systemdlete | This might or might not be off-topic. I am trying to recall how, back in "the day," one was able to peruse (I think a file similar to utmp) to get the exit code of a process that terminated. I am not seeing this in today's utmp or wtmp files, they seem to only log login sessions these days. | 05:19 |
---|---|---|
systemdlete | I used to work on a AT&T 3B20, and there was a command line tool that you could use to get information about any process, including things like start time, end time, exit code, user and group IDs, etc, etc. | 05:20 |
systemdlete | Istr, though, that years later, that feature seemed to be slowly defecated out of existence. | 05:20 |
systemdlete | Any other old farts like me who remember something like this? Probably c. 1985 | 05:21 |
systemdlete | I had a hung system, shelled out to a console because the desktop was not responding, and found out it was thunderbird hanging up everything. So I terminated it. But when I returned to the desktop, all my open terminals were gone. So I thought if I could use that old feature somehow, I could find out why they all disappeared even though I only terminated the one process (t-bird). | 05:23 |
systemdlete | Well, give me some idea anyway. | 05:23 |
systemdlete | I suppose oomkiller might be the culprit here, but still, it would be nice to know exactly how/why this happened. | 05:24 |
* systemdlete thinks sometimes the old ways were better for certain things... | 05:24 | |
systemdlete | IOW, I am looking for a tool that has historical information on terminated processes. That might be impractical these days, and maybe why it has been defecated. | 05:26 |
onefang | I suspect you meant "deprecated" not "defecated". It's important that us old farts get that right. | 05:48 |
rwp | systemdlete, Do you mean something more than $? which is the exit status of the last foreground pipeline? | 06:00 |
rwp | Or do you mean using "wait" to get the exit status of the background process? | 06:00 |
systemdlete | This was a system-wide (kernel) facility that logged this info to--I think, but not sure--the utmp file. Again, my memory is VERY fuzzy about this. | 06:01 |
rwp | AFAIK the exit status of processes have not been stored anywhere other than passed to the parent when the parent wait(2)'s on the child. | 06:01 |
systemdlete | Even on very old svr5 systems? | 06:01 |
rwp | I worked for 20 years on HP-UX systems. I don't recall any way to do it there. | 06:02 |
systemdlete | This is not a Linux-specific thing, which is why I started out saying I am not sure if this is the right venue for my question. | 06:02 |
rwp | HP-UX is a System V userland with a kernel based upon the BSD 4.2 kernel. | 06:02 |
systemdlete | It might have been something specifically configured on 3B20, but I really don't remember the details. | 06:02 |
rwp | I never worked on VMS or many of the other classic systems. I could certainly believe that the IBM mainframe OS had a storage for it. | 06:03 |
systemdlete | I've worked on HP-UX also. As well as Solaris, SunOS, AIX, and a number of other *nix systems from the mid-80s, which were referred to as "supermicros" | 06:03 |
systemdlete | I've worked with IBM MVS a little bit, I got to the point where I could almost write JCL | 06:04 |
systemdlete | But this was def not MVS | 06:04 |
rwp | I have worked on a few SunOS, Solaris, and AIX systems too. They were our competitors. But some necessary things were vendor locked. So we used them. | 06:04 |
systemdlete | So no one here recalls a facility like this on old *nix systems. I should expect that... | 06:05 |
rwp | I could also believe that some utility in the "accounting" system of tools would log this information too. Seems plausible. | 06:06 |
systemdlete | Oh, that much I *do* remember--it logged this info. | 06:06 |
systemdlete | I just don't remember exactly where or how it was accessed. That was 35+ years ago. | 06:07 |
systemdlete | I was still learning at that point. | 06:07 |
rwp | Perhaps pcacct process accounting? https://stackoverflow.com/questions/29246631/how-does-linux-process-accounting-psacct-work | 06:08 |
systemdlete | ah! | 06:08 |
systemdlete | yes, ty ty ty | 06:08 |
rwp | But if we are going to talk about that let's do it in -offtopic. | 06:08 |
systemdlete | sure. No need. Now I recall what I was looking for. Thanks again for stimulating some old neurons for me. | 06:08 |
rwp | You are welcome. Happy to help. I very often need a second brain to get a neuron to connect. | 06:09 |
systemdlete | Might be a lack of certain minerals that make the synapses fire. | 06:10 |
noogie | The main repo key expired? | 12:41 |
hagbard | The error message says it did. | 13:01 |
onefang | Oops, I see that to. | 13:09 |
u-amarsh04 | I'm getting signature errors when attempting an aptitude -u - EXPKEYSIG BB23C00C61FC752C Devuan Repository (Amprolla3 on Nemesis) <repository@devuan.org> | 14:09 |
lts | Yes | 14:09 |
lts | I trust devs are very much on it | 14:11 |
onefang | Yep, we are. | 14:11 |
blarglibloup | hi, i assume i'm not alone but I can't update any of my devuan machines since a few hours now: the repo key is apparently expired? | 14:24 |
onefang | We are working to fix that. | 14:25 |
blarglibloup | ok thanks | 14:25 |
u-amarsh04 | and for those using chromium on unstable I filed this bug report: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1018937 | 14:28 |
Chain-Q | yeah, I just ran into apt keys expiring too, glad to know it's being worked on. | 14:33 |
av6 | time to write a cron script to check the expiry date? | 15:32 |
onefang | Perhaps a job for apt-panopticon. | 15:34 |
Guest9050 | Hi it seems like gpg key expired, when new keyring will drop? | 15:43 |
onefang | We are working on it now. | 15:43 |
Guest9050 | I see, thx for info | 15:44 |
rustyaxe | suddenly my apt update tells me all my dpkg keys are expired. Tried installing the keyring deb from here https://beta.devuan.org/os/keyring to no resolution. Ideas? | 16:05 |
rustyaxe | not uising the 2017 keyring of course | 16:05 |
iceman1 | hi same here | 16:05 |
iceman1 | me too | 16:06 |
rustyaxe | aha | 16:06 |
iceman1 | I check the dev1galaxy and its seems to be happening to other users | 16:06 |
iceman1 | someone mention New keyring just got built and needs to get into the repos and propagate to the mirrors. | 16:06 |
onefang | The ideas I have been telling everyone - wait for us to fix it. We are working on it. | 16:06 |
iceman1 | @onefang , gotcha | 16:07 |
sgage | before descending on the IRC and repeating the report of the key being expired: | 16:07 |
sgage | you really ought to consider reviewing the IRC log before, because this | 16:07 |
sgage | has been reported and discussed now for a while. | 16:08 |
sgage | Log at http://reisenweber.net/irclogs/libera/_devuan/latest.log.html | 16:08 |
iceman1 | sgage we just hope and it was a simple question, I understand everyone is most likely overwhelm | 16:08 |
iceman1 | sgage we just hop on I mean | 16:09 |
onefang | Well it's past midnight here, so I hope someone else can take over keeping people informed that we are working on it. | 16:10 |
iceman1 | I can do that | 16:11 |
rustyaxe | apologies for asking something apparently been asked a bunch this morning. can i tell apt to ignore signing for now? i rather need to install bluez-cups so i can carry on with this morning. | 16:11 |
bb|hcb | rustyaxe: yep, that will help | 16:13 |
lts | Perhaps use the channel topic? | 16:13 |
rustyaxe | lts: Might help, was the first place i looked before asking, actually. Didnt see anything there ;) | 16:14 |
iceman1 | I just the dev1galay and updating the chat their | 16:15 |
iceman1 | I am keeping people informed at https://dev1galaxy.org/viewtopic.php?pid=37308#p37308 | 16:15 |
bb|hcb | https://pkginfo.devuan.org/cgi-bin/policy-query.html?c=package&q=devuan-keyring&x=submit | 16:20 |
iceman1 | also maybe we meantion this in reddit, | 16:20 |
bb|hcb | ^^^ here you can check and in case you manually install the 2022.09.03 version, it will not work. wait for the 2022.09.04 | 16:20 |
onefang | Hopefully I'll still be awake when the fix hits the mirrors, and can let people know when all the mirrors have updated. | 16:24 |
iceman1 | reddit posted, hopefull this will reduce the load of questions a little | 16:29 |
APic | Heya | 16:30 |
APic | How can i let the Kernel-OOM-Killer trigger a Coredump of the killed Process please? | 16:30 |
iceman1 | echo 1 > /proc/sys/vm/oom_dump_tasks | 16:32 |
APic | Thanks | 16:33 |
APic | Already have that set | 16:34 |
APic | Where does the Kernel dump the Process? | 16:34 |
iceman1 | To check where the core dumps are generated, run: sysctl kernel.core_pattern or | 16:41 |
iceman1 | cat /proc/sys/kernel/core_pattern | 16:41 |
APic | Thanks! | 16:42 |
APic | Err:1 https://deb.devuan.org/devuan/merged chimaera InRelease | 16:42 |
APic | Certificate verification failed: The certificate is NOT trusted. The name in the certificate does not match the expected. Could not handshake: Error in the certificate verification. [IP: 2a0a:e5c0:2:2:400:c8ff:fe68:bef3 443] | 16:42 |
APic | How can i fetch a new Cert please? | 16:42 |
APic | And trust it? | 16:42 |
APic | [5.8]root@APic:~# cat /proc/sys/kernel/core_pattern | 16:43 |
APic | core | 16:43 |
* APic needs a Directory | 16:43 | |
APic | My irssi got killed. I cannot find the Coredump in /usr/local/bin where the Binary resides or in /tmp | 16:44 |
iceman1 | dev are working on it https://dev1galaxy.org/viewtopic.php?pid=37308#p37308 | 16:45 |
APic | Ok | 16:45 |
APic | Thanks | 16:45 |
iceman1 | need to step away but I will come and check in later | 16:55 |
iceman1 | need to step away but I will come and check in later | 17:18 |
debuse42 | Someone recognized signing keys end of life? | 17:20 |
iceman1 | yes | 17:21 |
iceman1 | https://dev1galaxy.org/viewtopic.php?id=5211 for details | 17:21 |
iceman1 | please pass this along | 17:21 |
iceman1 | https://www.reddit.com/r/devuan/comments/x4vr62/apt_update_error/ reddit feed | 17:21 |
onefang | I was about to thank iceman1 for helping to spread the word, so I can sleep now. lol | 17:22 |
APic | ktnx | 17:41 |
APic | Works! | 17:46 |
APic | Epic! | 17:46 |
APic | ♥♥♥ | 17:46 |
___used | The latest Windows update which reflashes BIOS "Secure Boot" invalidating certs on which Linux signed bootloaders rely, is the BEST advertisement for not dual booting windows. | 18:27 |
___used | Has anyone here been affected? | 18:27 |
rustyaxe | ___used: no i dont own windows | 18:57 |
___used | Microsoft owns Windows, everyone else "rents" from them... | 18:58 |
shibboleth | used: wat? | 18:58 |
shibboleth | a windows update that touches spi flash? | 18:59 |
___used | UEFI flash | 18:59 |
shibboleth | which is spi-connected flash | 18:59 |
___used | apparently. News headline is from heise.de in German | 18:59 |
shibboleth | still, i doubt this is the case | 18:59 |
shibboleth | there are a plethora of uefi/mobo vendors and even more flash vendors | 19:00 |
___used | Been trying to find an external ref, apparently this is a thing, updates for W10 and 11 tend to applt crl's | 19:00 |
shibboleth | i find it exceedingly unlikely that this is the case | 19:00 |
___used | https://www.heise.de/hintergrund/Bootloader-Signaturen-per-Update-zurueckgezogen-Microsoft-bootet-Linux-aus-7250544.html the article proper is paywalled alas | 19:00 |
___used | https://www.ghacks.net/2022/08/15/kb5012170-windows-update-error-0x800f0922-uefi-bios-update-may-resolve-it/ closest I could find | 19:02 |
shibboleth | ehm | 19:02 |
___used | ? | 19:03 |
shibboleth | german isn't my first language but my understanding of the language in that article is the opposite case | 19:03 |
rustyaxe | no clue, windows never appealed to me, ended up moving to linux from unix proper in the late 90s. tried windows some times, wasnt impressed. I treat it like demo OS on new machines. try the hardware then install something useful | 19:03 |
___used | Right, read the ghacks article instead please. | 19:03 |
shibboleth | msfts new bootloaders are signed with a new cert not necessarily trusted by older mobos | 19:04 |
___used | "An advisory page on Microsoft's website provides additional information on the issue. According to Microsoft, the security issue was found in the GRUB bootloader, which is commonly used by Linux." | 19:05 |
shibboleth | also, the update doesn't "flash your bios", it makes a few changes in efivars | 19:05 |
___used | please use the provided info: @ghacks | 19:05 |
___used | It installs a DBX crl list which disallows all known linux signed bootloaders. | 19:06 |
___used | After that you can only boot into linux with secure boot disabled. | 19:06 |
golinux | ___used: This subject isn't for devuan support. Please take it to OT | 19:11 |
___used | ok. But is Devuan affected? As in, not booting when on a dual boot machine or live image? | 19:12 |
xrogaan | I hate it when I have text editor trouble, and it removes part of my text. Sent an email with 'dpkg package.deb' instead of 'dpkg -i package.deb'. | 19:30 |
joerg | https://www-heise-de.translate.goog/hintergrund/Bootloader-Signaturen-per-Update-zurueckgezogen-Microsoft-bootet-Linux-aus-7250544.html?_x_tr_sl=de&_x_tr_tl=en&_x_tr_hl=en-US&_x_tr_pto=wapp | 19:32 |
joerg | for those who got trouble with german language in that article | 19:33 |
joerg | and yeah, sorry, as golinux already said | 19:34 |
debdog | the question was whether devuan is affected by that. | 19:36 |
joerg | I don't see a reason why devuan would be immune to that | 19:43 |
* joerg just reads "secure boot2 and feels nausea | 19:50 | |
joerg | s/2/"/ | 19:50 |
___used | They basically found one (?) version of grub which could "install malware"?! and put all the grub signatures on the DBX crl list | 19:51 |
___used | I suggest reciprocation, put all Windows sigs on the DBX crl list and push a linux update >;-) | 19:51 |
___used | Taking it to #offtopic | 19:51 |
brocashelm | The following signatures were invalid: EXPKEYSIG BB23C00C61FC752C Devuan Repository (Amprolla3 on Nemesis) <repository@devuan.org> | 19:52 |
brocashelm | ^ just got this while updating | 19:52 |
___used | neat. Not. And you can't secure boot? | 19:52 |
joerg | please see /topic ! | 19:54 |
joerg | apt-get update keyring errors? See https://dev1galaxy.org/viewtopic.php?pid=37309#p37309 | 19:54 |
joerg | it's probably not helping to discuss it here, read the thread linked in topic | 19:56 |
brocashelm | ok, just installed the new keyring and it worked | 19:58 |
joerg | :thunbs-up: | 19:58 |
mshinas | Thanks guys for your hard work, I applied the solution from https://dev1galaxy.org/viewtopic.php?pid=37309#p37309 and it worked for me!:] | 20:01 |
brocashelm | same here | 20:02 |
bb|hcb | About the expired key - devuan-keyring 2022.09.04 has made it to the repos (maybe not on all mirrors yet). The best way to fix it is: | 20:25 |
bb|hcb | apt update --allow-insecure-repositories && apt install devuan-keyring --allow-unauthenticated | 20:26 |
bb|hcb | Please someone update the forum post - this avoid the manual step and does not rely on a fixed version of the keyring package | 20:26 |
rwp | xrogaan, Please see bb|hcb's update above. ^^ | 20:34 |
brocashelm | bb|hcb: posted :) | 20:35 |
bb|hcb | :) | 20:35 |
iceman1 | hi | 20:37 |
iceman1 | back | 20:37 |
iceman1 | did everything went ok? | 20:37 |
Xenguy | Thanks brocashelm | 20:41 |
brocashelm | np | 20:43 |
iceman1 | was the package signed for this? wget http://deb.devuan.org/devuan/pool/main/d/devuan-keyring/devuan-keyring_2022.09.04_all.deb do we have a check sum for it? sorry to ask a lot | 20:50 |
joerg | brocashelm: private message? | 21:05 |
D-HUND | what's the state with the signatures? in case it's fixed yet, do I have to run something? | 21:15 |
joerg | http://reisenweber.net/irclogs/libera/_devuan/_devuan.2022-09-03.log.html#t2022-09-03T20:25:58 | 21:15 |
D-HUND | ok, that worked, thanks! | 21:20 |
joerg | [3 Sep 2022 20:25:58] <bb|hcb> About the expired key - devuan-keyring 2022.09.04 has made it to the repos (maybe not on all mirrors yet). The best way to fix it is: | 21:21 |
joerg | [3 Sep 2022 20:26:01] <bb|hcb> apt update --allow-insecure-repositories && apt install devuan-keyring --allow-unauthenticated | 21:21 |
joerg | iceman1: I noticed your question. Still checking what's the situation | 21:27 |
iceman1 | @joerg ok, should we post the apt update ... in reddit | 21:28 |
joerg | please not yet | 21:29 |
joerg | no idea if this helps in any way https://termbin.com/wa1z it's the md5sum I see here, doesn't mean this is correct and in any way certified, I could not tell from here | 21:34 |
fsmithred | joerg, I got the same md5sum and I downloaded the package shortly after it went into pkgmaster | 21:36 |
joerg | fsmithred to the rescue :-) Glad you chime in | 21:37 |
joerg | weird I can't reproduce same checksum via https download | 21:39 |
fsmithred | ??? | 21:39 |
fsmithred | I used wget http://... | 21:39 |
fsmithred | no | 21:39 |
fsmithred | https://... | 21:40 |
joerg | prolly PECAK, I might need more coffee | 21:40 |
joerg | PEBCAK | 21:40 |
fsmithred | I do get the same. Just downloaded it again in browser this time. Same md5sum. | 21:41 |
joerg | ugh indeed, I didn't notice the 404 | 21:41 |
joerg | fsmithred: https://termbin.com/uhz2f | 21:46 |
rwp | The deb.devuan.org is unsuitable for https due to it being a round robin of many other servers behind it. | 22:04 |
rwp | For https one would need to specify a server name that matches a particular server. Otherwise it is http for deb.devuan.org round robin dns. | 22:05 |
fsmithred | rwp, I used pkgmaster.devuan.org for both downloads (1 wget, 1 browser) | 22:27 |
joerg | rwp: many thanks :-) absolutly plausible | 22:27 |
brocashelm | i use beard.ly as a secondary mirror, just in case deb.devuan.org fails (usually the blendbyte server in taiwan) | 22:30 |
rwp | I am currently seeing that mirrors are not yet all in sync yet. Which is expected. So at this moment I can only reliably use pkgmaster for it. | 22:32 |
rwp | But I don't want to use pkgmaster all of the time. Since that would overload it. | 22:33 |
rwp | I am not in am emergency here. I have one copy of the deb. I can manually install it around. Or! I can wait a few hours and I am sure the mirrors will be in sync then. | 22:33 |
blastwave | test | 23:15 |
blastwave | Devuan Repository GPG key expired today ? | 23:15 |
blastwave | apt-key adv --recv-keys --keyserver keys.gnupg.net BB23C00C61FC752C | 23:15 |
blastwave | that says the GPG keys expired today | 23:16 |
sixwheeledbeast | read topic? | 23:17 |
brocashelm | blastwave: solution provided by bb|hcb: http://reisenweber.net/irclogs/libera/_devuan/_devuan.2022-09-03.log.html#t2022-09-03T20:25:58 | 23:21 |
fsmithred | or read this post: https://dev1galaxy.org/viewtopic.php?pid=37315#p37315 | 23:37 |
AndroUser | I'm trying to connect to a Hotel WiFi network with no password but despite that ESSID network being scanned by iw, it refuses to connect via iw nor nmcli at all. | 23:44 |
AndroUser | I'm typing this on my Android device. | 23:45 |
gnarface | anything in the log? | 23:45 |
AndroUser | can you elaborate more in regards to the log? | 23:45 |
gnarface | check /var/log/daemon.log for error messages coming from wpa_supplicant or your dhcp client | 23:46 |
gnarface | if you tail it (tail -f /var/log/daemon.log) there should be some noise from both during your connection attempt. you might need to edit the configs for both to increase verbosity to get anything meaningful. | 23:47 |
AndroUser | https://ibb.co/1XBhJLP | 23:50 |
AndroUser | here's my output | 23:51 |
AndroUser | after I tailed daemon.log | 23:51 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!