[00:10] Hi, does anyone know what was the unofficial deb making tool that allowed you to create debian packages while compiling from source ? I think it contained the words deb build and check [00:12] bbblowjob: Was it for any language specifically? [00:13] bbblowjob: that sounds a bit like checkinstall [00:13] I was thinking https://github.com/nylas/make-deb [00:13] Examknow, no it was a one size fit all type of software [00:13] Oh [00:13] sarnold : this actually rings a bell man, lemme check it out [00:14] sarnold, It is , actually !!! thanks a lot ! [00:14] yay! :) [00:15] bbblowjob: the next choice is something a little different, give it a look too https://fpm.readthedocs.io/en/latest/intro.html [00:17] thanks sarnold I'll give it a try [00:17] Hi, after a while, logs stop showing e.g. journalctl -u mysql shows nothing, until i reboot the server. and systemctl status mysql, or systemctl status systemd-journald, both show the message "Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable." --- What do I do [00:32] sarnold, how is this any better than checkinstall ? [00:33] bbblowjob: iirc checkinstall does some ptrace / debugging kinds of things to try to spot what's been written, where, and that kinda worries me for reliability reasons; iirc the fpm tool is more like traditional declarative packaging, just much less frustrating [00:34] thanks sarnold, I tried both on a folder rn and I prefer checkinstall tbh, more accurate descriptions and the package actually contains everything I need [00:34] bbblowjob: cool cool, that's good enough :) I just wanted to make sure you knew there were options [00:35] sarnold, yes thanks man :) [01:11] Is there a persistent way of disabling services ( specifically tracker* services )? [01:12] When I set the services to "tracker*.service.disabled" they get replaced when I release-upgrade [01:16] DarkTrick: you could mask them, that's like disabling them, but a bit more stern :) [01:16] sarnold, what does "mask" mean, specifically? [01:17] overwrite the service file with an empty file? [01:17] DarkTrick: these days there's a systemctl mask subcommand [01:17] DarkTrick: once upon a time you had to make the symlinks to /dev/null yourself [01:19] sarnold, man says "option to only mask temporarily until the next reboot" [01:21] DarkTrick: yes, if you use the --runtime option [01:21] If I get it right, it would just create a link with the name of the service file, that's redirected to /dev/null [01:21] yes [01:22] sarnold, thank you! === johuck2 is now known as johuck [01:25] is wacom supported right out of the box for ubuntu 20.04 [01:26] sarnold, essentially systemctl creates a link in /run/systemd/system. Is my assumption correct, that files in /run/systemd/system "overwrite"/"mask" files in /usr/lib/systemd/user ? [01:26] i notice it is not smooth [01:26] pi0 I have a tablet here, it works fine so far [01:26] pi0, Though, I don't use the buttons [01:26] i do not use the buttons and its is fully charged [01:26] and its lagging when i draw [01:27] ubuntu with gnonme [01:27] pi0, Hm, never experienced that. How about if you're using the cord? [01:27] bluetooth [01:29] pi0 With wacom you should also have the option to connect is via USB [01:30] DarkTrick: here's a table of where the unit files can be located and the default order in which they override each other: https://www.freedesktop.org/software/systemd/man/systemd.unit.html#Unit%20File%20Load%20Path [01:31] DarkTrick: I expect the order on ubuntu is the same, but I haven't checked myself [01:32] wow what a major difference with usb [01:32] i did not consider that [01:32] is there a tweak tool for it [01:32] to configure buttons and such [01:37] pi0, I'm glad that it works with USB. However, it SHOULD work properly with bluetooth. Maybe it's worth filing a report? As reference: In case you use Xorg, I also experience regular problems with non-std I/O devices. [01:37] pi0, I don't know of any tool, but maybe some else does [01:39] sarnold, thank you for the list! Very helpful! [01:39] An interesting lot of paths, though... especially the run/etc/usr-differences are... too much imo... but well... [01:44] Question: in unattended-upgrades, does "${distro_id}ESMApps:${distro_codename}-apps-security"; or "${distro_id}ESM:${distro_codename}-infra-security"; auto install kernel updates as part of security updates? [01:44] DarkTrick: yeah, I can understand -- however, eg the apparmor system defined paths back around 2000 or so, and just used /etc same as everything else did at the time [01:45] DarkTrick: and imho it's been a PITA ever since -- it'd be so much nicer to have a well-defined override mechanism, or an add-in mechanism, like systemd's designed. systemd's design feels super-complicated at first but it solves a *lot* of problems and does so with a simple consistency [01:47] sarnold, I'm relieved I'm not the only one. Btw, what is a "PITA"? [01:48] I guess it's no "is a family of yeast-leavened round flatbreads baked from wheat flour" :D [01:48] from "apt list --installed" some say "installed,automatic" what does this mean? [01:48] systemE: esm-infra-security has some kernel updates in it, yes [01:48] DarkTrick: PITA == Pain In The Ass :) [01:49] :D thank you [01:49] I imagined something like "a change request" :D [01:49] systemE: apt keeps track of which packages were explicitly installed by the user and which ones were dragged in as dependencies, so when you remove packages later, it can say "these packages were auto-installed, should we remove them?" [01:51] sarnold: so installed,automatic does not mean it was automatically installed by unattended-upgradees-- totally different? [01:51] systemE: yeah I think the 'automatic' means it was installed as a dependency of something else [01:53] sarnold: Ok, thank you. So I have this issue at work, our journald logs for our services stops working, e.g. journalctl -u mysql shows and systemctl status mysql, or systemctl status systemd-journald, both show the message "Warning: Journal has been rotated since unit was started. Log output is incomplete or unavailable." [01:53] Logs return when the server is rebooted [01:53] (until it happens again) [01:54] after reading https://community.centminmod.com/threads/error-journal-has-been-rotated-since-unit-was-started.16664/ it appears that maybe the kernel is auto updating and breaking the logger? [01:54] uname -a shows Linux db1 4.15.0-112-generic #113-Ubuntu SMP Thu Jul 9 23:41:39 UTC 2020 x86_64 x86_64 x86_64 GNU/Linux [01:55] ls -al /vmlinuz shows lrwxrwxrwx 1 root root 31 Oct 14 14:45 /vmlinuz -> boot/vmlinuz-4.15.0-121-generic [01:55] systemE: I think that question about kernel updates in that thread is misguided [01:55] Could I be onto something? [01:55] systemE: it is important to reboot into kernel updates periodically, yes, but it shouldn't be related to the usual journald log rotations [01:55] any ideas? [01:59] systemE: honestly it's probably juts a matter of making sure there's loads of storage space avalable for the /var/log/journal filesystm, to reduce the frequency / time / etc of journal rotations [01:59] 01:49 < DarkTrick> I imagined something like "a change request" :D [02:00] pi0, sorry that sentence was about "PITA", not addressed to your issue [02:05] sarnold: 8.2G free space, there's only one file in /var/log/journal [02:05] the file in in journal is 1.4 G [02:08] journalctl -u systemd-journald shows, -- Logs begin at Thu 2020-10-01 10:49:50 CDT, end at Tue 2020-10-27 21:06:46 CDT. -- Oct 01 10:49:50 db1 systemd-journald[510]: System journal (/var/log/journal/612892249edc4cf7b40cdfd1534feded) is 1.4G, max 1.4G, 18.0M free. [02:08] sarnold: Could this be the problem ^ ? [02:08] systemE: hmm, I would have expected that to be closer to 4g if it's rotating already .. check journald.conf(5) for the full set of things you can tweak [02:09] systemE: yeah that's itexactly! :) [02:09] WOOT [02:09] my boss is gonna be happy [02:09] thanx sarnold . Any ideas how to resolve this? === zbenjamin_ is now known as zbenjamin [02:11] systemE: the journald.conf settings SystemMaxUse= and SystemKeepFree= are my first guesses [02:12] systemE: I don't know how to dump the current settings :( [02:29] sarnold: so according to arch wiki, "If the journal is persistent (non-volatile), its size limit is set to a default value of 10% of the size of the underlying file system but capped at 4 GiB." Now our /var/log/journal is on a 15G disk, so 1.4G is the max the journal can be, confirming the wiki. [02:30] sarnold: I'm surprised the journal just "stops working" when it maxes out its size? Why doesn't it rotate and archive or something, do you know? [02:33] systemE: it does [02:33] systemE: because it throws away old contents it shows you the warning about incomplete output [02:36] but it "stops working" e.g. currently typing systemctl status mysql shows NOTHING [02:43] "SystemMaxUse= and RuntimeMaxUse= control how much disk space the journal may use up at most." "...defaults to 10%...of the respective file system, but each value is capped to 4G." [02:50] systemE: perhaps mysql just hasn't emitted anything recently? [02:50] systemE: can you cause mysql to emit messages on demand to test? [02:55] seems unlikely considering we have tons of stuff like mattermost,redmine and dozens of drupal/civicrm websites utilizing this service [02:57] I don't actually know mysql, what kinds of things would it print to the journal once it's started and running? [02:58] I'd sort of expect it to emit a bunch of tihngs like limits, configs, listenings ports, etc at startup, and then I'd expect it to be silent until something goes wrong [03:26] Do we still need to install something to watch netflix on ubuntu (18.04)? Or has that changed? [03:31] google-chrome [03:33] systemE: Is that intende for me or someone else? [04:27] blahboybaz: nope, you [04:27] That's what I always did anyways, netflix, amazon prime, etc, google-chrome [05:22] mgedmin: FYI, my two-line patch worked. My battery indicator hw is still misbehaving, but at least it no longer shuts down my machine. Thanks for the tips. [05:22] Do you know if it's worth trying to upstream my fix (or something similar)? And if so, where? [05:38] jmspeex: do you have a current bug filed in ubuntu about it? === PowerTower_121 is now known as PowerTower_120 [07:13] Hi. If something says "Requires: Java 1.8_92+", what package of java do I use to run it? [07:21] Industrial: if this software is from the ubuntu repos, the dependencies will sort that out [07:22] ducasse: no it's not. It's https://minion.mmoui.com/?download [07:22] So a jar file [07:22] I tried the latest openjdk but it's giving me errors, quite possibly a too new version of java [07:23] Error: Could not find or load main class gg.minion.Minion [07:23] Caused by: java.lang.NoClassDefFoundError: javafx/application/Application [07:23] That's what I get [07:23] in that case we don't support it, so you should ask whoever provides the software which java version(s) it works with [07:25] it says so .. Java 1.8_92+ .. [07:26] if it says 1.8.92+ i would guess it works with any version higher than that, which is obviously not the case [07:26] So whats the difference between openjdk 8 vs openjdk 14 vs "java 1.8" ? [07:30] according to this it needs java 8 - https://www.reddit.com/r/elderscrollsonline/comments/fg37dp/minion_305_java_wont_start_in_linux/ [07:30] nothing later [07:32] Industrial: ^^ [07:46] ducasse: thanks [07:49] yw [07:54] I make an nfs share from 1 ubuntu to another [07:55] are you sure you created the nfs share successfully? [07:56] do we have a video viewer/editor capable of single frame scrub forward/backwards AND ability to zoom/pan around a UHD (4K) frame to look at detail? VLC, Parole and Openshot-qt don't [07:56] Industrial: you need to set the classpath [07:56] the folder on the server belongs to userid 1001 . on client that user is uid=1000 . On server in etc/exports: /home/foobar 192.168.X.Y *(all_squash,rw,anonuid=1001,anongid=1001) [07:56] but when I mount on client, I only get a readonly file system [07:57] I guess I am missing a step with user mapping ? [07:57] ducasse: I have done something wrong, but I'm not sure how to fix it. https://gist.github.com/Industrial/45287c7cebde0ab3db3f8fb6348d2c9d [07:57] TJ-: I'm not sure how to do that [07:59] TJ-: maybe https://shotcut.org/features/ ? [07:59] TJ-: -classpath . or $PWD ? doesnt seem to work :-) [08:01] Industrial: that pretty much tells you what to do [08:03] Industrial: classpath is not a /search/ path - you may need to put multiple directory paths on it [08:03] tatertots: do you have an idea what is going wrong with my mount options? === Scotty_Trees2 is now known as Scotty_Trees [08:09] renn0xtk9: do you have access to the nfs server right now to try different options? [08:11] tatertots: yes [08:11] renn0xtk9: and your subnet mask is most likely /24 [08:12] renn0xtk9: x.x.x.x/24 [08:12] yes it is I thought it was optional [08:13] renn0xtk9: try with the following syntax [08:14] renn0xtk9: /home/foobar x.x.x.x/24(rw,sync,no_root_squash) [08:14] renn0xtk9: you'll need to exportfs -ra afterward [08:16] renn0xtk9: if you do exactly as above without deviation it's expected to be read/write [08:17] tatertots: hmm no it says touch: cannot touch '/mnt/dell/a': Read-only file system [08:18] renn0xtk9: then you're probably making some mistake client side [08:18] Hi. I moved /var/lib/snapd to another place and create a soft link to the var/lib/. After reboot I found that all snapd apps cannot lanuched via 'snap run'. And all mounts were lost in the report of 'df -h'. Now I'm using 'mount --bind new_path /var/lib/snapd' to mount the new path to the old one. I found that reinstall a snap app can make it work work again. I wonder that if there any easier way to repair all snap apps? [08:18] If think the fact that users on client is ID=1000 and that the folder on server belongs to UserID=1001 a problem no ? [08:18] Good morning. My ethernet stopped working 2 days ago (ubuntu 20.04, I217-V). Tries to connect for a couple seconds then notifies "Activation of network connection failed". Hardware is fine (works in Live Distro and Windows). Manual configuration fails, manually compiling current and older versions of the e1000e module failed. Any pointers what else [08:18] I could try? [08:19] hi all, I can ask about apache installed in a ubuntu VM? [08:19] tatertots: client side I mount like that sudo mount -t nfs -o rw,nolock,udp 192.168.178.29:/home/foobar /mnt/dell/ [08:20] renn0xtk9: in terminal> cat /proc/fs/nfs/exports|nc termbin.com 9999 [08:21] renn0xtk9: share url/link here [08:23] tatertots: https://termbin.com/kt9t <- it is still ro in there . weird :S [08:24] renn0xtk9: that's why it's read only [08:24] hmm but where does it come from [08:24] renn0xtk9: your configuration [08:25] hmm but I do write rw in /etc/exports [08:26] renn0xtk9: the current config is read only and contradict or opposite of the advice i give you for r/w "NO" root squash [08:27] lotuspsychje: I don't. Was just trying to scratch my own itch. Mostly, I'm not sure if it would be considered a bug of a feature. As background, the problem is that a hw glitch makes it impossible to read the BAT0 status, which makes upowerd conclude that the battery is at 0%, triggering an immediate shutdown. [08:27] tatertots: I have rw sync no_root_squash in there.. [08:27] renn0xtk9: in terminal> cat /etc/exports|nc termbin.com 9999 [08:28] https://termbin.com/rjtf [08:28] jmspeex: well if something doesnt work by default on ubuntu, i would consider that as bug, would that be your case? [08:29] renn0xtk9: in terminal> sudo exportf -ra [08:29] renn0xtk9: in terminal> sudo exportfs -ra [08:30] renn0xtk9: completed? [08:31] yes [08:32] renn0xtk9: in terminal> sudo systemctl restart nfs-server [08:33] done [08:33] renn0xtk9: in terminal> cat /proc/fs/nfs/exports|nc termbin.com 9999 [08:34] https://termbin.com/otmu7 [08:35] renn0xtk9: in terminal> sudo exportfs -ra [08:35] renn0xtk9: in terminal> cat /proc/fs/nfs/exports|nc termbin.com 9999 [08:35] https://termbin.com/cdn1 [08:36] renn0xtk9: in terminal> sudo systemctl status nfs-server|nc termbin.com 9999 [08:38] https://termbin.com/qs9t [08:39] If you start an ubuntu 20 desktop version without monitor and arrive at login screen you shold press alt+tab to be able to click on your user and input the password... IT SHOULD BE available with ESCAPE or ENTER better === rubdos_ is now known as rubdos [08:42] renn0xtk9: in terminal> stat /home/mhaselbauer|nc termbin.com 9999 [08:42] https://termbin.com/5n4em [08:44] renn0xtk9: you can't do that [08:45] renn0xtk9: do this [08:46] renn0xtk9: in terminal> sudo mkdir /var/test [08:48] renn0xtk9: in terminal> sudo chmod 777 /var/test [08:48] renn0xtk9: completed? [08:49] done [08:50] renn0xtk9: /var/test x.x.x.x/24(rw,sync,no_root_squash) [08:50] renn0xtk9: you know where that should go right? [08:51] yes [08:51] renn0xtk9: save changes [08:51] renn0xtk9: in terminal> sudo exportfs -ra [08:52] renn0xtk9: in terminal> sudo systemctl restart nfs-server [08:53] renn0xtk9: in terminal> sudo systemctl status nfs-server|nc termbin.com 9999 [08:53] Do you agree with "if you start ubuntu 20 desktop version without monitor and arrive at login screen you shold press alt+tab to be able to click on your user and input the password... IT SHOULD BE available with ESCAPE or ENTER better" [08:53] monitor can be unplugged for example or maybe a failure graphics card of first monitor [08:54] there should be an easy click to "know" you can type the password [08:54] tatertots: https://termbin.com/uic0 [08:55] renn0xtk9: in terminal> cat /etc/exports|nc termbin.com 9999 [08:56] https://termbin.com/3xuh [08:57] tatertots: I will be unavailable for ~30 min [08:57] renn0xtk9: wrong [08:57] renn0xtk9: you didn't enter the correct syntax i gave you [08:57] renn0xtk9: do NOT deviate [08:58] renn0xtk9: human error [08:58] renn0xtk9: there's nothing technically wrong besides the fact that you insist on injecting incorrect syntax [08:59] hmm ok sorry [08:59] where is incorrect? [08:59] renn0xtk9: no where did i ever use "*"...you insist on using it and i demand that you stop [08:59] renn0xtk9: /var/test x.x.x.x/24(rw,sync,no_root_squash) [09:00] no where in that does "*" appear...it comes from YOUR brain...and shouldn't exist [09:00] Ah sorry I :(( [09:00] no should there be any "white space" [09:00] nor should there be any "white space" [09:00] I will remove and retest afterwarsd [09:00] I am busy now [09:00] no "white space" [09:00] thanks a lot for your help ! [09:02] I always thought the * was to restrict incomming username (and * would mean no restriction basically) [09:03] renn0xtk9: /var/test 192.168.178.24/24(rw,sync,no_root_squash) [09:04] renn0xtk9: don't think...simply create that exact line and save it [09:08] renn0xtk9: completed? [09:10] done [09:12] Then I restart nfs-sever and exportfs -ra right? [09:13] renn0xtk9: in terminal> sudo exportfs -ra [09:14] done [09:14] renn0xtk9: in terminal> sudo systemctl restart nfs-server [09:14] renn0xtk9: in terminal> sudo systemctl status nfs-server|nc termbin.com 9999 [09:14] done [09:16] https://termbin.com/zinh [09:18] renn0xtk9: in terminal> cat /proc/fs/nfs/exports|nc termbin.com 9999 [09:19] https://termbin.com/nxbf [09:21] is there a difference beteween nfs-server and nfs-kernel-server? [09:24] renn0xtk9: in terminal> sudo systemctl restart nfs-kernel-server [09:24] renn0xtk9: in terminal> sudo exportfs -rav|nc termbin.com 9999 === OEIRAS is now known as OERIAS [09:26] https://termbin.com/9k7j [09:27] renn0xtk9: in terminal> rpcinfo -p localhost|nc termbin.com 9999 [09:29] one question about fan networking https://wiki.ubuntu.com/FanNetworking where do these names come from? fan-250-0-127 [09:29] and how to they relate to examples [09:29] https://termbin/hupj [09:29] renn0xtk9: in terminal> sudo cat /proc/fs/nfs/exports|nc termbin.com 9999 [09:30] renn0xtk9: that link is broken, but easily fixable [09:30] StyXman: they're in-addr.arpa (reverse order) so 250-0-127 would be 127.0.250.0/24 [09:30] TJ-: so they are not related to the previous examples at all, right? [09:32] ? i have a ubuntu 20.04 lts running in virtualbox [09:32] i changed the networking from NAT to a Bridged connector [09:32] now i cannot update with `apt update' i get a lot of hash sum mismatches [09:33] that sounds more a vbox bug than *buntu's [09:33] OK but i got archlinux working the same way [09:33] space1: can you restart the networking service? sudp systemctl restart networking [09:33] sudo*, or use NM if you have a desktop [09:34] tatertots: https://termbin.com/6ltam [09:34] StyXman: i get failed to restart networking.service not found [09:34] StyXman: not sure but that article infers that for LXD there's a fan bridge using 127.0.250.0/24 or similar [09:35] renn0xtk9: in terminal> rpcinfo -p localhost|nc termbin.com 9999 [09:35] space1: is the VM using Ubuntu desktop with GUI, or ubuntu server? [09:35] TJ-: yeah, that's what confusing me [09:35] StyXman: me too! [09:35] TJ-: it is using the default GUI [09:35] i haven't changed anything [09:36] space1: ok, so you'd do "systemctl restart NetworkManager" [09:36] space1: desktop uses NetworkManager not ifupdown (networking.service) [09:36] TJ-: OK good to know [09:36] i am not used to debian based distros yet [09:37] StyXman: may be some clues here; there's mention of fan and defaults https://lxd.readthedocs.io/en/latest/networks/ [09:37] TJ-: I didn't know the service was not installed, I use ubuntu at work and debian at home and never noticed the fdifference [09:39] TJ-: yeah, thanks, I wanted to start at the bottom, but yes, I'm aiming for LXD config [09:39] maybe there are clashing mac addresses? [09:40] tatertots: https://termbin.com/yyx1 [09:40] StyXman: also, this is generally informative as to how to do fans, though I don't see that specific subnet mentioned http://manpages.ubuntu.com/manpages/bionic/man8/fanatic.8.html#use%20with%20lxc/lxd [09:40] StyXman: TJ-: your advice didn't help i think there might be a MAC address clash somewhere [09:41] space1: check the log: "journalctl -b 0 -u NetworkManager -n 100" to review that last 100 messages - NM is very verbose and helpful in logs [09:41] renn0xtk9: now mount the nfs share on the client using ONLY "mount -t nfs x.x.x.x:/var/test /client/mount/point"....do NOT deviate [09:43] renn0xtk9: now mount the nfs share on the client using ONLY "mount -t nfs x.x.x.x:/var/test /mnt/dell"....do NOT deviate [09:43] tatertots: I have to use sudo otherwise it says mount: only root can use "--types" option [09:44] renn0xtk9: yes use sudo..that's acceptable [09:44] than it works [09:45] renn0xtk9: did you mount on client already? [09:46] tatertots: yes [09:46] renn0xtk9: in terminal on client> mount|grep nfs|nc termbin.com 9999 [09:47] https://termbin.com/zly1 [09:47] i have a package named "smartgit", why doesn't this command match smartgit? > apt purge "*smartgit*" [09:48] renn0xtk9: in terminal on client> touch /mnt/dell/test [09:48] done [09:48] renn0xtk9: in terminal on client> ls -lh /mnt/dell/test|nc termbin.com 9999 [09:49] renn0xtk9: in terminal on client> ls -lh /mnt/dell/|nc termbin.com 9999 [09:49] https://termbin.com/bqgm [09:50] renn0xtk9: this nfs share is r/w [09:50] yes [09:50] renn0xtk9: agree? [09:50] absolutely [09:51] renn0xtk9: you cannot use https://termbin.com/5n4em [09:51] hansh: i dont see smartgit in the repos, did you download it via external ppa's? [09:52] tatertots: however when I go on server the /var/test/test has the userid of another user so I cannot write into it with my user [09:52] renn0xtk9: you have to configure id mapping [09:52] lotuspsychje, no i downloaded it via a web url, https://www.syntevo.com/downloads/smartgit/smartgit-20_1_5.deb [09:52] tatertots: I though it would be possible to use https://termbin.com/5n4em using id mapping [09:52] Tj yes that is what I want [09:52] i found the solution guys https://askubuntu.com/a/1242739 [09:53] hansh: perhaps there's something in the dir, try ' echo "*smartgit*" ' to see how bash is expanding it [09:53] EriC^^, i tried that, so i wrote "sma" and tried tab-autocomplete, tab had nothing to suggest [09:53] I don't remember if apt purge uses glob patterns, regexes, or what [09:53] (implying that there was nothing starting with the name "sma" ) [09:53] try apt list|grep smartgit, to see if it's even installed? [09:53] hansh: https://www.syntevo.com/smartgit/download/#installation-instructions [09:54] renn0xtk9: sudo unmount /mnt/dell [09:54] renn0xtk9: sudo umount /mnt/dell [09:54] lotuspsychje, i know how to install it. i've been using bleeding-edge beta versions, now i want to go back to the stable versions, so i uninstalled it an re-installed it [09:54] tatertots: done [09:54] hansh: what does 'dpkg -l | grep smart' give? [09:55] EriC^^, https://paste.debian.net/plain/1168909 [09:56] renn0xtk9: in terminal> sudo chown nobody:nogroup /var/test [09:56] hansh: suround it by quotes '*smartgit*' [09:56] renn0xtk9: in terminal> stat /var/test|nc termbin.com 9999 [09:56] otherwise bash might be expanding that to some file names [09:57] also, please report the rror message too [09:57] tatertots: http://termbin/com/s0ei [09:57] renn0xtk9: typo [09:58] renn0xtk9: double check your link [09:58] hansh: try 'sudo apt-get purge smartgit' [09:58] tatertots: http://termbin.com/s0ei [09:58] ahm https://paste.debian.net/plain/1168910 [09:58] hansh: tabcompletting 'sma' won't work if there is a file whose name *contains* smartgit, but doesn't *start* with it [09:59] renn0xtk9: in terminal> sudo exportfs -rav [09:59] renn0xtk9: in terminal> sudo systemctl restart nfs-kernel-server [09:59] StyXman: really it should though, it's completing the packages name not files in dir [09:59] EriC^^, yes that works. i'm just surprised because "*smartgit*" didn't work [09:59] renn0xtk9: mount on client [10:00] the folder has nothing to do with it, i think; https://paste.debian.net/plain/1168911 [10:00] tatertots: done [10:01] renn0xtk9: in terminal on client> touch /mnt/dell/test [10:01] renn0xtk9: in terminal on client> ls -lh /mnt/dell/|nc termbin.com 9999 [10:01] done [10:02] hansh: yup, odd behavior on apt's part [10:02] tatertots: https://termbin.com/5xln [10:03] hansh: for me it seems to work if i do apt remove "*package*" [10:03] ooh good catch! [10:03] renn0xtk9: in terminal on client> id|nc termbin.com 9999 [10:04] I didn't even think the glob could accidentally match something in $PWD [10:04] https://termbin.com/tuyjm [10:05] renn0xtk9: you CAN write/edit/modify test [10:06] tatertots: on client yes on server no [10:06] renn0xtk9: you can on server also [10:07] no [10:07] renn0xtk9: administrators can write all/any [10:08] but on sever I have id=1001 and filec created from client have id=1000 [10:08] renn0xtk9: and NFS "clients" are the consumers of NFS server shares [10:09] renn0xtk9: administrators are not prevented from access [10:12] renn0xtk9: the test nfs share is r/w, use what you learned when attempting to create shares in the future [10:12] I have a file: /var/lib/gdm3/core can i delete this file? it's huge [10:12] The issue is about usermapping tatertots [10:12] renn0xtk9: ok then map users [10:12] renn0xtk9: or sync users [10:13] file created by user "foo" on Machine A mut be seen as created by user "bar" on machine B [10:14] dexterfoo: find out what it is for [10:14] renn0xtk9: since you don't have a centralized directory like LDAP or Active Directory, it'll be a by hand manual process which aren't forgiving to typographical errors and or syntax [10:14] dexterfoo: deleting random files purely based on size is a fine way to completely break the system [10:15] renn0xtk9: consider some form of directory for auto mapping/sync of user id [10:16] what version of apt is 20.04 supposed to be running on? [10:16] !info apt focal [10:16] apt (source: apt): commandline package manager. In component main, is important. Version 2.0.2ubuntu0.1 (focal), package size 1259 kB, installed size 4182 kB === OERIAS is now known as CristobalColon [10:18] thanks; don't suppose there's any easy-ish way to upgrade to >=2.0.3? [10:18] > Recent versions of apt changed the way patterns are specified, and apt no longer supports regular expressions directly or wildcards, other than * for which support was restored in 2.1.0 and backported to 2.0.3. [10:20] hansh: use package versions meant for your ubuntu release [10:20] !mix | hansh [10:20] hansh: it is usually a very bad idea to mix packages from different releases (or Linux distributions), and it is completely unsupported [10:21] hansh: if you want, there is version 2.1.1 on ubuntu 20.10 [10:23] kk [10:24] renn0xtk9: do you have access to any directory services on your network ? [10:24] what is the 20.04 equivalent of 18.04 & 20.10's apt purge "*package*" [10:24] ? [10:27] (eg how do i write >>apt purge "*mysql*"<< in 20.04? that command should work fine in 18.04 and 20.10) [10:28] sudo apt purge packagename [10:28] same as always [10:29] hansh: why do you think it's different in 20.04? [10:29] Maik: I think he wants an alternative to using * because it apparantly doesn't work in 2.0.2 [10:30] ah, right... not fully awake yet. My bad [10:30] (yes ThinkT510 is correct) [10:31] you can leave out the () [10:31] hansh: do you even use ubuntu right now? [10:32] lotuspsychje, actually no, i'm using Xubuntu (20.04), i'll check if it's Xubuntu specific, have a Ubuntu 20.04 nearby as well [10:33] hansh: doesn't matter, it's still ubuntu only with the Xfce desktop. Hence the X [10:34] so it works the same [10:34] Maik, (actually with Xubuntu it's slightly more than that, at least if you install from the official Xubuntu cd, rather than the Ubuntu netinst cd) [10:34] hansh: apt is the same in both [10:34] fuuuu [10:34] too much lisp, sorry [10:35] why you type with ()? it's annoying [10:35] Maik, Lisp programming, https://i.imgur.com/XkBSx01.png [10:35] and hansh, after more than 13 years i know what Ubuntu and all of it's flavors are. [10:36] :) [10:36] anyway i just tested, it's exactly the same behaviour in Ubuntu 20.04, >>apt purge "*mysql*"<< does not match mysql-client [10:36] what i was trying to say is that the base is still ubuntu and work the same [10:37] ^^ [10:37] kindof unfortunate to break this in a LTS release imo (only to quickly fix it again in the next minor release) [10:38] I want to see a pastebin of apt purge "*mysql*" not working to believe it [10:38] hansh: is it only apt? or apt-get too? [10:39] hansh: and lsb_release -a && uname -a too :p [10:39]