[00:33] hey [00:34] hello [00:46] I'm not sure if this is the best place to ask this, but I'm having a strange problem with rsync. I've got DirA locally, and DirB on a NAS, which I have mounted using CIFS. When I run rsync, no matter what, it copies ALL the files from DirA to DirB, and sets the 'modified' time of the files in DirB to the current time. [00:47] The command I am using is `rsync -av /home/kn100/DirA/ /media/DirB/`but I have tried using --ignore-times and --checksum and get the same result. [00:49] kn100, first thing I'd check is that the clocks are set right on both the local system and the NAS [00:51] kn100: copying *all* the data from source to destination may make sense over a network filesystem, there's not a whole lot of point in copying all the data from the NAS to the local machine just to checksum it... [00:52] kn100: normally when moving data between two machines you'd rather use ssh or something similar, so an rsync process can run on the remote system, and do checksums over there, etc [00:52] sarnold, the --checksum flag was purely just me experimenting, I'm aware it's crazy to do that normally. I was just getting rather frustrated [00:52] kn100: ah, okay :) [00:54] rfm I think you may actually have hit the nail on the head there: https://kn100.me/Random-files/loltime.png . I'm definitely not in Beijing! [00:59] so, times now match up on client and server, going to let rsync copy all the files once more and see whether now maybe the issue is resolved! [01:01] kn100: it likely won't, i think it has to do with what 'modify' means and its relation to cross-filesystem copying [01:02] actually nevermind [01:03] It's a consumer grade NAS with an ext4 filesystem, and the client is also on ext4 - but I wonder if the CIFS stuff in the middle or some crappy software on the NAS is causing the problem. Know of any alternatives to rsync/other avenues of exploration - assuming this problem persists? [01:04] The use case is I use Shotwell on my Ubuntu box to sync photos to my local library on my desktop, and then run rsync nightly to copy them to my NAS, where they're accessed by a home server and indexed for Photoprism [01:04] it should work, i confused ctime with mtime [01:04] should I be expecting the ctime to match across the files? [01:05] ctime will be different cause it depends on the inode, but mtime should match [01:05] because it appears to be setting that to the current time too, mtime is 1 second later than ctime for the files after rsync has finished with them [01:05] oh [01:07] yeah, so on the client, the file was created sometime in feb, but has a modified time in 2020, but on the NAS it has a ctime of march 5th, along with a modified time of march the 5th :( [01:08] this seems relevant kn100 https://bugzilla.kernel.org/show_bug.cgi?id=198967 [01:08] bugzilla.kernel.org bug 198967 in File System "Modification times not preserved correctly" [Normal, Resolved: Code_fix] [01:08] the guy fixed it with rsync -avP though [01:09] I am running with -av currently, and -P is just progress - I wonder if this is something to do with how I've mounted the NAS or something [01:11] kn100: if you do rsync -avP file ...file1 && stat file1 does it show the correct time? [01:11] I just ran an update and initramfs said it will attempt to resume from /dev/sdb2, which is my /swap --- is that normal? [01:17] if i have a server with a single network card and I create a network bridge using netplan(I'm trying to enable bridge networking for kvm guests), how am i supposed to connect to the host if I have to disable dhcp4 for the card to enable the bridge? [01:18] dhcp4 on the bridge [01:19] I tried that and didn't get an address, so I assumed since multiple articles show dhcp4: false and dhcp6: false it seemed to be no coincidence [01:20] ah, possible. the clear majority of setups I've seen documented were similar, people seem to like static configs for their infrastructure things [01:20] EriC^^: apologies for how long this took: https://paste.ubuntu.com/p/dM4j2f3mw4/ [01:22] same output with `cp -p` too [01:22] hello, i am trying to use a low latency kernel for my system. however when I build the kernel from the focal git repo, it is not setting the kernel configuration option, and builds the dpkg w/o kernel preemption enabled. how do i enable low latency kernel from debian configs? [01:25] kn100: it seems here that setting the gid and uid in the mount options fixed it for this guy, https://unix.stackexchange.com/questions/281491/prevent-updates-to-modified-time-when-copying-files-to-a-mounted-samba-folder [01:27] the uid and gid I expect need to match the user using the mount on the client right? As in, I am kn100 on my computer, my uid and gid are 1000, an therefore in the cifs mounting options in fstab I should set uid=1000,gid=1000 [01:27] oingoboingo: if you're building the kernels yourself rather than using packaged kernels, it's probably easier to just skip the ubuntu-supplied sources entirely [01:27] yeah kn100 i think that's right [01:27] if so, that's how it's currently configured. In fact, `credentials=,cache=strict,uid=1000,gid=1000,file_mode=0664,dir_mode=0775,iocharset=utf8` are the mounting options I am currently using, if that's helpful to know [01:28] kn100: maybe playing with the cache option might help? [01:28] kn100: try cache=none just to see [01:29] I'll try removing it altogether. I previously had it set to cache=none a while ago, but that obviously kills performance - but I suppose right is better than fast! [01:29] sorry im just guessing at this point, never really used cifs [01:29] I really appreciate you helping! [01:29] no problem [01:29] because your guesses are far better than mine so far. I keep suspecting my crappy NAS, given the software it is running is terrifyingly insecure... [01:34] sarnold: i already tried on latest stable (5.16.12) but I have kernel panic w/ the identical boot config. but I will most likely try the same HWE version and build it from kernel src tree. thx will give it a try [01:34] oingoboingo: aww :( [01:35] same problem with cache=none, and also with no configuration change to cache at all. Let me try from a different client, perhaps the behaviour is different there [01:37] same behaviour on a different linux client. NotsurewhatIwasexpecting.jpg [01:39] at this point i am half tempted to just give up the encryption feature on my NAS and just mount the damn thing via NFS and hope that works better lol [01:41] kn100: this isnt a great solution, but you could do 2 runs, the 2nd being rsync --size-only -t file ..file1, it should not copy the file, but only the mtime [01:46] hey, interestingly, that DOES actually set the modified time correctly [01:47] that's so strange, must be a bug in rsync right [01:48] https://paste.ubuntu.com/p/TNdJM4YCJP/ - how strange [01:53] try to do 'while true; do stat file1 2>/dev/null && break; done' [01:53] then rsync -av file ..file1 [01:53] that should show the stat when its still barely started copying [01:54] aHA [01:55] i think it must be as you said something to do with the nas [01:55] I think I found something... rsync --itemize-changes tells you WHY rsync wants to change the file, and every time it says `.f...p.....` - which as far as I can tell means the permissions are different, and assumedly it wants to replicate the permissions between the two, which it probably isn't able to [01:56] gnome-shell: Failed to resolve address for stun.services.mozilla.com., why is gnome-shell still using the obsolete stun.service from mozilla [01:57] yep, think I got it! THe solution is --no-o - which means no-owner. Seems to now set the modify time, and I don't actually care about replicating the owner info in the NAS :D [01:57] ah great! [01:58] now to copy 50+gb across a 1gbit network for the fifth time tonight :D [01:58] hope this works [01:59] god damn it nope [01:59] you dont need to recopy them, if you want you could do rsync -vrt --size-only and have it only copy permissions [01:59] erm [02:02] it did work for 1 file though right? the -no-o [02:02] *--no-o [02:04] it may have... honestly it's 2 in the morning here and I'm struggling to think straight. -vrt --size-only ALMOST works to set the times for all files, except for 8 (out of 20k...) which is mega strange. [02:07] oooh even more interesting. the files it's failing on for some reason are perfectly duplicated, but one is called file.jpg, the other is called file.JPG... [02:08] decisions from the 90s strike again! [02:10] whatever, this works well enough. Thanks again everyone. I think I'll investigate this more deeply in the future and write something up on it, some funky business going on here for sure [02:11] yup, pretty odd === FlannelKing is now known as Flannel [03:50] im trying to use autokey. im not sure which channel to ask for help in so here i am. i set my hotkey and the command. it works from the notification bar but not when i use the hotkey. [03:55] oscurochu: try running journalctl -f in a terminal when trying it out; also maybe tail -f ~/.xsession-errors -- one of those might have handy error messages [03:59] nope. nothing useful [04:00] i get warnings only for discord. nothing happens when i press my hotkey [04:02] plus when i use the notification icon to do my macro, the pasted text gets scrambled. its really weird [04:05] /j #python [04:07] oscurochu: dang :( [04:07] Null_A: any chance you can fix your connection? [04:16] hello === jdmark2 is now known as jdmark === gdb_ is now known as gdb [05:40] assalamualikum === Jubes4 is now known as Jubes === scjosh8 is now known as scjosh [08:23] hi [08:24] ?/ [08:24] !ask | mint__ [08:24] mint__: Please don't ask to ask a question, simply ask the question (all on ONE line and in the channel, so that others can read and follow it easily). If anyone knows the answer they will most likely reply. :-) See also !patience [08:25] i am using mint in a live usb of 32 gig wha games can i run? [08:26] we only support ubuntu and its official flavours mint__ [08:26] !mint | mint__ [08:26] mint__: The Ubuntu channels can only provide support for Ubuntu and its official flavors, since other distributions and derivatives have repository and software changes. So please use their dedicated support venues, for example: Linux Mint (#linuxmint-help on irc.spotchat.org), Kali Linux (#kali-linux), and LXLE (#lxle) === pnwise__ is now known as pnwise [09:04] Hi [09:05] To create a template for VMWare vCenter, should I delete the file `/etc/netplan/00-installer-config.yaml`? === Jubes3 is now known as Jubes [11:40] scortal. [11:40] :-) [11:52] ubuntu. === q5812 is now known as deb01__ [12:28] To create an Ubuntu template for VMWare, should I delete the file `/etc/netplan/00-installer-config.yaml`? (based on the advice given here: https://docs.microsoft.com/en-us/azure/cloud-adoption-framework/manage/hybrid/server/best-practices/vmware-ubuntu-template ) [12:29] if they say so ? [12:55] vmware is it free? [12:55] np. [12:56] ehm, no. [13:00] scortal, no but there is virt-manager, virtualbox or gnome-boxes, all free === Jubes9 is now known as Jubes === lotuspsychje_ is now known as lotuspsychje [13:36] Anyone know why Ubuntu shits itself when I plug a second WiFi adapter into my computer? [13:38] Crappy FOSS drivers? [13:38] lets keep the language polie please Kolusion [13:39] *polite [13:39] lotuspsychje shut the fuck up you little pansy bitch [13:39] Kolusion: please stop it [13:39] Maik: Please mind your own business. [13:40] !ops [13:40] Help! Channel emergency! (ONLY use this trigger in emergencies) - CarlFK DJones el Flannel genii hggdh ikonia krytarik mneptok mwsb nhandler Pici popey sarnold tomreyn Unit193 wgrant [13:43] Back. I got disconnected. My network equipment is garbage. [13:44] Tp-Link [13:45] So why could Ubuntu be shitting itself over a network adapter? Terminal locks up until remove it from the USB port. [13:45] This suggests a driver issue. [13:46] 'sudo bash' just hangs and won't prompt me for a password until I remove the device from the USB port. [13:47] All network activity freezes as well. It's like the kernel is stuck with the driver. [13:47] Kolusion, could you reboot with usb wifi adapter on and see how is going on [13:49] Is this that great open source community I hear about that are here to help. Oh okay. [13:49] Windows users don't know what they're missing. [13:49] ledeni. If I plug in the WiFi adapter by itself, its fine. If I insert it after an existing on is plugged in, the problem happens. But had them both working at the same time earlier today. I am going to try plugging them in reverse order now..... [13:51] It made no difference. I am going to close down the virtual machine now and try again. [13:53] Kolusion, what i asking is turn off pc and plug in both adapters boot up like that [13:53] Same problem. I am now going to remove the device and plug it in again. I have seen WiFi adapters pack up like this before. [13:54] Oh hang on, it has unfrozen. [13:55] Success. If I plug in the troublesome one first, then the better one after, it works. [13:55] Strange. [13:56] Kolusion, great 👍 and be patience sometimes [13:57] ledeni. I was? [13:57] Kolusion, yes but language ... [13:58] ledeni what about it? [13:59] Kolusion i will not discuss about it [13:59] ledeni its better you don't. You would only be wasting your life. :) [13:59] Thanks for your time in assisting. [14:00] It seems the bad driver was causing a denial of service with kernel I/O. [14:01] It is a community driver. [14:01] But in saying that, official Tp-Link drivers are terrible. [14:03] When I am using aircrack-ng, I run out of memory and Ubuntu starts killing apps before the host virtual machine VirtualBox crashes. Will running 'sync' at the end of my loop script fix that? [14:05] What's up with everyone being so offended by the swearing? Do you guys realise this is the internet, right? If you talk like this on a Windows forum or something, people will laugh at you. [14:06] see the rule and guidelines [14:06] That won't be happening. [14:07] then just behave :) [14:07] I am. [14:08] Expecting zero swearing in a support chat just isn't going to happen. Especially when it is related to Linux. [14:08] People get frustrated. [14:09] Well, I am going now. Bye all. [14:10] Hi all === BlessJah_ is now known as BlessJah [15:55] hi [17:43] h [17:43] i [18:03] tap tap this thing on? [18:04] Yes. [18:06] hello guys [18:07] im creating a shell script [18:07] and wanted some help on ot [18:07] it [18:07] please [18:21] Type your question [18:21] You may need to be a little more specific. === coc0nut_ is now known as coc0nut [19:33] hi. [19:40] I am unable to access any websites via Firefox or Google Chrome on my Ubuntu MATE install. I am able to ping 8.8.8.8 but cannot ping www.google.com. I have also noticed that apt cannot access the internet either. I have a feeling that somehow the ProtonVPN Client might have corrupted something. [19:41] Paddy_NI, does dns work? [19:41] When I do "ip addr" I noticed a connection called "ipv6leakintrf0". [19:41] JoeBk: How do I test that? [19:41] also, get a real vpn [19:42] nslookup google.com [19:42] JoeBk: I used PIA for years which worked just fine and recently switched to Proton [19:42] I used to have PIA [19:42] I switched to expressvpn [19:43] does nslookup work? [19:43] nslookup google.com tells me that the connection times out, no servers could be reached [19:43] not good [19:44] I use pihole running on a RPi3b plugged in to my router, every other device on my network works perfectly [19:44] nslookup google.com 8.8.8.8 [19:45] That instantly returned a result [19:47] Paddy_NI, see if you can configure DNS. [19:47] I've never had to do it on ubuntu. [19:49] Yeah I tried manually using 8.8.8.8 and 1.1.1.1 and it's still the same. [19:52] Paddy_NI, what is the same? working? [19:52] Not working [19:52] did you configure DNS with 8.8.8.8? [19:53] Yes I also tried 1.1.1.1 [19:53] I don't know, I'm far from being an expert on Ubuntu. [19:54] No worries, I have a feeling that either Pi-Hole suddenly does not like my PC for some reason or that ProtonVPNs client software has left some dodgy config on my system to do with leak protection or kill switch. [19:55] The client wont load now that I cannot properly access the internet so that option is out the window [19:55] Fun times [19:55] :-) [19:55] https://phoenixnap.com/kb/ubuntu-dns-nameservers [19:57] so, I have a NEW install of 20 when I try and apt install tor I get the errors shown at https://paste.ubuntu.com/p/bbbjdqPJ5w/ - how do I fix this. All my other Ubuntu 20s install it with NO issues [19:57] apt install has always worked so well [19:58] that phoenixnap link looks totally bogus to me, on ubuntu one does not configure stuff in netplan if NetworkManager is in use === beaver is now known as pong [20:06] looks like that libevent thing is not installable.. try to install it and see what's the error [20:07] maybe you have a newer version and that program wants an older one [20:09] If all else fails and you have more than one kernel, you could always try the alternate kernel. This isn't always a fix but will sometimes work. [20:09] jpmh: It's useful to show the whole version number, ie. 20.04 or 20.10. [20:10] jpmh: I would suspect that you have some incompatible repos/PPAs configured. [20:13] ubuntu. [20:14] jpmh, did you fully upgrade before installing tor? [20:14] also, ubuntu recommends the tor package from their site [20:14] !tor [20:14] Tor is a program to route connections through several servers for anonymity. It is in Ubuntu's repositories, but the Tor Project recommends using their Tor packages due to past issues with Ubuntu's. For setup info, see option (2) of https://www.torproject.org/docs/debian.html.en [20:16] TOR is rort [20:16] oerheks: if you mean did I do apt update; apt upgrade; the answer is YES! [20:16] scortal: please stop that [20:16] 2.0.10 libevent is such old package, are you sure you run 20.04? [20:16] https://launchpad.net/ubuntu/+source/libevent [20:16] 20s seems like you mean 20.04, right? [20:16] Are the official tor packages a smooth upgrade from the Canonical packages, or do you have to uninstall the Canonical stuff before you install from tor sources? [20:16] scortal: if you have a Ubuntu support question please ask and wait for someone to respond :) [20:16] jhutchins: OK -so if incompatible, why on this ONE machine and not others and how do I fix it [20:18] jpmh: You have different packages on the different machines, possibly from different sources. [20:19] jhutchins: they are ALL on KVM instances installed from the same ISO [20:20] libevent-2.0-5 is from the 16.04 era [20:20] so, your bug suggests you are on 16.04 [20:20] oerheks: YES - so why does the official repo reference it [20:20] 16.04 is dead, eol [20:21] oerheks: I am not, at least according to what I installed from, what lsb_release, uname -a etc all tell me [20:21] unless you use ESMn oerheks [20:21] *ESM [20:21] could be esm, but we do not support that. [20:21] true [20:22] and he claims 20s .. no, he is joking [20:22] oerheks: https://paste.ubuntu.com/p/xRkRmx2ZY3/ [20:23] oerheks: and uname provied: Linux another9 5.4.0-100-generic #113-Ubuntu SMP Thu Feb 3 18:43:29 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux [20:30] Such a weird error, file a bug against that tor package? also, use the package from their site [20:31] It might be interesting to see sources.list and sources.list.d/* [20:34] yes, when I look at tor in focal/universe it depends on libevent-2,1-7 so I don't know where jpmh is getting tor [20:35] jpmh, besides the sources.list, show us "apt-cache policy tor" [20:35] jhutchins: there are no entries in the .d directory [20:37] https://paste.ubuntu.com/p/942WBZkFcs/ - the cache output === JoeBk_ is now known as JoeBk [20:38] rfm: what do you mean? [20:38] debian/ubuntu mix? [20:39] oerheks: not consciously - but I MAY have done something, although I have no idea what [20:41] jpmh, you have specified the torproject repo to get tor from, which is fine, but you're pointing to the stretch release which is Debian. you want focal not stretch. (I assume tor has a focal repo) [20:42] rfm - where did I do that - and how do I undo that, if I did it? [20:43] Tor on ubuntu is the universe package [20:43] !info tor focal [20:43] tor (0.4.2.7-1, focal): anonymizing overlay network for TCP. In component universe, is optional. Built by tor. Size 1,377 kB / 4,958 kB [20:44] !info tor bionic [20:44] 'bionic' is not a valid release [20:44] sure it is, ubottu [20:45] LOL - at you arguing with the bot [20:46] jpmh, edit /etc/sources.list. You can either remove the torproject line and use the ubuntu universe version, or change stretch to focal to use the torproject version [20:48] rfm: TY SO MUCH - that did it - really appreciated - this is why I LOVE Ubuntu === EricFan1 is now known as EricFan [21:05] is there a way to check which UBUNTU version offers up to WHICH version of a package? [21:05] say python3 or postgresql-client [21:06] !info postgresql [21:06] postgresql (13+226, impish): object-relational SQL database (supported version). In component main, is optional. Built by postgresql-common. Size 5 kB / 69 kB [21:07] olspookishmagus: What version are you looking for? [21:08] that's the server version right? [21:09] I have a postgresql server v12 running on a 18.04.06 LTS server and it's incompatible with the postgresql-client (v10) [21:10] it produces an error with the \d command, namely: ERROR: column c.relhasoids does not exist [21:11] olspookishmagus: That can be a real mess, I've had to deal with old MySQL databases that weren't upgraded as the versions changed. [21:14] i usually search launchpad with +source [21:16] https://launchpad.net/ubuntu/+source/postgresql-10 [21:16] so, you would have added a source to get 12? [21:17] apt-cache policy postgresql [21:19] it's really a mess [21:20] I'm currently making a dot diagram as to the situation [21:20] it involves python, virtualenvs, pip modules, postgresql, postgresql-client, matlab [21:20] chaos [21:25] why 18.04? [21:26] oerheks: Probably the same reason as RHEL5 === JoeBk_ is now known as JoeBk [21:40] <_KaszpiR_> anyone knows if netplan allows to set static/predictable mac address in netplan (long story short, I have some hosts with multiple interfaces and I want to make a predictable netplan config for each host) [21:43] _KaszpiR_, there shoudl be examples for such stuff on the netplan.io page [21:49] <_KaszpiR_> *should* fits perfectly in theory, but in practice... well I guess Im to tired to check it now, will try in the morning :) [22:04] ubuntu [22:11] I am using Ubuntu 20.04 LTS and Wayland. Today when using Vivaldi, my desktop froze completely. I was able to ssh to my machine from another, and glean a little bit. vivaldi and gnome-shell were using lots of CPU, and gnome-shell was using lots of RAM (10 GB and growing). [22:12] I forced everything to restart, and my desktop went back to normal. I tried repeating the things I was doing in Vivaldi, and my desktop hung in the exact same way. [22:13] Should I collect more information and/or report this elsewhere..? [22:13] brb [22:17] I am using Ubuntu 20.04 LTS and Wayland. Today when using Vivaldi, my desktop froze completely. I was able to ssh to my machine from another, and glean a little bit. vivaldi and gnome-shell were using lots of CPU, and gnome-shell was using lots of RAM (10 GB and growing). I forced everything to restart, and my desktop went back to normal. I tried repeating the things I was doing in Vivaldi, and my desktop hung in the exact same [22:17] way. Should I collect more information and/or report this elsewhere..? [22:24] plujon: I would start looking for trouble reports on Vivaldi. Ubuntu/Gnome is too general to get good results. [22:25] (Oh, it's a browser.) [22:25] plujon: It might be useful to try the same sequence with a different browser. [22:52] gnigh === white_rose[m] is now known as whiterose[m][m] === whiterose[m][m] is now known as white_rose[m] === Jubes91 is now known as Jubes9 === CodeMouse92 is now known as Guest5228 === Guest5228 is now known as CodeMouse92