=== genii is now known as genii-core | ||
=== Bebef9 is now known as Bebef | ||
=== cpaelzer_ is now known as cpaelzer | ||
jamespage | icey: hey | 10:57 |
---|---|---|
jamespage | what's the rationale for adding the version suffix to the luarocks rocks- subdirectory in your patch? | 10:57 |
jamespage | the conditional that's wrapped in lacks the -5.1 | 10:58 |
icey | jamespage: ah - should have added it to the conditional as well; the version suffix is because that's what's added to the `--tree` path | 10:59 |
jamespage | icey: https://paste.ubuntu.com/p/SbMR3tBSBF/ | 11:45 |
icey | jamespage: lgtm | 11:46 |
icey | jamespage: I'm seeing errors like https://www.mail-archive.com/debian-bugs-dist@lists.debian.org/msg1827415.html on ceph as well | 13:06 |
jamespage | icey: uploaded | 14:05 |
jamespage | (luarocks) | 14:05 |
icey | jamespage: \o/ | 14:05 |
jamespage | icey: pls can you submit those changes back to debian as well :) | 14:27 |
icey | jamespage: did a report2debian :-D | 14:38 |
=== genii-core is now known as genii | ||
=== Kamilion|ZNC is now known as Kamilion | ||
smoser | rbasak: swtpm import please ? | 18:55 |
smoser | and libtpms | 19:01 |
=== paride4 is now known as paride | ||
Zaliek | Somehow one of my server's time is not NTP syncronized and the time has drifted enough that TOTP doesn't work anymore | 22:13 |
Zaliek | How do I force a time sync from the terminal? | 22:13 |
Zaliek | Oddly the RTC clock is the correct time | 22:16 |
=== Voeid8 is now known as Voeid | ||
Edgan | Zaliek: sudo apt-get update ; sudo apt-get -y install ntpdate ; sudo ntpdate pool.ntp.org | 22:18 |
Zaliek | It seems systemd-timesyncd is supposed to be doing this | 22:18 |
Zaliek | I tried restarting the service but it still won't update the time | 22:18 |
Edgan | Zaliek: system status systemd-timesyncd ; journalctl -u systemd-timesyncd.service -f | 22:19 |
Zaliek | Nothing in there, just the service started/stopped messages | 22:20 |
rfm_ | Zaliek, what does "timedatectl timesync-status" say? | 22:21 |
Zaliek | Not a valid command | 22:22 |
Zaliek | status says NTP syncronized: no | 22:22 |
Zaliek | Network time on: yes | 22:23 |
JanC | it might not change time if the difference is too large | 22:23 |
JanC | ntpdate should do that though | 22:24 |
Zaliek | it's 2 minutes | 22:27 |
tomreyn | i think there are some commented out options in the timesyncd configuration file you can use to get more verbose info. | 22:28 |
tomreyn | that's /etc/systemd/timesyncd.conf | 22:28 |
tomreyn | hmm no, nothing about verbosity | 22:29 |
Zaliek | This system is 16.04 if that makes a difference | 22:29 |
Zaliek | Honestly surprised it took this long for the clock to drift by 2 mins | 22:30 |
tomreyn | oh 16.04, do you have ESM then? | 22:30 |
Zaliek | Yeah, but no actual support | 22:30 |
Zaliek | Just the free ESM | 22:30 |
tomreyn | hmm, i think 16.04 only used a smaller subset of sytemd services | 22:31 |
Zaliek | It's being replaced soon but I can't login to the app it's hosting because the clock has drifted | 22:31 |
tomreyn | i don't remember whether it already did timesyncd | 22:31 |
Zaliek | Yeah it's using the systemd service | 22:31 |
Zaliek | All of the old ntp tools are not installed | 22:32 |
tomreyn | you could override the ntp server it's synching to in the configuration file, then restart the service, wait 5 minutes and see whether it caught up. | 22:32 |
tomreyn | if not, i'd disable timesyncd and use a different ntp client, such as chrony | 22:33 |
JanC | also, for a quick workaround you can set the time manually, of course... | 22:36 |
Zaliek | I figured it out... | 22:36 |
Zaliek | I never allowed NTP through the firewall | 22:36 |
Zaliek | :facepalm" | 22:36 |
JanC | hehe | 22:36 |
Zaliek | Thanks for your help. Gonna need to do the same for my other servers | 22:38 |
Zaliek | Luckily I never setup TOTP with SSH on this one | 22:38 |
tomreyn | you just lost an excuse to not upgrade just yet ;-P | 22:38 |
Zaliek | Well it still needs to because I need a newer version of PHP | 22:39 |
JanC | I have the other problem: I need a 16.04 system because I need an older PHP version... :-/ | 22:41 |
tomreyn | i assume you know about https://deb.sury.org/ | 22:45 |
JanC | I did not | 22:47 |
tomreyn | that's debian developer ondřej surý providing co-installable php versions for several debian + ubuntu releases | 22:47 |
JanC | could be useful, I guess :) | 22:47 |
tomreyn | those can break release upgrades, of course, so be careful about that | 22:48 |
JanC | I want to get rid of that website though, so maybe not going to use it for this | 22:49 |
tomreyn | this, too, can't prevent the inevitable or provide full security support for bugs in software long EOL'd upstream. | 22:51 |
JanC | supposedly PHP 5.x is still supported upstream? | 22:52 |
JanC | apparently not, although that repository says it only has supported versions? | 22:53 |
JanC | (it clearly does have EOL versions) | 22:55 |
tomreyn | i think he means to say he does provide support for them | 22:56 |
JanC | """Only Supported Versions of PHP (http://php.net/supported-versions.php) for Supported Ubuntu Releases (https://wiki.ubuntu.com/Releases) are provided.""" | 22:56 |
tomreyn | oh, that seems to be misleading, yes | 22:56 |
tomreyn | https://www.patreon.com/posts/may-2021-update-50798437 | 22:59 |
Zaliek | Odd, apparently LXD containers keep their time in sync with the host via some other mechanism than NTP | 23:07 |
tomreyn | cat /sys/devices/system/clocksource/*/current_clocksource # https://www.kernel.org/doc/Documentation/timers/timekeeping.txt | 23:16 |
sarnold | perhaps by the expedient of not having virtualized clocks? | 23:22 |
rfm_ | LXD containers (as oppsed to LXD VMs) use the host kernel instance, no? So it's lierally the same clock | 23:23 |
rfm_ | Which is why systemd-timesyncd.service has "ConditionVirtualization=!container" | 23:25 |
=== rfm_ is now known as rfm |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!