| gry | on a scale from "don't get it entirely" to "know it well and can teach it to others", i'm currently "do not get it entirely" about the clock | 01:28 |
|---|---|---|
| gry | «delete the localhost cmos clock stratum(?) from NTP's config» — this line is not there | 01:28 |
| gry | «depending on which virtualization system you are using» — it is vmmd, and other openbsd guests work fine | 01:28 |
| manuel | Hello, where can I find the recommended requirements for installation: D | 08:29 |
| rgh[m]1 | manuel: In what sense? | 09:19 |
| fsmithred | Pentium III or better. | 11:06 |
| used____ | I have a problem on Beowulf, a user crontab entry which should run every 15 mins does not run. It never worked. | 12:43 |
| used____ | Do I have to do something special for user crontabs to run? The executable is a script in the user's $HOME and it works fine, and is a+x | 12:44 |
| used____ | I think I know what causes this, my run time spec was slighlty dodgy in the contrab -e installed table, let's see if it works now | 12:45 |
| used____ | There is no /etc/cron.allow or .deny installed | 12:47 |
| mro | used____ what's the result of $ sudo service cron status | 12:47 |
| used____ | cron is running | 12:47 |
| used____ | I have other crontabs executing and logged in syslogd | 12:48 |
| used____ | Not this user's though | 12:48 |
| mro | does that fire via crontab: $ touch /tmp/hello-dron | 12:48 |
| used____ | Say again | 12:48 |
| used____ | ? | 12:49 |
| mro | if you enter into crontab -e: * * * * touch /tmp/hello-cron | 12:49 |
| used____ | Let's wat 15 mins until the real installed script should fire. | 12:51 |
| used____ | It ran. Okay so I fumbled the crontab entry time spec, and it probably had no terminating newline. Thanks for the patience. | 12:52 |
| used____ | wait even | 13:00 |
| * used____ reads incredulously: /etc/default/anacron -> ANACRON_RUN_ON_BATTERY_POWER=no -- there's a howto fix it in the anacron readme in shared docs | 13:10 | |
| used____ | This means that none of the anacron jobs run on laptops which are not on mains power for a long enough time on any systems where anacron is shipped with these defaults. | 13:11 |
| used____ | Which include systemd and non systemd systems. | 13:11 |
| used____ | HELLO?! | 13:11 |
| used____ | doc: /usr/share/doc/anacron/README.Debian (Beowulf) | 13:12 |
| used____ | More f***ery: /etc/cron.daily/mlocate : if which on_ac_power ...; then ON_BATTERY=0; # sabotaging mlocage on battery yet again | 13:16 |
| used____ | desktop pc here: /usr/bin/on_ac_power -> 255; # could not determine power status | 13:17 |
| used____ | So the script above determines it is not on mains power (lies) and does not run mlocatedb update. | 13:18 |
| used____ | Guys, upstream is a fountain of crap, PLEASE, PLEASE, PLEASE keep track of crap like this somewhere? The list of crap to undo to make systems work as they did in 1998 is becoming endless. | 13:18 |
| used____ | There are 17 installed scripts in /etc/cron.daily/ and I have to check each, in addition to anacron and others, to make sure they run, because some dweeb upstream did not bother to check that on_ac_power can return something other than 0 or 1 and assumed "other" is "not on ac power" | 13:27 |
| used____ | I was wondering why the heck ntp is not updating and so on. | 13:28 |
| used____ | Set in: /etc/default/anacron : ANACRON_RUN_ON_BATTERY_POWER=yes | 13:29 |
| used____ | Disable the stanza checking on_ac_power in /etc/cron.daily/mlocate | 13:29 |
| used____ | The on_ac_power fiasco continues in for example: /etc/cron.daily/apt-compat and many other places | 13:30 |
| used____ | Solution: retire on_ac_power (on mains powered host), replace with link to /bin true | 13:31 |
| used____ | /bin/true | 13:31 |
| used____ | The people who write these rules are chock full of hoey and hubris. Nobody ever checked these things? Machines with broken ACPI and thus mains power reporting are like 50% of older PCs/ | 13:32 |
| used____ | If you think you are affected, on a PC on mains (what else?!), run: /usr/bin/on_ac_power; echo $? | 13:33 |
| used____ | If the result is anything other than 0 you are affected. See mods above. | 13:34 |
| djph | bash: /usr/bin/on_ac_power: No such file or directory | 13:34 |
| djph | "meh" | 13:34 |
| used____ | Even better | 13:34 |
| used____ | djph: try to run /etc/cron.daily/mlocate manually see if it executes the mlocatedb update | 13:35 |
| used____ | It runs with set -e so ... | 13:35 |
| djph | I also don't have an /etc/cron.daily/mlocate | 13:35 |
| djph | I do have a (non-broken) /etc/cron.daily/locate script though | 13:36 |
| used____ | Try on your linux host, not on windows ;) | 13:36 |
| used____ | What locate are you using? | 13:36 |
| used____ | Also what Devuan? Beowulf? Other? | 13:36 |
| djph | Chimaera | 13:36 |
| djph | looks like the installed package is locate 4.8.0-1 | 13:37 |
| used____ | Right, look in the script, see if it references battery or on_ac_power | 13:39 |
| used____ | grep -Rl on_ac_power /etc/cron* -> /etc/cron.daily/mlocate /etc/cron.daily/apt-compat | 13:40 |
| used____ | Restart cron service after editing /etc/default/anacron | 13:40 |
| * used____ is pissed at these issues | 13:41 | |
| used____ | Discovered by accident too, no systematic search | 13:41 |
| used____ | pet peeve #2 for today: no explicit way to tell crond that the MAILTO= is NEVER to be tried or used, since most systems do not run their own mail server now. Use syslog only. It is "implicit". Rephrased: you can't turn that off. | 13:43 |
| used____ | Not to mention logrotate was also broken for the same reason, wrong assumption on on_ac_power | 13:49 |
| used____ | <$#!!&&*%%^!!!> | 13:49 |
| used____ | My /var/log/daemon.log is full of entries like: 'Mar 23 12:26:24 beowulf acpid: received input layer event "jack/lineout LINEOUT unplug"' -- 'Mar 23 12:26:24 beowulf acpid: 0 total rules matched' -- 'Mar 23 12:26:24 beowulf acpid: completed input layer event "jack/lineout LINEOUT unplug"' -- how do I silence this? There are thousands of them, less than a second apart. The jack is not physically inserted | 13:53 |
| used____ | or removed. Kernel updates did not help. | 13:54 |
| used____ | Looks like something I could address with a dummy event handler in /etc/acpid* ? | 13:55 |
| used____ | * /etc/acpi/* | 13:55 |
| rrq | if you use rsyslogd you might rather add a filter for ignoring those logs | 14:01 |
| used____ | I have to set acpid to not log events since they are piped to debugging scripts in /etc/acpi/ already | 14:17 |
| used____ | The fake runs on battery thing prevented log rotation for over a year, other than manual rotations. | 14:25 |
| used____ | I do not really expect to manually set up everything in 2022, on non Slackware systems, you know. I did not bother to check until things got really ugly. Like 82 MB syslog | 14:25 |
| used____ | Break time | 14:26 |
| gnarface | used____: do you have the same issue with the regular cron package? | 15:31 |
| djph | used____: no, there is no reference at all to "battery" or "ac" (exclusive of a closing 'esac' ) in /etc/cron.daily/locate | 15:50 |
| djph | and the case statements are looking for kernel versions / ionice implementations. | 15:53 |
| fsmithred | used____, did you install from a live-iso or from installer iso? If live, anacron might be diverted. | 16:20 |
| fsmithred | if you have /usr/sbin/anacron.orig.anacron, then run these... | 16:21 |
| fsmithred | dpkg-divert --no-rename --remove /usr/sbin/anacron | 16:21 |
| fsmithred | mv /usr/sbin/anacron.orig.anacron /usr/sbin/anacron | 16:21 |
| fsmithred | and restart anacron | 16:21 |
| used____ | re. | 18:07 |
| used____ | I have /usr/sbin/anacron.orig.anacron | 18:07 |
| used____ | What does "reverted" mean, exactly? | 18:07 |
| used____ | I restarted *cron* which runs anacron | 18:08 |
| used____ | I hope nothing broke, after doing the above | 18:09 |
| used____ | -divert --remove and then mv | 18:09 |
| used____ | Yes I installed from a live iso image of Beowulf in cca. 2020 | 18:09 |
| fsmithred | used____, live-config moves anacron out of the way so it can't run in the live session where it would be useless. (on a read-only system) | 18:19 |
| fsmithred | and the installer carries that over. (oops.) | 18:19 |
| fsmithred | fwiw, live-tools will disable update-initramfs | 18:20 |
| used____ | update-initramfs runs every time I update the kernel | 18:21 |
| fsmithred | I think that doesn't get carried over. | 18:21 |
| used____ | ;) | 18:21 |
| fsmithred | I usually avoid installing it. | 18:22 |
| used____ | it? | 18:22 |
| fsmithred | live-tools | 18:22 |
| fat-avocado | Hello Everyone :) I am troubleshooting my Luks/LVM install with /boot on a separate device. My fstab and crypttab | 21:31 |
| fat-avocado | What else do I troubleshoot? | 21:32 |
| fat-avocado | My fstab and crypttab all check out. I replaced all the entries with UUIDs and changing boot options didn't help. I boot to a blinking cursor. | 21:37 |
| gnarface | there might be some package you're missing i think? | 21:41 |
| gnarface | cryptsetup-initramfs maybe? | 21:42 |
| gnarface | i forget, this keeps coming up though... stick around someone will have an answer | 21:42 |
| fat-avocado | I don't have crypt* in my /bin, so I think you are right. | 21:51 |
| fat-avocado | Can | 21:55 |
| used____ | Didn't someone say devuan servers use OVH... great environmentally friendly building, wooden ceilings? ... https://www.theregister.com/2022/03/22/ovhcloud_fire_datacenter_report/?td=rt-3a | 22:35 |
| used____ | Admittedly, it looked nice from the outside. Before the fire. | 22:37 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!