maemo/community/contributor/Pali) |
07:58.04 | gnarface | woops |
07:58.05 | bozonius | 16MB? That's a joke these days |
07:58.06 | gnarface | heh |
07:58.23 | gnarface | i think you need 64MB in /tmp for some types of mysql config |
07:58.37 | gnarface | you said you made a dns change? |
07:58.53 | gnarface | only thing i could think of is if the new server refuses reverse lookups... |
07:59.01 | gnarface | ? |
07:59.23 | gnarface | if you're parsing logs and doing reverse lookups on the ip addresses |
07:59.54 | gnarface | hmmm... also makes me wonder if the script has an execution time limit |
08:00.07 | gnarface | or ram limit, that may be getting hit |
08:00.14 | bozonius | time limit... hmmm |
08:00.40 | bozonius | again, even though this works... sometimes, with the right "karma" (options) |
08:00.46 | bozonius | lol |
08:00.48 | gnarface | you said you switched to a local dns through dnsmasq |
08:00.50 | gnarface | right? |
08:01.04 | bozonius | not on the VM where this bug is happening, no |
08:01.17 | bozonius | basically, my VBox host runs a dnsmasq server |
08:01.20 | gnarface | just one more thing to check anyway |
08:01.32 | bozonius | I have all my guests pointing to the host as their DNS server |
08:01.32 | gnarface | also dnsmasq has like a 3 hour delay or something like that by default |
08:01.57 | bozonius | well, I think more than 3 hours have passed since this issue began... |
08:02.02 | gnarface | if you make changes to the dns server, you might want to restart dnsmasq |
08:02.10 | gnarface | just saying |
08:02.21 | bozonius | I did, actually, when I created the new VM, but that was weeks ago |
08:10.02 | bozonius | default range IS "yesterday" .... so then why does it only seem to work when I pass the range explicitly I wonder? |
08:10.31 | bozonius | I see this will take some serious debugging. And I'm tired. I think I shall get some sleep for now. |
08:10.43 | gnarface | maybe just uh... fix the script, report a bug? |
08:10.49 | gnarface | that came stock from devuan jessie? |
08:10.57 | bozonius | thank you for all your suggestions. |
08:11.04 | gnarface | or is it from a 3rd party repo? |
08:11.19 | bozonius | yeah, If I figure this out, I will post a bug to Devuan and the logwatch dev |
08:11.31 | bozonius | I believe it is in the devuan repos |
08:12.00 | gnarface | hmmm, weird, seems odd we're just noticing this now |
08:14.42 | bozonius | apt-cache showpkg logwatch says it comes from devuan repo |
08:14.50 | bozonius | and why it is so flakey, too... |
08:16.38 | bozonius | what time zone are you in; I can look for you then |
08:16.43 | bozonius | I'm Pacific Time here |
08:17.26 | gnarface | you will never predict my arrival like that :) |
08:17.31 | gnarface | i'm in pacific time too |
08:17.32 | bozonius | lol |
08:17.33 | bozonius | gotcha |
08:17.38 | gnarface | have a good one |
08:17.59 | bozonius | well, I'm going to sleep now. night, and thanks a lot! Really appreciate all your suggestions. |
08:19.51 | gnarface | it was no problem |
08:20.37 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
08:29.05 | *** join/#devuan pav5088_ (~pav5088@101.165.10.120) |
08:29.46 | *** join/#devuan bluemarlin (~bluemarli@unaffiliated/bluemarlin) |
08:32.37 | *** join/#devuan peetaur (~peter@p57AAA0E1.dip0.t-ipconnect.de) |
08:37.34 | *** join/#devuan peetaur (~peter@p57AAA0E1.dip0.t-ipconnect.de) |
08:38.49 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
08:41.23 | *** join/#devuan MinceR (~mincer@unaffiliated/mincer) |
08:42.53 | *** join/#devuan Mike_Hamza (~Texn@105.110.162.172) |
08:47.17 | *** join/#devuan debdog (~debdog@ip-37-201-6-149.hsi13.unitymediagroup.de) |
08:49.49 | *** join/#devuan mobinmob (~mobinmob@2a02:582:7417:3700:815b:7387:26eb:6829) |
08:51.09 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
08:54.08 | *** join/#devuan eliasr (uid27497@gateway/web/irccloud.com/x-owtusyijamvztkba) |
09:00.44 | *** join/#devuan pav5088_ (~pav5088@101.165.10.120) |
09:13.17 | *** join/#devuan rrq (~ralph@60-242-139-200.static.tpgi.com.au) |
09:14.24 | *** join/#devuan debdog (~debdog@ip-37-201-6-149.hsi13.unitymediagroup.de) |
09:28.00 | *** join/#devuan konsolebox (~konsolebo@112.198.75.111) |
09:32.14 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
09:39.15 | *** join/#devuan debdog (~debdog@2a02:8070:418b:1400:7a24:afff:fe8a:d04d) |
09:58.37 | *** join/#devuan Oldmoss (~Oldmoss@anon-41-191.vpn.ipredator.se) |
09:58.53 | *** join/#devuan Besnik_b (~Besnik@ppp-94-66-221-122.home.otenet.gr) |
10:06.35 | *** join/#devuan gattuso (~gattuso@pompel.me) |
10:14.03 | *** join/#devuan Mike_Hamza (~Texn@105.110.22.23) |
10:26.16 | *** join/#devuan eracc_lappy (~Captalist@unaffiliated/eracc) |
10:33.18 | *** join/#devuan theclash (~luca@57.red-88-12-180.dynamicip.rima-tde.net) |
10:39.02 | *** join/#devuan rdav (~rdav@61.181.148.122.sta.dodo.net.au) |
10:42.20 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-84-220-218-69.clienti.tiscali.it) |
10:42.35 | AntoFox | o/ |
10:47.18 | *** join/#devuan Chanku (~Chanku@2601:704:2:a700:21b:77ff:fea3:4a99) |
11:02.50 | *** join/#devuan zyliwax (~zyliwax@unaffiliated/zyliwax) |
11:12.21 | *** part/#devuan Oldmoss (~Oldmoss@anon-41-191.vpn.ipredator.se) |
11:34.45 | *** join/#devuan rdav (~rdav@61.181.148.122.sta.dodo.net.au) |
11:34.47 | *** join/#devuan Oldmoss (~Oldmoss@anon-43-208.vpn.ipredator.se) |
11:50.57 | *** join/#devuan Levure (~quassel@211.110-242-81.adsl-dyn.isp.belgacom.be) |
11:56.51 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-84-220-218-69.clienti.tiscali.it) |
12:23.38 | *** join/#devuan thaller (~thaller@host17-168-static.38-79-b.business.telecomitalia.it) |
12:36.23 | *** join/#devuan Maarten (~smuxi@185.47.133.96) |
12:53.20 | *** join/#devuan Xenguy (~Xenguy@unaffiliated/xenguy) |
13:07.15 | *** join/#devuan Infiltrator_reco (~Infiltrat@unaffiliated/infiltrator) |
13:22.09 | *** join/#devuan catprints (~realperso@172.58.83.0) |
13:25.09 | *** join/#devuan bazquxer (~bazquxer@gateway/tor-sasl/bazquxer) |
13:29.57 | *** join/#devuan ltem (~ltem@h081217057026.dyn.cm.kabsi.at) |
13:51.04 | *** join/#devuan ltem (~ltem@h081217057026.dyn.cm.kabsi.at) |
13:52.19 | *** join/#devuan aristene (~aristene@187-13-211-182.user.veloxzone.com.br) |
13:54.27 | *** join/#devuan petergozz (~Peter@ppp14-2-61-185.lns21.adl2.internode.on.net) |
13:54.55 | *** join/#devuan telst4r (~telstar@a91-153-137-214.elisa-laajakaista.fi) |
13:54.55 | *** join/#devuan telst4r (~telstar@fsf/member/telst4r) |
14:08.28 | *** join/#devuan cyteen (~cyteen@cyteen42.plus.com) |
14:26.05 | *** join/#devuan ItsHipTobe (~ryder@CPEf0f2490e0e43-CMf0f2490e0e40.cpe.net.cable.rogers.com) |
14:29.10 | *** join/#devuan Oldmoss (~Oldmoss@anon-39-169.vpn.ipredator.se) |
14:35.06 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
14:40.40 | *** join/#devuan Xenguy (~Xenguy@unaffiliated/xenguy) |
14:52.58 | *** join/#devuan Centurion-Dan2 (~Thunderbi@office.centurion.net.nz) |
14:57.49 | *** join/#devuan Xenguy (~Xenguy@unaffiliated/xenguy) |
15:12.47 | *** join/#devuan Akuli (~Akuli@mobile-access-bceeac-157.dhcp.inet.fi) |
15:15.25 | *** join/#devuan Maarten (~smuxi@185.47.133.96) |
15:32.01 | *** join/#devuan g4570n (bee46a02@gateway/web/freenode/ip.190.228.106.2) |
15:59.27 | *** part/#devuan catprints (~realperso@172.58.83.0) |
16:22.34 | *** join/#devuan mike_gentoo (~Texn@105.110.157.15) |
16:22.59 | *** join/#devuan wigums (~wigums@unaffiliated/wigums) |
16:25.55 | *** join/#devuan menip (~menip@73.53.122.132) |
16:29.14 | *** join/#devuan rsx (~rsx@ppp-46-244-242-21.dynamic.mnet-online.de) |
16:38.36 | *** join/#devuan Guest1176 (~Guest@83.220.239.243) |
16:46.36 | *** join/#devuan Schallaven (~Schallave@grad336-296.resnet.yorku.ca) |
16:46.43 | Schallaven | hello freedom squad o/ |
16:46.49 | *** join/#devuan guru (~guru@2601:196:8700:c91:3252:cbff:fec5:6f94) |
16:47.06 | *** join/#devuan furrywolf (~randyg@me00536d0.tmodns.net) |
17:05.38 | telst4r | 'lo o/ |
17:10.19 | *** part/#devuan Oldmoss (~Oldmoss@anon-39-169.vpn.ipredator.se) |
17:14.46 | eracc_lappy | Hey folks. I've been watching my logs a bit more closely since just switching to Devuan 1.0 this past week. I see this in the kernel log from yesterday: https://pastebin.com/raw/AGWwCkBx |
17:15.21 | *** join/#devuan deep-book-gk_ (~1wm_su@194.242.11.14) |
17:15.29 | *** part/#devuan deep-book-gk_ (~1wm_su@194.242.11.14) |
17:15.37 | eracc_lappy | Looks like that may be a bug that has been around a while: https://duckduckgo.com/?q=kernel%3A%5B57187.176352%5D+Uhhuh.+NMI+received+for+unknown+reason+2c+on+CPU+0&t=hu&atb=v72-7&ia=web |
17:20.04 | eracc_lappy | Heh, Red Hat suggests just ignoring the message: https://bugzilla.redhat.com/show_bug.cgi?id=538739#c16 |
17:20.07 | eracc_lappy | Nice. |
17:30.20 | MinceR | lol |
17:30.50 | *** join/#devuan Guest1354 (~Guest@83.220.239.0) |
17:31.55 | muep | doesn't sound unreasonable to me if it is a difficult-to-reproduce warning related to some specific hardware |
17:37.20 | *** join/#devuan [0__0] (~0__0]@ec2-54-85-45-223.compute-1.amazonaws.com) |
17:42.43 | bozonius | gnarface: logwatch has (mysteriously) started working again. |
17:42.54 | bozonius | and no, I cannot explain it |
17:43.00 | *** join/#devuan cyteen (~cyteen@cyteen42.plus.com) |
17:43.11 | golinux | gremlins |
17:43.17 | eracc_lappy | ^^^ |
17:43.27 | bozonius | is that you doing it? |
17:43.45 | bozonius | you messin' with my system? |
17:43.56 | bozonius | it is so weird. |
17:44.16 | bozonius | I looked through my bash history to see if I could figure out what I might have done |
17:44.27 | eracc_lappy | Probably old gremlins from WWII. They are bored, so messing with Linux. |
17:44.56 | bozonius | I'd think those would be dead by now. |
17:45.57 | *** join/#devuan Dantalion (~Dantalion@D978CDAF.cm-3-1d.dynamic.ziggo.nl) |
17:55.23 | MinceR | are they even mortal? |
17:56.37 | *** join/#devuan debdog (~debdog@2a02:8070:418b:1400:7a24:afff:fe8a:d04d) |
17:58.03 | bozonius | is there an inside joke here? I am really curious why this happened. I did not system re-config I can think of that would have caused this. |
17:58.37 | bozonius | If this really is a security hole of some kind, as I think you are implying, then I may need to drop Devuan and go with something else instead. |
18:09.29 | KatolaZ | bozonius: ??? |
18:12.36 | bozonius | well, KatolaZ, "gremlins" and such are not exactly the most technical explanation here |
18:13.00 | bozonius | I did absolutely nothing to my system either prior or after this one-off issue |
18:13.20 | bozonius | so if not "I" then what/whom? |
18:17.38 | KatolaZ | o_O |
18:17.53 | KatolaZ | so people from this channel would have got into your machine and solve the problem? |
18:18.13 | KatolaZ | without you noticing it... |
18:18.41 | KatolaZ | I like more the "gremlins" explanation, TBH |
18:19.17 | KatolaZ | BTW, what was your problem about? |
18:19.27 | bozonius | why does it have to be people from THIS channel, but I won't rule it out either. Look, I think it unlikely, but at this point, there doesn't seem to be any rational explanation |
18:19.38 | bozonius | go back and read the log from last night |
18:19.57 | bozonius | logwatch stopped sending mail, even if I ran it from command line |
18:20.30 | bozonius | if I explicitly passed an option, it would send mail, but no other way I could see. Strangest part about that was that the option I passed is the default! |
18:21.01 | bozonius | I browsed the web for any bugs like this, but I found no hits that explained MY situation. |
18:21.09 | bozonius | spent a good 4 hours or so on this |
18:21.34 | KatolaZ | bozonius: maybe you had some specific config which was not working? |
18:23.19 | KatolaZ | between a non-working logwatch configuration and the alternative you proposes (that Devuan has a remote vulnerability which allowed somebody to remotely log into your system and solve the problem with logwatch) |
18:23.27 | KatolaZ | I tend to prefer the first one... |
18:23.35 | KatolaZ | just Occam's razor, you know... |
18:28.03 | bozonius | non-working? It is working now. It was working before. But it did not work for about 5 days. |
18:28.15 | bozonius | without me changing anything in logwatch's config |
18:28.44 | bozonius | I have a question: Why is there a /lib/systemd on Devuan -- what is this used for? |
18:29.01 | KatolaZ | uh? |
18:29.13 | KatolaZ | bozonius: that's a folder |
18:29.34 | bozonius | I ran dpkg -l on some of the files there and they don't seem to be related to any package |
18:29.44 | KatolaZ | bozonius: that's a folder |
18:29.56 | KatolaZ | and contains systemd units |
18:29.58 | bozonius | there are files in it |
18:30.00 | bozonius | yes |
18:30.01 | bozonius | why |
18:30.04 | KatolaZ | for package that provide systemd units |
18:30.13 | bozonius | such as... |
18:30.14 | KatolaZ | because some packages provide also systemd unit files |
18:30.25 | bozonius | that are not used on Devuan |
18:30.27 | KatolaZ | on top of init scripts |
18:30.32 | KatolaZ | correct |
18:30.39 | bozonius | so why are they still around? |
18:30.48 | muep | the units don't do anything if you do not have systemd and I'd guess there's more useful work for packagers than removing a few small unused files |
18:30.53 | KatolaZ | because we should otherwise rebuild those packages |
18:31.05 | KatolaZ | and those files are useless if systemd is not installed |
18:31.21 | *** join/#devuan Guest1206 (~Guest@83.220.237.72) |
18:31.29 | bozonius | "because we should otherwise rebuild those packages" -- that is an awkward sentence; I don't follow that part |
18:31.35 | KatolaZ | removing those files would entail forking something like 300-400 more packages |
18:31.49 | *** part/#devuan Guest1206 (~Guest@83.220.237.72) |
18:31.50 | bozonius | script? |
18:31.56 | KatolaZ | just to remove an unused file |
18:31.58 | KatolaZ | uh? |
18:32.04 | KatolaZ | script for what? |
18:32.16 | bozonius | thanks, KatolaZ |
18:32.35 | bozonius | where are you from? US, UK..? |
18:32.42 | bozonius | EU? |
18:32.48 | bozonius | (just curious) |
18:33.08 | KatolaZ | bozonius: why is that relevant for you? (just curious) |
18:33.55 | bozonius | no one I know says "because we should otherwise" |
18:34.09 | bozonius | so I am wondering if this is some local dialect |
18:34.15 | bozonius | that's all |
18:34.38 | KatolaZ | -_- |
18:34.48 | muep | I did not find that an odd expression at all |
18:35.00 | bozonius | I've never heard it, that's all |
18:35.07 | KatolaZ | bozonius: you asked |
18:35.10 | bozonius | at any rate, I've got things to do |
18:35.12 | KatolaZ | 19:30 < bozonius> so why are they still around? |
18:35.16 | KatolaZ | and I replied |
18:35.19 | bozonius | seee you later |
18:35.19 | muep | though I am not not a native speaker so I am not fully idiomatic anyway |
18:35.22 | KatolaZ | 19:30 < KatolaZ> because we should otherwise rebuild those packages |
18:35.57 | bozonius | is away from kb |
18:41.09 | *** join/#devuan knidos (~knidos@85.100.166.140) |
18:53.07 | *** join/#devuan tarbz2 (~Thunderbi@186.23.155.100) |
19:03.33 | *** part/#devuan stybla (stybla@pylon.zeratul.org) |
19:03.54 | *** join/#devuan Drugo (~Drugo@62-11-1-95.dialup.tiscali.it) |
19:09.49 | *** join/#devuan Jjp137 (~Jjp137@cpe-75-83-16-81.socal.res.rr.com) |
19:11.19 | *** join/#devuan kelsoo1 (~kelsoo@dragora/developer/kelsoo) |
19:20.38 | *** join/#devuan Oldmoss (~Oldmoss@46.165.242.166) |
19:33.40 | *** join/#devuan ephemer0l (~ephemer0l@pentoo/user/ephemer0l) |
19:38.06 | *** part/#devuan Oldmoss (~Oldmoss@46.165.242.166) |
19:38.29 | DocScrutinizer05 | ln -s /dev/null /lib/systemd |
19:38.55 | DocScrutinizer05 | ;-P |
19:39.10 | DocScrutinizer05 | (just kidding) |
19:47.39 | bozonius | find / -name '*systemd*' -exec rm -f {} \; |
20:00.25 | DocScrutinizer05 | that's what a sane sysadmin would like to do, yes. But ut won't do any good in the end |
20:01.26 | DocScrutinizer05 | e.g. there's still libsystemd0 that is basically needed to keep systemd out of the system |
20:06.15 | fsmithred | udev brings in /lib/systemd and uses it |
20:06.35 | DocScrutinizer05 | dang |
20:06.59 | fsmithred | yeah, even with libsystemd0 excluded it's still there |
20:07.05 | fsmithred | even with eudev it's still there |
20:21.58 | *** join/#devuan GoatAvenger (~GoatAveng@unaffiliated/goatavenger) |
20:22.19 | *** join/#devuan cyteen (~cyteen@cyteen42.plus.com) |
20:23.26 | *** join/#devuan Capricornus (~HexChat@5.86.98.177) |
20:49.06 | *** join/#devuan rdav (~rdav@61.181.148.122.sta.dodo.net.au) |
21:06.45 | *** join/#devuan fsda6342 (~User@host-69-145-45-32.shr-wy.client.bresnan.net) |
21:12.51 | *** join/#devuan mchasard (~devuan@48.233.8.109.rev.sfr.net) |
21:14.51 | *** join/#devuan fsmithred (~fsmithred@68-184-46-18.dhcp.oxfr.ma.charter.com) |
21:30.57 | *** join/#devuan matlock (~matlock@24.96.131.20) |
21:58.16 | *** join/#devuan Oldmoss (~Oldmoss@193-123-11.connect.netcom.no) |
22:04.49 | *** part/#devuan Mike_Hamza (~Texn@105.110.157.15) |
22:19.49 | *** part/#devuan Oldmoss (~Oldmoss@193-123-11.connect.netcom.no) |
22:24.00 | *** join/#devuan rann (sid175221@gateway/web/irccloud.com/x-fzvstysxobogpnyt) |
22:49.49 | *** join/#devuan kelsoo (~kelsoo@dragora/developer/kelsoo) |
22:59.02 | *** join/#devuan Striktarn (~Striktarn@unaffiliated/striktarn) |
23:37.27 | *** join/#devuan Drugo (~Drugo@62-11-1-95.dialup.tiscali.it) |
23:52.40 | *** join/#devuan mushmouth (~Patrick@bas1-peterborough62-74-12-146-200.dsl.bell.ca) |