[00:16] whoa. well, it seems the ssd is busted, dmesg heckling out constant ata errors and it won't reboot after initial install. weird how I got everything up and running yesterday, then the lvm-apt package by accident, then the whole thing went to hell :D [00:16] :( [00:17] sarnold: it was a cheapo WD Blue 250GB ssd, around 50 EUR or so [00:18] still :) [00:18] either my mobo's dying from that workstation or the lvm yesterday put a proper zap into it [00:19] that other guy who was here earlier on was agreeing with my notion that LVM can be lethal for SSD's since it has so many different ways it can "overwrite" it; i.e. go past its boundaries and whatnot [00:20] and I said about this earlier but, nonetheless, a short recap; after that lvm apt package froze up the entire machine (I googled that one up too, seems like a common problem), no matter how I tried to reformat the ssd, it still kept on bouncing me into grub rescue [00:21] ... which means that it probably overwrote some stuff into places it shouldn't have [00:22] ,..aand that was probably a bit too much for the ssd to handle :ʒ [00:23] it's a bit of a bummer in a sense that I got everything working pitch-perfect within a couple of hours [00:24] did you do the encryptedinstall? iirc that does a write to the whole drive; it might have found it didn't have enough rewrite blocks [00:28] sarnold: nope, did not do that. [00:29] sarnold: it was working very nicely before that lvm package came by and then everything went to hell, I mean I was able to reboot time after time with absolutely no problems [00:30] sarnold: at least the zfs gives me a nicer version of grub when it fails to boot :--) [00:31] spelter: hah :) I'm glad you can see an upside, hehe [00:31] sarnold: the weird thing is that even when it fails to reboot, when I disconnect the drie and open it up on another linux machine, all the system files etc are there [00:32] spelter: o_O very strange [00:32] sarnold: I mean that motherboard isn't exactly new, and by that I mean it's clocking at around 11 years, built to last, top of the line i7 core with a few components changed over the years [00:33] sarnold: I'm going to sacrifice another ssd to the great lord moloc---I mean ubuntu [00:33] I have a leftover Kingston 128GB SSD that should also be in pristine shape [00:33] but capacitors are gonna capacitate [00:34] I wonder if doing an entire hour's computing was too much for it? [00:35] sarnold: ran thru the dmesg messages and it did give out a bunch of ata errors but i'm not quite sure if they're the ssd or the mobo, yet. I mean the whole thing runs beautifully from a usb stick [00:35] the livecd (or in this case, livestick) runs perfectly fine, and fast [00:35] alright... maybe the sata controller? [00:36] sarnold: well the thing is that I had a fried SSD on that workstation around halloween (who needs horrors when you've got real stuff to work on, right?) [00:37] that was a Windows drive and I've been now scrambling to get the data rescued, which I miraculously have done, by the power of R-Studio [00:37] (available for linux also) [00:37] spelter: i think i fear for the safety of your kingston 128 [00:38] sarnold: well the harsh fact is that the lm1 apt package screwed up the entire thing ,and there's been many forum posts cursing LVM to hell when it comes to using it with SSD's [00:38] since it literally can overwrite blocks that it is not supposed to etc [00:39] sarnold: I wouldn't fear too much for the safety, I mean the WD blue still comes up pitch-perfect on this linux endpoint [00:39] I do run the whole shabang in BIOS mode for legacy support [00:40] but yep, seriously put , the lvm1 package started to install via apt and then the whole OS froze up [00:40] had to really piledrive it down, I think I didn't use a hard boot tho, just a sudo shutdown [00:41] and even that was like a reaaaaaally long process, and after that, the ssd was kinda like, as it is now, "working but not-working" [00:42] a.k.a. I can fill it up with zeros with dd on another endpoint and run the live install .iso that's been stuck on a stick, it will install just fine ... up until the first reboot [00:42] before that whole lvm1 ordeal, i was able to just reboot at will with absolutely no problems at all [00:43] and yeah, to clarify: when that lvm1 package started to install, and the machine froze up, it didn't boot except into the grub/grub rescue mode after that [00:43] I've been googling up that thing and it seems to be a common thing [00:44] you know, SSD's are a thing in of themselves and add to that all the quirks that each manufacturer has in their models [00:46] spelter: hmmm, I wonder, zeros compress pretty well.. maybe try badblocks with -w or similar to try to poke it in the eye differently.. [00:47] sarnold: I'm just about ready dumping the crap off the kingston [00:48] it's a smaller ssd tho, only 128GB [00:49] iirc I've got two 128s in my big server in the basement; it's plenty for root, homedirs, etc [00:50] I wonder what would be the safest bet in order to have it "future-proofed" so that if that thing works out, I can use it as a basis to hop on to a larger one [00:50] sarnold: hehe, I prefer to run every time-critical thing off of ssd's [00:51] 250gb is plenty for a bunch of devkits and all that jazz [00:52] although my ssd system drives on windoze side tend to be 2tb and 4tb ssd's :--D [00:52] hey, you could spend 1k eur/usd into worse things [00:53] That's a lot but then again both the devkits and software side altogether is so bloated that "IT'S NOT MY PROBLEM! Every night at 8PM" :D [00:54] oh believe me .. thatsystem in the basemtn was *not* cheap [00:54] sarnold: computin' ain't cheap yo [00:54] sarnold: if you can afford a truck you can afford a --- linux box [00:54] :D [00:56] also I gotta admit that SSD's ain't what they're used to be; I mean I've heard so many horror stories recently of people's brand new SSD's just taking that one final dump after mere days of use [00:57] that's also my rationale for going for the 128GB kingston; at least then -- something something, insert rationale here :D [00:57] no but that's an older drive that apart from it being in the shelves for a year or two, hasn't been used very much [00:59] the quality has been dwindling, same goes for previously good sd-card and usb stick brands [00:59] ah see I never trusted the sd cards and usb sticks [00:59] but ssds, those should be okay [01:00] I don't know if modern qlc ssds are okay or not [01:00] sarnold: in my case I don't always have a choice, it's a camera thing [01:00] yeah [01:01] but yep, even sandisk extrems can go *poof* just like that. it sucks. especially when you're really paying to get something decent. [01:02] sarnold: oh trust me, I've been riding on the SSD insanity for almost a decade now and I've had such revelations with the quality and the way different manufacturers cut corners etc [01:03] and I mean; drives that were over 500 EUR at the time and "whoops, we didn't see that one coming in our memory cell architecture" [01:03] :D [01:04] Sorry to interject a little here, but is this really still related to Ubuntu for the last few pages of scrollback? [01:05] seriously, it was like that. intel used to keep up a pretty solid track record until they noticed that they can't compete on the consumer markets because they didn't want to tarnish their brand, and hence they switched to enterprise-intended ssd manufacturing only , could've stopped that one too, haven't checked up on them in a while. [01:06] krytarik: oh damn, I thought I was on the open discussion channel, not at the helpdesk. my bad [01:06] krytarik: sorry about that. [01:06] Heh, no worries. [01:06] that's okay, spelter. i've done the same thing. #ubuntu-offtopic is more your jam. [01:07] that aside, i have a question. whenever I open the official Discord app (a snap, if that matters), I'm immediately logged out. what would cause that? [01:07] krytarik: it was related to my woes with my recent attempt to install ubuntu 20.04.1LTS on an almost-pristine SSD, which worked for a while and then it went to hell, and now I've been scratching my head over it [01:08] signofzeta: what ubuntu version? [01:08] signofzeta: also, snaps are a hit-or-miss kinda thing imho. "oh snap!" *rimshot* [01:08] spelter: 20.04.1 LTS, fully updated, kernel 5.4.0-52 [01:09] I mean I've literally given up on most snaps and just compile with my own scripts from the latest git source or so if I need to [01:09] yeah, I know snaps can snap at you, but this is new. just started today. may or may not be related to me installing amdgpu drivers from AMD, not sure. i can't read the debugging output because i'm logged off. [01:09] signofzeta: yeah you should defo see what it's doing by running it from cli [01:10] tried it, spelter. it logged me out so fast i couldn't read the output. [01:11] wait what [01:11] I type "discord" in a terminal (or click the icon), then i'm immediately back at the logon screen. [01:12] logged you out? out of your xwindow session? [01:12] wow. [01:12] oh wow [01:12] yeah, just like that. let me check journalctl for anything interesting. i haven't had time to troubleshoot yet. [01:12] is there anything in journalctl / dmesg / audit logs? [01:12] tell ya what, as soon as I get that kingston ssd tried (OR fried) out, I'll try me some bad mojo-dickord myself [01:12] discord , sorry [01:13] journalctl has nothing for discord. some AppArmor stuff about the Ubuntu Store, but nothing damning. [01:14] signofzeta: you could try starting a script (With i.e. 'script whatsgoingonwithmydiscord.txt') in your terminal before you go-go [01:14] signofzeta: discord is the kinda software that I really don't, for some reason, trust that much [01:15] I mean there's been some datasec news about the owner selling some private data or some stuff like that [01:15] yeah, but my friends are on there. i use it carefully. but I also use Linux. [01:16] signofzeta: :D maybe better to firejail the entire thing and see what it does then [01:16] and do the script thing before you run it so at least you can then get a whiff is it says something inside term [01:16] s/is/if === johuck5 is now known as johuck [01:17] The only thing I see in dmesg about Discord is an AppArmor warning: audit: type=1326 audit(1604451649.764:15197): auid=1000 uid=1000 gid=1000 ses=3 pid=26580 comm="Discord" exe="/snap/discord/115/usr/share/discord/Discord" sig [01:17] signofzeta: "my friends are there" reminded me of that Mark Twain quote: "Go to heaven for climate, hell for the company." [01:18] I wish I could get my friends to use PGP, but that's a whole different channel. :-) [01:18] signofzeta: and if you start a script before you run it...? anything? [01:18] signofzeta: it looks like it was truncated at 'sig' [01:18] Is it? That was the whole line. [01:19] sarnold: that was the whole line, unless it wrapped mid-screen [01:19] Discord does some odd stuff on Windows side too, I've been probing it every now and then [01:19] signofzeta: oh btw you can use discord with just about all the features via www.discordapp.com/login [01:19] yeah, I'm no stranger to Discord oddness. I run it on my macOS laptop and i can watch the battery drain. [01:19] I think that was the url [01:20] signofzeta: how'd you see it? journalctl does some stupid line truncating thing, I wonder if that's what you hit [01:20] I mean if you have an up-to-date web browser, even screen sharing, webcam, stuff like that works [01:20] and hence you don't need to install any of their wonky cr*p on your PC [01:20] sarnold: maybe. I did it in a terminal window. i've used Linux for 15 years, but i'm back to Ubuntu and trying to get the hang of systemd. [01:21] sarnold: and no, that pasted line came out of dmesg | less [01:22] signofzeta: weird. I've never seen an audit message not stick to the key=value pattern [01:23] sarnold: maybe it was truncated before reaching dmesg, i'm not sure. [01:23] possible, I've been logging mine to /var/log/audit/audit.log for so long I haven't got a clue what happens when they are sent to dmesg [01:23] but the apparmor tooling can scrape messages from that, so they shouldn't be truncated there, I don't think [01:23] sarnold: nope, i'm an idiot, hold on [01:24] sarnold: https://discord.com/login <= and if it tries to redirect you to the discord app, there should be a selection "open in browser instead" [01:24] yeah, I saw the Discord browser login. I'll use that, but I'd like to the fix the app eventually. [01:24] sarnold: voice chat, video chat/screen share etc works [01:24] straight outta browser [01:24] signofzeta: yes, I understand that woe [01:25] Here's the full line: udit: type=1326 audit(1604451812.515:22787): auid=1000 uid=1000 gid=1000 ses=3 pid=34999 comm="Discord" exe="/snap/discord/115/usr/share/discord/Discord" sig=0 arch=c000003e syscall=203 compat=0 ip=0x7f262b995b9f code=0x50000 [01:25] I'm distracted by some political theater tonight, my bad. [01:25] political theater? not on my watch! [01:27] i'll try logging Discord output to a file soon, I'm chatting in another channel right now [01:27] $ ausyscall 203 [01:27] sched_setaffinity [01:28] Anyone know how to enable Mobile Networking from command line? I see two options Enable Networking and Enable Mobile Networking. I want to do this from command line. For Enable Networking I can do "nmcli networking off or on" [01:28] i installed the closed-source AMD GPU drivers a few days ago, because I needed OpenCL support. i assume it's related. [01:28] How do you toggle Mobile on off [01:31] Guma: there's this thing called google and a thing called Ubuntu's Official Documentation, you might've heard of them ; there's the instructions at least for older versions ; https://help.ubuntu.com/community/NetworkManager#Using_NetworkManager_on_the_command_line [01:34] I'm installing the 20.04.1LTS desktop now on that beforementioned Kingston 128GB SSD to see how that works out [01:35] I wonder if I should use ZFS or not, hmm [01:35] ZFS it is! Sun Microsystems! Zardoz! [01:35] I mean, sorry. [01:35] spelter: I do. Works great so far. it's not a boot volume, though. [01:36] no errors yet [01:37] signofzeta: I had the SSD issue just before this one, hmm, thjis also seems to be a way lot faster than the WD Blue 250GB cheapo one, although this is like, years old and so [01:37] but hasn't been used much, all the memory cells should be in pretty good shape as for r/w wear [01:43] i just had a PCIe M.2 SSD report that it was starting to fail. poor old thing. i guess its last owner abused it. though my motherboard's M.2 slot is immediately under the GPU slot, so it gets hot. [01:43] signofzeta: that's the thing with solid state drives [01:44] it's wild e. coyote stuff, works like a rocket engine, until you hit a wall [01:44] signofzeta: at least you got a forewarning [01:45] Hard drives would slow down and maybe make noise back in the day, before SMART kicked off. SSD's are more reliable, but tend to die quietly. [01:45] I mean, usually, when SSD's or other solid state media goes, it just goes [01:45] signofzeta: quietly AND suddenly at times [01:47] btw is there any benefits for a PC desktop user from trying out ubuntu 20.10? I saw some initial reviews that weren't very enthusiastic about 20.10 [01:47] does 20.10 use the 5.5 kernel? my computer has an AMD fTPM. [01:47] I did a pervert thing here on this Raspberry Pi4B endpoint and tried out 20.10 and then reverted back to 20.04.1LTS but got to keep the 5.8 kernel [01:47] :D [01:48] Kernel: 5.8.0-1002-raspi aarch64 bits: 64 Desktop: Xfce 4.14.2 Distro: Ubuntu 20.04.1 LTS (Focal Fossa) [01:48] Machine: Type: ARM Device System: Raspberry Pi 4 Model B Rev 1.1 Kernel: 5.8.0-1002-raspi aarch64 bits: 64 Desktop: Xfce 4.14.2 Distro: Ubuntu 20.04.1 LTS (Focal Fossa) [01:48] ;) [01:49] well, if you have an AMD or Intel fTPM, 20.10 will let your apps use it -- if it has Linux 5.5, that is. i forget. [01:49] signofzeta: my mobo is old a.f. but I do have an i7 core with plenty of ram onboard etcetera, and a good gpu [01:50] but I doubt it supports any of the new motherboard gimmicks [01:50] nvidia quadro drivers seem to be up to date [01:50] It's short for Firmware TPM. it's a TPM emulated in the CPU microcode. might just be an AMD Ryzen thing, who knows? [01:54] aaand after everything was installed, I'm now waiting in front of a black screen for something to happen [01:54] just like what happened with the other SSD [01:54] error: compression algorithm inherit not supported [01:54] Is there an easy way (in focal) to find all process IDs of all processes that are running as user "_apt" and whose executable's path starts with "/usr/lib/apt/methods/"? [01:54] error: you need to load the kernel first [01:54] (??? or should I pipe ps to sed) [01:55] hi again [01:55] managed to upgrade it [01:55] yep doesn't work from the other ssd either :| [01:55] error: attempt to rea or write outside of disk 'hd0'. [01:55] Sven_vB: that's probably a one-liner if you install https://github.com/osquery/osquery .. but ps to sed isn't a bad start [01:55] (Read) [01:56] card reader still not working [01:56] spelter: sweet! can you file a bug report on that while yo've got it running? :) [01:56] and now (gnome-disks:12708): GNOME-Disks-ERROR **: 01:52:59.913: Error getting udisks client: Timeout was reached [01:56] disks dont work after upgrade to 20 [01:56] sarnold: While I got it running? :D you mean the GNU GRUB screen that is now once again staring me in the face [01:57] sarnold: I'm typing this from another endpoint [01:57] funny thing is that also, GNU GRUB doesn't let me scroll up or down with arrow keys [01:57] only lets me press enter [01:57] sarnold, sounds like ps|sed is the easiest way with standard equipment then. thanks! [01:58] I seriously wonder what the hell was going on, I got everything to work perfectly yesterday, and then boom [01:58] or was = is [02:01] yeah so anyway it's kinda hard to file in a bug report on that one as it is, I guess I could see if I can shut that down and dump the logs to you [02:01] the install in of itself went without errors tho, I've had errors in that process too during my re-install attempts [02:02] actually I think I had some errors during the very first install attempt but i just did sudo apt updates and upgrades and a couple of reboots later it was all good === zbenjamin is now known as Guest70393 === zbenjamin_ is now known as zbenjamin [02:04] I'm re-running the install from the boot installer, again everything's working just fine for now. I'll try with the default filesystem selection this time. [02:05] I chose "erase disk and install ubuntu" and no advanced/extra options this time... [02:07] I do have some extra USB sticks, wonder should I try to just go for the 20.10 and see what happens [02:08] it might be that the i7 core generation I have is too old or something for Ubuntu to handle it tho, wonder also if I should set stuff up in BIOS somehow differently. but well, at least now I know it wasn't (just) the WD Blue SSD [02:09] afaik all the i7s should work fine [02:09] btw my wd blue ssd that is now out of that pc opens up just fine here on this ubuntu 20.04.1lts (running on aarch64) [02:10] I mean I can get the entire filestructure opened up [02:10] no wait, I can't :D [02:10] there's a 537 MB volume, then "bpool" and "rpool" [02:11] /dev/nvme0n1p2 2048 1050623 1048576 512M EFI System [02:11] that's my guess for your 537 MB volume [02:11] sarnold: if you don't mind 6 rows of text .... [02:11] Device Boot Start End Sectors Size Id Type [02:11] /dev/sda1 * 2048 1050623 1048576 512M b W95 FAT32 [02:11] /dev/sda2 1052670 488396799 487344130 232.4G 5 Extended [02:11] /dev/sda5 1054720 5249023 4194304 2G 82 Linux swap / Solaris [02:11] /dev/sda6 5251072 9445375 4194304 2G a5 FreeBSD [02:11] /dev/sda7 9447424 488396799 478949376 228.4G a5 FreeBSD [02:12] apparently, it's all there, grub just goes haywire and dumps me into the rescue prompt [02:13] and I've tried different fs options from the setup choices on the 20.04.1LTS [02:13] I wonder if the zfs install option expects uefi rather than legacy [02:14] it all worked out just fine, well, apart from few glitches that I did send a bugreport out of === EuphOria is now known as AbandonHope [02:14] okay so now I got an "Installer crashed" errormsg [02:15] still managed to get to the desktop, should've sent an error report, or at least it said it would [02:16] running apt update / upgrade, any good tips what I should do before I attempt to reboot? [02:16] ^ on [02:18] just my take on things but perhaps the installer altogether should have a logging+crash log send functionality straight from the get-go [02:19] especially when it's such a rickety roller coaster ride to install [02:19] spelter, shameless plug: when you have doubts about rebooting, it's always a good idea to have a SuperGrub Disk at hand. #sgrub [02:19] Sven_vB: thanks [02:20] I dunno if it's a licensing thing or what but the whole troubleshooting as for the boot failures and such could be made a way lot easier [02:20] and coherent at least [02:21] yeah there's a ton of stuff that could still be way easier in Ubuntu. at least, over the years it has made lots of progress already. [02:22] Sven_vB: and don't get me wrong, I do realize the fact that I get what I pay for -- which is nothing :D [02:22] no, but all joking aside, it has a huge potential [02:23] oh btw with spare USB thumb drives, did you mean to try different live ISOs? because the SuperGrub Disk can boot them from HDD or SSD, and the latter usually is a lot quicker to boot. [02:23] linux altogether and so forth, it just seems so scattered, add to that the fact that indeed packet managers can go wild and whatnot [02:24] Sven_vB: oh, ok, thanks for the tip... [02:24] now I've ran apt updates and upgrades [02:24] any last words before I reboot and face the grub? [02:24] :D [02:25] if you install Ubuntu a lot, it might be worth considering options for automatic install [02:25] Sven_vB: yep I've done that a multiple ties [02:27] what do you use then? [02:29] How do I re-mount an external LUKS-encrypted drive that accidentally became disconnected? [02:29] spelter: fwiw we get a *lot* of automated bug reports during install https://bugs.launchpad.net/ubuntu/+source/ubiquity/+bugs?orderby=-id&start=0 [02:32] sarnold: where should I send the entire dmesg dump [02:33] i'm a bit worried about thoes ata errors tho [02:34] kvndy, afair you have to dismount its logical volumes, deavtivate them, close the LUKS, ensure its mappers are released for real, then open the LUKS as usual, activate the logical volumes, then mount their file systems. [02:34] kvndy, ideally you have a helper script for that ;) [02:35] with mappers I meant dm devices that show up in /dev/mapper. [02:36] spelter, some ATA errors can be normal part of drive type detection. if they occurr while the disk has been properly detected already, check the cables. if you have doubts about disk health, smartmontools can help. [02:37] Sven_v8 So... restart the computer every time then? [02:38] Sven_vB: https://pastebin.com/LWBGdEWD stuff like that ... just posting a part of the dmesg there but [02:39] then again it makes absolutely no sense at all that I'm able to install the entire thing on to an ssd and play around with it until I hit a brick wall [02:39] I mean it runs right now as I type this, but I'm pretty sure if I reboot it now, it'll be toast [02:40] if there's a dev around who wants to take a peek at the entire dmesg dump from the boot, I'll be glad to send it over [02:41] Emask 0x52 (ATA bus error) seems to be prevalent [02:41] [ 117.081313] sd 0:0:0:0: [sda] tag#8 Add. Sense: Unaligned write command [02:42] I wonder ... I used to hear reports that some install software would misalign the starting sector, causing all writes on the drives to be insanely painful [02:42] https://bbs.archlinux.org/viewtopic.php?id=174335 <= "Problem solved. The solution was to switch which SATA ports my drives were connected to. Thanks for the help! I also updated the title to better describe the actual problem. [02:42] I thought that hadn't been an issue in half-dozen years or something, but maybe you've got a fun way to trip it? [02:42] In case anyone else has this problem, I'm using an ASRock Z87 Extreme4 motherboard, and the solution was to move my SATA cables from the SATA3_A0_A1 ports to the SATA3_0_1 ports." [02:42] that's from archlinux but [02:42] sarnold: well like I said, I'm running an old-gen i7 core [02:43] and by old, I mean they stopped manufacturing this intel motherboard in 2009 [02:43] wow I think this guy's errors are much scarier, he's got *all* the errors, heh === PowerTower_121 is now known as PowerTower_120 [02:44] kvndy, if reboot is a cheap option and it's easier than re-mounting, that's a valid approach of course. [02:44] sarnold: well, I didn't post all the dmesg errors in that one [02:44] I got a bunch of that same-kind-ish ata error and so forth [02:45] kvndy, in case it is the disk where your lvm program resides, it's usually also the only option. [02:45] Sven_v8 annoying but eventually I want to figure it out. It's going to take some research [02:45] "Disabling lock debugging due to kernel taint" :D [02:46] "Module license 'MIT' taints kernel" [02:46] :D [02:46] ok "I/O error, dev sda, sector 234441248" is one of those ATA errors that I expect to not see. [02:46] Sven_v8 Ubuntu is installed on my internal SSD, but storage on an external LUKS-encrypted drive with a flaky connection [02:46] let's hope it's just your cables having problems [02:47] kvndy, I see. then a script for cleaning up the debris seems to be a good idea. [02:48] and thus I reboot aaand. [02:48] back to grub. "error: attempt to read or write outside of disk 'hd0'. Entering rescue mode..." [02:49] I still don't get it how did it work without any issues initially, I rebooted it like ten times [02:50] when have you last run the long S.M.A.R.T. test? [02:50] and I mean, by all logic, the entire OS is already installed on the ssd [02:50] these are both near-mint ssd's, the other's a bit older [02:51] but hasn't been used much at all, same thing goes for the 250GB WD Blue [02:51] did you verify their capacity with F3? maybe they have a surprise for you. [02:51] my point here is that I'm now trying this same thing out on another SSD and I've got two that have simply just went zonk [02:52] as in? that they're fake? [02:52] maybe. or just broken. [02:52] oh c'mon [02:53] two in a row and the other one, the 128gb ssd that I just tried out just now, I emptied it from around 110GB of files [02:53] Sven_v8 I'll have to wait until the next accident before I'm motivated to figure it out [02:53] I always test my new SSDs and thumb drives. bugs can be very hard to find if the lowest basics are at fault. [02:53] so, I mean ... and these are both from official retailers that don't sell copies so ... [02:53] these are not new ones [02:53] been trying to explain that [02:54] I mean I can rip out a brand new, still sealed and whatnot ssd and put that in if that satisfies the doubt [02:54] well, afair F3 was invented because a well-known retailer had a bad batch. [02:55] and well, both of these hard drives have been used as material transfer mediums between endpoints with very little overall use, SMART health excellent and so forth, and been filled up to capacity , so [02:55] how many ssd's do you want me to pull out and stick in there before you would conclude that it's not the drives [02:55] if you want to believe, I accept that. I just wanted to inform that you can know instead of believing. :) [02:56] two different drives from two different manufacturers from two different years [02:57] so if I stick 10 brand new different SSD*s in there, you still think that the statistical odds without checking are on the side of the SSD's and not because there's something going haywire in i.e. the code or elsewhere [02:57] makes perfect sense, except that it doesn't [02:58] what I still don't get is how it could work yesterday and be screwed up like that today [02:59] I did check the "install updates while installing" and "allow turdparty proprietary drivers" etc yadayada [02:59] maybe let's get back to that error message. so you assume you have an Ubuntu on that SSD, but the GRUB looks like it's broken? then it would be interesting to see what a known-good GRUB (e.g. SuperGrub Disk) sees. [03:00] yeah, it might not be the SSD's. i bought big hard drives ages ago, so I use those for my storage. this one SSD i found acts as a totally-unnecessary L2ARC. [03:01] those two ... were ... perfectly working ... literally almost good-as-new ... ssd's. [03:01] and I can still open up the contents of both on this other endpoint [03:02] from some of the error messages I got the impression that it might be a problem related to the order of the sata ports [03:02] and I MEan for christ sakes, I can run the entire OS just fine after installing it but with the first reboot it does that now [03:02] how would that affect GRUB? do you have them in a hardware RAID? [03:02] it's gotta be some update that has been introduced to ubuntu within the past 24hs [03:03] that's my take on it [03:03] since I haven't changed the cabling, it goes straigh into that one docking slot, and since it WORKS [03:03] kinda [03:04] except for the boot-up ending up in grub's rescue mode [03:05] maybe. can you pop these into a (ugh) Windows system and set up a Storage Space or something, just as a test? skip this if your data is already on these. [03:05] hi again [03:06] a Storage Space? as in, clone the entire drive as an image for people to probe? [03:06] spelter, perfect use case for SGD [03:06] im trying to get access to my sd card to put some things on it and i cant format it, it keeps giving me this message [03:06] rather not d o that [03:06] error creating file system command-line mkexfatfs-n exited ith non-zero exit status 1 [03:07] flushing stderr error fsynk failed imput/output error udisks error qark 0 [03:07] nothing on google [03:08] spelter, actually, if a broken GRUB is your only obstacle atm, I'd just install Ubuntu without a GRUB, and install GRUB later from the inside. [03:10] Sven_vB: and how do you suppose I do that with the desktop installer? there's no option for that? or did you suggest that I'd just grab Ubuntu Server and install the desktop separately? [03:11] spelter, I had just assumed that the desktop installer has that option. if it doesn't??? it won't make much of a difference. you'd backup your good grub, let the desktop installer destroy it, boot from SGD, restore the good grub. :( [03:11] I'm still contemplating on trying out 20.10 because I have this stupid wish that perhaps it's a kernel level bug that has been fixed in the 5.8.x kernel tree that 20.10 uses [03:11] should have been :) [03:11] but then again, mo' updates, mo' problemz [03:11] forget the Windows Storage Space if your disks are in use. it's the closest thing Windows has to ZFS. it might have been a good control group for a test. [03:12] spelter, now I'm confused. what's the current problem? wasn't it GRUB reading outside the disk? [03:12] signofzeta: how about the fact that I can fill up these SSD's to their final memory cell without an issue? [03:12] spelter: sorry, I haven't been paying close attention. never mind then. [03:12] Sven_vB: I install via the official .iso installer, everything works ... until I reboot and end up in grub's rescue mode. [03:13] if choose zfs as the filesystem in install options, I end up in a little bit fancier looking grub (btw the arrow keys don't work in the GNU Grub) [03:13] :D [03:14] spelter, ok. so it seems to be a GRUB problem, no kernel involved at that stage. SGD is made exactly for that situation. why not try it? [03:14] go fsck -f dev/sda and it ill repair the booting part [03:15] Sven_vB: I might as well do that by now, it seems just so ironic (or MORONIC, for that matter) that you get distros "broken out of the box" like that, exactly what i.e. Luke Smith, if you know the Unaboomer, has been ranting about in his youtube vids [03:15] more on the archlinux side tho [03:15] a house is only as solid as its foundation [03:16] there should be a proper recovery mode implemented straight on to the desktop version imho, everything is streamlined to look so good and then it hits a brick wall. [03:16] *just a kind side note greeting to the devs ;) I hope wimpy's not watching this! * :D [03:17] spelter, considering the complexity of modern computing, it's a marvel that the installer works a lot of the times. a certain rate of failure is to be expected. [03:18] also of course I'd like to keep a market for commercial installers. ;) [03:24] Sven_vB: I mean since you've obviously done such a good job in doing all that stuff, it makes me wonder why can't your code be implemented as a safety measure into existing distros [03:25] Sven_vB: should I try out Rescatux or what ? [03:25] spelter, I'm not sure I understand the question. I'm not a Ubuntu dev, and my patches for the official installer have been ignored so far. [03:25] Sven_vB: that figures :D [03:25] spelter, nope, I'd try the SuperGrub Disk. it's on the same website though. [03:28] for my commercial projects I usually use installers based on multistrap nowadays. once you learn the depths of the install process in detail, it loses its magic. :) [03:35] Hey im having issues finding where my games are being saved at on wine. I went to /home/.wine/drive_c/program files(x86)/ [03:35] dont see it unfortunately. But when im installing the game again it shows its right there. Completely confused. Using ubuntu20.04 === Mrokii_ is now known as Mrokii [03:41] random1: "du ~/.wine/drive_c/ | sort -n -r | head -n 20" will give you an idea of where the larger files are, which will probably tell you were your games' data gets stored. [03:42] appreciate it jordan_u [03:44] still dont see it. [03:44] It did the install o.O. [03:45] Going to try installing it again i guess o.O [03:46] macOS has a way to type accented characters by hitting Option/Alt + a key, then the letter. Is there anything like that in Ubuntu? [03:47] signofzeta: there's several, I think; 'deadkeys', 'compose key', 'altgr', hopefully these help you find something nice [03:50] thanks, sarnold. I'll check them out. macOS made it easy, and iOS made it dead simple. sure beats going Alt+numbers on Windows, that's for sure. it's not often i need to type them, but it's annoying when I can't. [04:08] Hey can anyone tell me how to search for a specific file in my hard drive? [04:08] Textures0.bsa [04:09] random1: https://www.howtoforge.com/tutorial/linux-search-files-from-the-terminal/ [04:09] random1: if it's been there a while, 'locate Textures0.bsa'; if it's new, sudo find / -name Textures0.bsa [04:10] sarnold: you can update the locate db [04:10] leftyfb: yeah but that's a pain :) [04:10] sarnold: nah, I run it all the time [04:10] sudo updateb ; locate [04:10] leftyfb: is it still a cron job or is it a systemd time now? [04:11] ahhhh, so you don't care about the prunepaths etc that the configuration uses [04:11] sarnold: looks like it's also in cron [04:12] sarnold: nope, half the things I search for using locate I want searching everywhere :) [04:12] leftyfb: have you seen this yet? https://blog.sesse.net/blog/tech/2020-10-12-20-27_plocate_1_0_0_released.html [04:12] leftyfb: it's several hundred times faster than 'locate' on my system with a lot of files [04:13] leftyfb: time for me to bail, have fun with it :) [04:13] win 164 [04:13] sarnold: it's not available in ubuntu yet :P [04:13] yeah :( === lavender- is now known as lavendereyes === ravustaj1 is now known as ravustaja === ravustaj1 is now known as ravustaja === ravustaj1 is now known as ravustaja [06:27] Hello! o/ [06:27] I have the known probem/bug with copy-paste (clipboard), in Ubuntu 20.04 LTS [06:27] what can i do to fix it? [06:27] welcome Tas-sos [06:27] Hello lotuspsychje :-) [06:28] I read somewhere about an alternative package that install (but now I can not find it now).. [06:34] "known probem/bug with copy-paste (clipboard)"? whom is this known to? is it known to the bug tracker? [06:35] Tas-sos: ^ [06:44] I mention it as known, because I believe that I am not alone... it is a known problem [06:44] https://bugs.launchpad.net/ubuntu/+source/virt-manager/+bug/1872527 [06:45] Launchpad bug 1872527 in spice-vdagent (Ubuntu Focal) "Clipboard doesn't work 100% of the time in Ubuntu 20.04 (in KVM guests)" [Undecided,Incomplete] [06:45] https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1852183 [06:45] Launchpad bug 1852183 in mutter (Ubuntu Groovy) "[X11] copy/paste (clipboard) with LibreOffice is broken in Ubuntu 19.10 & 20.04 (see comment 93 and 176)" [High,Fix released] [06:45] https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1879968 [06:45] Launchpad bug 1879968 in mutter (Ubuntu Focal) "Copy/paste still sometimes fails in LibreOffice and Wine apps" [Medium,Fix released] [06:49] Anyway, my problem is that sometimes (not always) I can not copy-paste from one application to another. [06:50] For example, from gedit to Firefox/Brave/Chromium [06:50] from a browser to LibreOffice etc.. [06:50] Tas-sos: hmm, the first bug report is said to be limited to running in a KVM VM (is this what you do?), the other two are solved (is your system fully up to date?). thanks for looking those up, though. [06:51] this problem is really very annoying when it happens because you can not work [06:51] i can imagine that's annoying [06:52] hi. in dpkg-query, what's the difference between Status-Status and Status-Want? my guess is that "Status-Status: installed" means that the package is installed right now, while "Status-Want: installed" means it's either installed now or will be installed after some sort of job runs, but I don't really know how that works exactly. [06:52] Tas-sos: please see the questions in parentheses above [06:52] !uptodate | Tas-sos [06:52] Tas-sos: To ensure you have all the latest known patches and security updates for your ubuntu installation, please update with the following command: `sudo apt update && sudo apt upgrade`. See also !upgrades and !security; you may also need to run `apt full-upgrade`. [06:55] yes of course you are right. I saw that the first one I sent referred to KVM which is not really in my case [06:56] I just have a HP laptop with Ubuntu 20.04 LTS [06:56] Tas-sos: is it fully updated and rebooted afterwards? [06:56] yes i updated my entire system yesterday.. with : apt update and apt dist-upgrade [06:57] okay, great. this is standard ubuntu with gnome-shell, not another flavor? [06:57] I also need to run "apt full-upgrade" ? [06:58] dist-upgrade and full-upgrade do the same thing [06:58] so, default gnome-shell / mutter? [06:59] And are you using Xorg, or XWayland? [06:59] No no another flavor, I have ubuntu! Nothing else [06:59] hm... about default gnome-shell... [06:59] What you mean mutter ? [07:00] you already answered this question basically [07:00] yes, most likely the problem will not exist again [07:00] gnome-shell and mutter are what forms the default gnome-based desktop on ubuntu 20.04 LTS [07:00] tomreyn: A! Ok! Thanks! [07:01] by default, 20.04 uses the XOrg X server, but you can also switch to XWayland. Did you switch to XWayland? [07:01] nc termbin.com 9999 < <(lsb_release -ds;cat /proc/{version,cmdline};echo "Session: $XDG_CURRENT_DESKTOP ($XDG_SESSION_TYPE)";echo Shell: $SHELL) [07:01] you can run this and get all those infos in a row. [07:07] cheater: with dpkg under-the-hood questions, you may be better served on #debian here or on oftc, or rather on the ubuntu or debian mailing lists [07:07] thanks [07:07] i asked in #debian but it's idle [07:07] i'll keep waiting thanks [07:08] i assume you read the man page and it didn't really help? [07:09] cheater: ^ those very short explanations there refer to the dpkg version which introduced these states, so you could refer to the changelog, which might have more info. [07:10] i mean the debian changelog [07:10] hmm good idea but that might be a little too deep for me [07:11] https://launchpad.net/debian/+source/dpkg/1.17.11 [07:12] " * Add new dpkg-query virtual fields db:Status-Want, db:Status-Status and db:Status-Eflag to allow fine-grained access to the Status values." [07:12] not super helpful either. i guess it says "inspect my source code". [07:24] tomreyn: Thanks for your time! I hope and believe that I will not face the same problem again! [07:24] Thanks! :-) [07:30] Only Youtube is slow. The Youtube homepage wont even load. All other websites are loading fine. Is there a way to troubleshoot this ? [07:31] john_rambo: disable browser extensions / plugins, try again [07:32] tomreyn, I just loaded FF with all addons disabled & Youtube loaded just fine. How to find out which addon is responsible ? [07:34] john_rambo: if "FF" is an abbreviation for "Firefox" which saved you 5 characters, but made it less obvious, then you can try re-emnabling extensions one by one and identify the one causing it this way. [07:35] alternatively, there's about:performance [07:35] tomreyn, Okay/Thanks === jeddo- is now known as jeddo === WalterWhitesCook is now known as WalterWhitesCock [08:39] hello #ubuntu. Does this channel also covers ubuntu server ? [08:39] basically I want to know how to show the motd again when I'm already logged in. Is there a command for this ? [08:40] ychaouche: it does, but you might usually have more luck in #ubuntu-server where the experts gather [08:40] thanks lotuspsychje, /topic should be updated === beaver is now known as headnod [09:51] Does anyone know how to combine multiple .vmdk files on Ubuntu? [09:52] I've found a lot of reference to the vmware tool but I can't find anything about it on vmwares site [09:52] RevertToCamelot: what do you mean combine? each of them would be a separate VM [09:53] They download as fragmented files within the zip sometimes when you need to download an entire vm [09:53] i've done it before, just never on linux === themill_ is now known as themill [11:58] Hi all [12:56] Hello! Still working on the grub issue that I talked about yesterday over here [12:57] I noticed that a lot of people have had the same problem since at least 18.04 days. After installation, the whole thing just boots into grub rescue [12:57] I mean plenty of material on that online, so it's definitely not a new problem [13:09] speltner: if you are looking for continued support, please ask your question again into the channel with all details so volunteers can pickup where you left [13:09] hi everyone. :-) welcome back, speltner. what operating systems was the SuperGrub Disk able to detect? [13:10] Hey all. I’m trying to add a CA cert to my Ubuntu 18.04 machine, as i understand all i need to to is put my .crt under /usr/share/ca-certificates and then run sudo update-ca-certificates ? However my cert doesn’t get picked up and added. Any ideas why? [13:11] If i run sudo dpkg-reconfigure ca-certificates my cert i listed and i can manually add it, however i don't want to do it interactivly [13:12] Notum, as for the non-interactive part, config automation tools like ansible could help. [13:13] Sven_vB Ansible is actually what i'm trying to use, this "add and import cert" thing is part of a larger playbook [13:14] Notum, I'm sure someone else had the problem before and has written an ansible module for it. :) [13:15] speltner, if you prefer to not try SGD, you could also boot into the live session and install GRUB manually. [13:16] Sven_vB Probably, but my problem isn't really related to Ansible. I just don't get why my cert isn't picked up with update-ca-certificates :/ [13:16] https://askubuntu.com/questions/645818/how-to-install-certificates-for-command-line I'm doing exactly as explained here [13:17] how do you test whether it is picked up? === yn_ is now known as yn [13:19] Notum, custom certs should go to /etc/ca-certificates/update.d/ ... (and then update-ca-certificates needs to be run) [13:20] Sven_vB When running sudo sudo update-ca-certificates i get '0 added, 0 removed' [13:23] ogra I dont think so? the ca-cert config utility (sudo dpkg-reconfigure ca-certificates) even says "This package installs common CA (Certificate Authority) certificates in /usr/share/ca-certificates." [13:24] Which is exactly where my cert is placed.. :/ [13:25] Notum, I think ogra's idea is worth a try. if that doesn't work, verify whether manually adding the cert works, to ensure it's not a problem with the test setup. [13:26] ah, no, i'm wrong updates.d is for hooks to run along with custom certs, sorry [13:26] but the update-ca-certificates manpage states: [13:26] Furthermore all certificates with a .crt extension found below /usr/local/share/ca-certificates are [13:26] also included as implicitly trusted. [13:26] try that dir [13:26] Sven_vB Manually adding the cert with dpkg-reconfigure ca-certificates works fine [13:26] at the bottom it also says: [13:26] /usr/local/share/ca-certificates [13:26] Directory of local CA certificates (with .crt extension). [13:27] ogra Ok, will give it a try :) [13:27] Notum, in that case you could dump the debconf database before and after, compare, and manually apply the change with debconf [13:29] or have a look at the source of update-ca-certs, maybe it turns out to be a very simple script and you see the trick part within minutes. [13:34] Adding the cert to /usr/local/share/ca-certificates works a liiitle better. I get 1 added and 1 removed so it seems to add and then remove it again... [13:42] So... regarding encryption with dual boot: https://askubuntu.com/a/442820/64513 Makes sense I can't encrypt full disk. In a nutshell I want ubuntu install to be 'well' protected while not caring bout windows/keeping win working (or the respective equivalent to that). Given I'd have a root, swap and home partition alongside n windows partitions. At the very minimum I want (and likely can) encrypt home [13:42] partition. Can I also encrypt root/system partition and swap (should I/ 'need' I to protect swap as well?) [13:47] nuala, I happen to have installed Ubuntu focal into an encrypted LUKS side-by-side a Windows yesterday. it works. :) [13:48] nuala, if you really need a swap at all, I think you should encrypt it. would be bad if your secrets would leak from that. [13:49] nuala, and yes you can have all the linux partitions on the LUKS, you just need to either adapt your GRUB config in arcane ways (may require using an old LUKS) or have the kernel and initrd unencrypted, but the latter could be on a different physical medium, maybe a USB thumb drive. [13:50] basically you have to ensure there's some way for the boot process to load the decryption program from some unencrypted memory. [13:57] Oh that's nice to hear! :D Hm. I dont think an unencrypted boot process would be an issue (I hope ><) like, threat level is generally low. [13:58] it should be safe against theft at least. just don't use a potentially-modified boot chain after someone had an opportunity to modify it. [13:59] Like if configuring GRUB would become too much of a hassle. I might also just overlooked the 'encrypt disk' part during installation. [13:59] Sven_vB: did you have to do manual partitioning to get a LUKS side by side with Windows? [14:00] mgedmin, no, I used a proprietary commercial installer that did it for me. [14:04] (just to clearify my thoughts and for later research: LURKS would be (part of) the 'default' encryption protocol? [never heard of it before]: AKA that's something i could expect to find on the general install image or do i need extra tools for it? ^^; [14:04] *LUKS :seenoevil: [14:04] at one point I had to manually install a couple of packages so I could mount a LUKS volume from nautilus [14:05] libblockdev-crypto2, to be specific [14:05] ah, but https://bugs.launchpad.net/ubuntu/+source/udisks2/+bug/1757321 is fixed so that should be always installed now (if you have udisks) [14:05] Launchpad bug 1757321 in udisks2 (Ubuntu) "udisks2 must depend on libblockdev-crypto2 and libblockdev-mdraid2 instead of suggests [Can't mount encrypted USB drive after upgrade to bionic]" [High,Fix released] [14:06] and I think always mounting LUKS volumes from /etc/fstab doesn't require this package anyway [14:06] tl;dr: LUKS should work out of the box using the general image [14:06] mgedmin: i see, sweet! i might give it a shot later this week then! [14:07] trying to uninstall linux-image-5.8.0-25-generic it insists on installing linux-image-unsigned-5.8.0-25-generic, but I already have linux-image-5.8.0-26-generic and linux-image-5.8.0-26-lowlatency [14:08] wouldn't either of the latter fulfill any dependencies for linux-image or whatever depends on the kernel package? === leaftype2 is now known as leaftype [14:18] heeen, you could mark both the normal and unsigned for uninstall, then apt should either do it or tell you why not. [14:20] hello everybody! I wanted to know if Thunderbird is going to be upgraded in Ubuntu 20.04? It is using the ESR version which is pretty obsolete [14:21] flaburgan: you can try the thunderbird snap if you like [14:21] hm I was wrong, there isn't any ESR version of thunderbird, the previous was 68 and the current is 78. So 20.10 has 78 but 20.04 is at 68. [14:22] Ubuntu doesn't upgrade software in old releases as a rule; browsers are an exception [14:22] for Thunderbird to be upgraded it would have to be very broken, or have a security bug that's very hard to backport [14:31] hiya, under what circumstances would 20.04 systems have an /etc/resolv.conf that's symlinked to /var/run/NetworkManager/resolve.conf yet have NM manage DNS via systemd-resolved anyway? [14:33] perhaps somebody manually created that symlink? [14:34] Could it be there becuse an user upgraded from an earlier verison that was NM-only or something like that? [14:36] There's a bug with NM and DNS wit WireGuard that I'm looking to work around, and I want to know if this is a widespread config or not [14:40] oh, if I remove /etc/resolv.conf and reboot, NM will create a file that's not a symlink in /etc/resolv.conf [14:41] hm [14:42] I can't say I remember when /etc/resolv.conf was ever a symlink to /var/run/NetworkManager/* [14:43] maybe it wasn't, but my gripe is that by default it's a symlink to /var/run/systemd/resolved/resolve.conf or such, but if a user removes it, it gets replaced with a file that's managed by NetworkManager. [14:45] interesting [15:04] ah, crap. I tried yet another re-install of 20.04.1LTS amd64, this time in UEFI and now, after installation is complete, it only boots to a still-non blinking _ <-cursor on the top left side of the screen [15:05] oh now it's starting to somewhere, it was just VERY stuck [15:05] took like 5 minutes on this i7 core, AH [15:05] YES! [15:06] that was the key, to switch from BIOS to UEFI mode [15:06] (and to reinstall) [15:06] now it's booting into zfs ssd [15:15] after running the initial install package and rebooting, I get the same stuck-on-the-top-left-underscore (_)-non-blinking [15:16] I'll wait it out like I did the last time, but that is very weird. when I first got that working within Legacy BIOS mode (before ??? happened and it didn't reboot except to grub/grub rescue anymore), it booted way faster [15:17] I wonder if that is normal that it gets stuck in that stage for 1-2 minutes or is that just because there was around 150megs of important updates like ubuntu base, linux-firmware, etc, etc. [15:17] now it's booting but yep, that took a couple of minutes of terror [15:19] no, getting stuck for a couple of minutes during boot is not normal [15:20] can you switch consoles during that time? or press esc to see boot messages? [15:20] mgedmin: nope, just did another run of apt update + upgrades just to be on the safe side. [15:21] mgedmin: rebooted, and now again I've got the _ on top left screen, probably will take that 2mins to dissolve [15:21] I'm connected to openvpn server but there is no internet nor I can ping the server's IP, I already allowed ip forwarding and added the iptables rule: -t nat -A POSTROUTING -s 10.2.0.0/24 -o eth0 -j MASQUERADE ; any idea? [15:21] mgedmin: i did choose to install the system as zfs, i wonder if that was a bit too experimental of a choice :P [15:23] yep it's defo stuck again the same way as previously. this is a fresh install, so should I try to install the whole thing with the default selection? or is it likely that it's related to something else? [15:23] how about figuring out what's happening instead? [15:24] is this blinking cursor thing happening after grub? [15:24] yep, booting up, now it's kinda sluggish on the rotating ubuntu circle [15:24] this didn't happen previously [15:24] I think there's usually some extra logs on tty12 if you can switch there? [15:25] /usr/share/doc/systemd/README.Debian.gz describes ways to debug startup/shutdown problems with systemd [15:25] won't take keyboard input atm [15:25] no keyboard input is bad :/ [15:25] yup can't switch tty [15:26] it did run from that drive like, just now, just fine :D then on the first boot, stuck cursor for a couple of mins in top left, then this [15:26] it's still in the low-res, single screen boot mode too. [15:27] the circle is still rotating and all that, but it is sloo-oooooooooo-ooow [15:27] WD Blue 250GB SSD [15:27] (and this is an i7 core, albeit older gen., with 16GB RAM...) [15:28] exception Emask 0x0 SAct failed command: WRITE DMA EXT [15:28] of ata1.00 [15:29] uhhh dying SSD? [15:29] yep it seems to be stuck,, 345 secs in and it gets stuck with exception Emask 0x0 SAct 0x0 SErr 0x0 action 0x6 frozen [15:29] INFO: task zpool:333 blocked for more than 241 seconds. [15:29] if it boots, check the disk's self-assesment with smartmontools [15:29] Tainted: P O 5.4.0-52-generic #57 Ubuntu [15:30] 437 seconds in [15:30] ......... [15:31] it seems to be going from one address to another very slowly as in there's the next spot, frozen, and the next one frozen [15:31] "echo 0> /proc/sys/kernel/hung_task_timeout_secs" disables this message. [15:32] yeah, and then you have a frozen server with no indication of why it's frozen [15:32] I/O error with the dev sda again [15:32] I/O errors ... 560 secs in. Still booting. [15:36] should I try enabling Intel's VT ? I've disabled it for security, how about execute disable bit ? [15:36] that one's enabled [15:37] first of all figure out if your SSD is really dying [15:38] yep that would be the ,.. thing to do [15:38] mgedmin: but I've tried with two ssd's now [15:38] (also, which Intel VT? if you plan to use virtual machines, you want it enabled) [15:38] and at least in bios mode the end result was the same with the both of them [15:38] install smartmontools and run smartctl -a /dev/sda to see the drive's self-assesment [15:38] I mean they booted to grub [15:39] speltner, i'd try to replace the cable [15:39] there's a GUI version of that in gnome-disks in case this is a desktop [15:40] SMART is handled by the disk firmware itself, if it says everything's fine, then maybe check cables [15:40] ioria: you think it might be something as ridiculous? I tried googling up some of the I/O error messages and actually found out that a lot of people have had similar booting problems if the boot drive hasn't been in a certain sata port of the motherboard [15:40] yeah I took it out of the computer and checking it now here on another endpoint [15:41] speltner, if you have a spare one, it's easy to check and rule it out [15:51] mgedmin: ok so this is interesting, sudo fdisk -l seems to point to multiple problems in the drive, but that they can be fixed with write? [15:51] ah , I don't have a screen here , brb [15:52] the only way fdisk could cause disk read errors is if you have a partition table pointing to areas outside the size of the disk [16:16] hi is this channel for lite as well? [16:17] Im wondering if someon could help me. I am trying to install ubuntu 20 LTS as a guest on Qemu, but on boot it cant find the LVM disks properly [16:17] joesgarage: lite? [16:17] lotuspsychje, linux lite [16:17] Is it possible to install ubuntu server without LVM? [16:17] uses ubuntu 20 [16:18] joesgarage: no, we can only support ubuntu and its official !flavours [16:18] lotuspsychje, thanks [16:19] endersending: you might want to try #ubuntu-server [16:19] thanks [16:54] https://www.brighttalk.com/webcast/6793/448578 [16:55] in #ubuntu-discuss please luna [16:56] lotuspsychje: is #ubuntu-discuss like #ubuntu-offtopic? [16:57] hylian: yes, but more ubuntu discussion minded [16:57] lotuspsychje: cool, thanks. [17:56] hello [17:56] so; I went to get a brand new SSD [17:56] any ideas on what's more preffered on Ubuntu 20.04.1LTS installation, for stability, compatibility, etc.... to use BIOS or UEFI mode? [17:57] if I remember correctly, UEFI made it a little bit easier for linux to see what's wrong if something goes haywire, or when [17:59] some years ago, the recommendation was that you should only use UEFI mode if you're double-booting? [17:59] UEFI is the modern option [18:01] mgedmin: yeap but then again I read from just a few years ago that BIOS would be less likely to cause problems, and that UEFI would be like , for dual boot purposes. And that Intel mobo is around 10 years old already so ... [18:01] megoix: I have a nagging feeling that UEFI + a mobo that old = uh oh [18:02] (sorry , mgedmin ) [18:02] I also did get a new SATA cable [18:02] although, well ... [18:05] ha, while I was away at the shops, the PC had booted into login screen :D [18:05] lol === denningsrogue3 is now known as denningsrogue [18:06] I still contend that I found the right answer the other day when I noticed that the grub-crash right after first reboot has been a problem across distros [18:06] for years [18:07] and the fault line in that one is if in some mobos, you've stuck the /dev/sda1 (or: the primary boot device) into a sata port that is not port#1 [18:07] it was something as trivial as that [18:07] speltner: please dont use this channel for every step you are doing [18:07] lotuspsychje: I'm describing the problem here [18:07] lotuspsychje: I thought this was the help channel? [18:08] lotuspsychje: or was there a separate channel for help requests? [18:09] speltner: try asking 1 question all in one line, then wait until volunteers pickup on your issue instead of monologues with the enter button [18:10] lotuspsychje: alright, I get it, sorry about that. I'm just a bit perplexed over the entire ordeal atm and really don't know wheere to start tackling the issue from. OH, btw, is there any possibility that Linux itself might do physical damage to SSD's i.e. by overwriting them (i.e. into the designated buffer cell areas) or so? That might be one culprit here. I read something about LVM that it might do [18:11] such a thing [18:11] speltner: try aproaching your issue systematicly, one step at time, where you are stuck now, and ask your question till someone can sort you out [18:31] I have to install about twenty packages from the commandline in Ubuntu 20.10. What is the best way to do this? [18:32] YWH_1: sudo apt install .... [18:32] They are all in the same directory [18:33] OK [18:33] YWH_1: sudo dpkg -I *.deb [18:33] sorry, -i, not -I [18:33] YWH_1: why do you have local packages to install? [18:34] Because the program in the Ubuntu repository is defective. [18:34] YWH_1: oh? Which program? [18:35] Libreoffice 7.02 I need to install 7.03 to fix the bugs [18:39] you can sudo apt install ./*.deb these days (and it's better than dpkg -i because it still checks dependencies) [18:40] it is important to use ./*.deb rather than just *.deb, otherwise apt will think you're passing package names rather than file names [18:41] Thanks [18:42] how to setup tor network so all my request goes via it? [18:48] Sometimes I will launch an app from the terminal, and it will pop up behind the terminal and notification will announce that the app is ready. Is there some way to make this not happen? I am running Ubuntu 20.04 [18:49] Please [19:03] hi, I want to install Ubuntu on Intel Compute Stick m3, what's the leanest and lightest flavor/configuration out there? [19:03] I need graphical environment and Firefox, that's all [19:19] jadax: i don't have experience with such sticks but Lubuntu seems to be the most lightweight flavor [19:23] jadax: there's #lubuntu for your lubuntu questions. #xubuntu would be another option. but you'll be very short of resources either way - don't expect to be able to run multi-tabbed browsers. [19:25] tomreyn I just need Firefox, 1 tab , youtube [19:25] that's the CPU I've got, https://ark.intel.com/content/www/us/en/ark/products/88198/intel-core-m3-6y30-processor-4m-cache-up-to-2-20-ghz.html [19:25] as well as 4GB RAM and eMMC as storage [19:26] oh it's got 4 GB, that's better than i assumed [19:26] yeah, it's one of the more capable sticks [19:27] it looks more capable than my old acer netbook which runs xubuntu/chrome ok (need to be patient during boot) [19:27] a share of the 4 GB will go to the GPU, though [19:28] anyways, should be ok, you could possibly even run the default desktop environment on it. [19:28] so would you recommend lubuntu or xubuntu at this point? [19:29] whichever you prefer [19:30] ok ok [19:31] lubuntu, xubuntu has become a bit heavier since the switch to GTK3 afaik === disillusion is now known as Nov1-Vaccine [19:32] jadax: you could try them both and then make up your mind [19:33] and just to confirm, lubuntu fully supports webbrowser (e.g. firefox) and should be capable of youtube playback? === Nov1-Vaccine is now known as disillusion [19:35] okay, the installer just crashed again with the brand new SSD and a SATA cable but this time I got a confirmation that it's a part of a critical bug in glibc === disillusion is now known as RussianInterfere [19:35] jadax: i don't see why not [19:36] jadax: yes and yes, it'S the same foundation as ubuntu, just the DE differs === RussianInterfere is now known as disillusion [19:36] Maik: I respectfully disagree strongly about lubuntu being the most "light-weight", imho it's among the most bloated ones [19:36] Maik: for lower-end computers I'd definitely recommend Xubuntu / Xfce [19:37] (goes for SBC/SoC computers as well) [19:37] oh, understand now! thanks, will try lubuntu then [19:38] jadax: lubuntu is ubuntu with the LXQt desktop [19:38] Maik: but what you're saying about Lubuntu being light-weight is simply not correct imho [19:38] Maik: ubuntu MATE / xubuntu are the most light-weighted ones out there [19:39] that's not an opinion, it's a fact [19:39] no need to rub it in dude [19:39] i asumed it was one of the most lightweight ones, if not, then not [19:40] Maik: I'm just saying. Been running XUbuntu-variants all the way from 18.04 LTS on a Raspberry Pi4B for what, almost 1,5yrs now [19:40] i never tried it myself to be honest, let me have a look at a live session [19:40] Maik: the heaviest ones are kubuntu and lubuntu, by far [19:40] lol [19:41] ever seen Gnome? [19:41] again, the only "subjective" part in that one is how well they are optimized for one's corresponding setup [19:41] Maik: yes [19:41] Maik: into the same category it goes [19:41] kubuntu uses up to 580MB idle, Gnome is around 890MB to 1.1GB [19:41] but obviously all of this depends on your hardware specs and what do you want to accomplish [19:42] of course [19:43] Maik: well, Ubuntu MATE and XUbuntu are a fraction of that. Right now I've got a bunch of Firefox tabs open as well as multiple other programs on a 4GB RPi4B. Yes, I am typing this from a RPi4B running Xubuntu. Feel free to ctcp version me [19:43] :D [19:43] i run main Ubuntu 20.04 on a Pi4 4GB [19:44] but we're chatting offtopic here imho [19:44] Maik: yep, same here [19:44] speltner I tried ubuntu mate on r pi 4b, I wasn't happy with youtube playback [19:44] pretty choppy, even scrolling up & down was laggy,, not smooth [19:44] it improved jadax [19:44] that's why I got compute stick with Intel m3 CPU [19:45] speltner: well yes and no, because when it comes to low-end performance, xubuntu and ubuntu mate are far lighter than any other distros, ubuntu Mate especially being perhaps the lightest version of them all [19:45] but let's take it to -discuss. This channel is for support with Ubuntu in the first place [19:46] jadax: did you try it out with wimpy's desktopify? it should give a bit of a boost. also, RPi4B's essentially don't do anything well if you don't overclock them a bit, and that's just my personal opinion ;) [19:46] speltner: "desktopify" from github [19:46] whoops, jadax ^ [19:47] jadax: I'd be interested to know what was your distro solution? [19:48] speltner I used this, https://www.tomshardware.com/how-to/play-netflix-raspberry-pi [19:48] Okay, I got a way faster boot result now with the new SSD+SATA cable. At least it didn't get stuck on that "_" underscore-cursor. booted up in around 10sec [19:49] I wonder if it really could've been the SATA cable tho, seems a bit crazy but then again I noticed that, since they're around 6-8 years old, the plastic in them felt kinda brittle [19:50] that workstation haven't been meddled around with too much tho so ... but weird. I'll keep you posted if the problem persists [19:51] is there any way to troubleshoot why `pam-auth-update --remove sss` actually adds the SSS module to PAM configs instead of removing it? [19:52] Now there's btw this 142,5 MB updated software package again, I believe that was the nail in the coffin the last time ... [19:53] I'll try it out nonetheless, I've got error reporting enabled and so forth. o7 [19:54] g3poandlsl: try adding --force and see if thta changes anything? perhaps it thinks you've made local changes that ask for sssd to be configured on [19:56] sarnold, same results with --force. It's weird because I can disable SSS via the TUI, but if I run the non-interactive command after that, it gets re-enabled [19:56] g3poandlsl: it's probably worth a bug report, anyway [19:56] sarnold, thanks [20:02] (update on my ssd/install woes; regarding 20.04.1LTS -- after the latest big update package it WORKS[for nnow, bought a brand new ssd+sata cable today]) ! perhaps something was fixed as well since now when the install process crashed, the bug it crashed into was marked critical and it might've been patched during the course of today [20:03] speltner: a SATA drive is the best 60 quid you'll ever spend on computer bits [20:03] gordonjcp: well that is not my main workstation and I just spent 600-something euros on a 4TB Samsung Pro SSD, so ... :D [20:04] (I'm running this one on a brand new cheapo WD Blue, tho) [20:05] speltner: :- [20:05] speltner: :-) [20:07] I have a problem, someone can help me? [20:09] Fabgiese: not at this point. you'll need to discuss the problem first of all. ;) [20:10] Fabgiese: in other words: you're generally welcome to just come here and immediately start describing the problem you're facing and are looking for assistence with. === KindTwo is now known as KindOne === JanC_ is now known as JanC [20:38] Hi, I got this error when trying to upgragde from 18.04 LTS to 20.04 LTS: https://paste.ubuntu.com/p/63m6c7RMMq/ [20:39] I can't delete all ppa packages [20:40] blackbird_0: why can't you delete all the ppa packages? [20:41] leftyfb: Because I need them (there is a lot) [20:42] blackbird_0: you need to remove the packages and the PPA entries. After you finish the upgrade you reinstall them [20:43] blackbird_0: upgrading your release is not supported when PPA's are enabled and packages installed from said PPA's [20:44] blackbird_0: use "ppa-purge" to make the process easier [20:44] leftyfb: Is there a way to do that in one command line? [20:45] blackbird_0: no. Use ppa-purge on each of the PPA's individually [20:47] That way I'll stay all night [20:48] blackbird_0: if you have that many PPA's, it's a perfect time to re-evaluate the usefulness of them. I have 9 PPA's myself. It doesn't take that long to handle them. [20:49] Is there a way to list all PPAs? [20:50] blackbird_0: apt policy [20:50] that shows all enabled repositories including PPA's [20:51] leftyfb: Sorry, I meant all installed PPAs? [20:51] blackbird_0: look in your sources lists [20:52] blackbird_0: /etc/apt/sources.list and /etc/apt/sources.list.d/* [20:52] blackbird_0: awk '$1 ~/deb/ && $2 ~/ppa/ {print $2;}' /etc/apt/sources.list /etc/apt/sources.list.d/*.list | sort -u [20:53] blackbird_0: sarnold: the line sarnold just posted will only show you PPA's. Not all 3rd party repos you might have added [20:53] ah good point [20:53] can purge-ppa cope with non-ppas? I can't recall [20:53] don't think so [20:53] not sure though, never tried [20:53] I was wrong, there is not a lot :) [21:05] pseudo-urgent question (since I can't see my display :D), is there a method to make the WinKey+arrow combo skip over displays in a multi-display setup? [21:06] (on Ubuntu 20.04.1LTS). Thanks [21:06] (^ I mean one window hopped into a monitor that I'm physically unable to see ...) [21:09] leftleg, sarnold: thanks a lot guys, I will remove the listed PPAs, and tru again [21:09] okay, a temporary solution for the winkeys+arrows not skipping from one display to another was to just disable the screens that I can't see, a few important windows hopped into there and that was the workaround for it :) [21:10] blackbird_0: cool :) [21:11] geez, I'd hope there was a way to hop from display to another using winkeys and arrows; sigh [21:14] Shift+WinKey+Letf or Right Arrow [21:16] @speltner [21:30] rootsandculture: I think I tried that but it didn't work. But thanks a lot, I will try it out; it's always an exciting moment when I reboot the system and see whether or not the whole thing collapses on itself :D so far so good! at least 5 reboots done! [21:32] rootsandculture: yep it works! thanks for the tip! ^__^ (p.s. just learning to use the new bells'n'whistles of 20.04.1 here so I'm sorry in advance) [21:38] Fabgiese: can you describe the problem? [21:45] yes, sorry, my son needed help, he is one year old [21:45] My problem is, my son accidently shutdown my computer [21:45] when it restart, I needed to use fsck manually [21:46] And it don't solved the problem [21:46] when it restart again [21:46] Enter in emergency mode [21:46] And my ubuntu don't start again [21:47] and I use sudo grub-install /dev/sda after [21:48] but it don't solved the problem as well. [21:49] sorry for my english, it's quite bad. === nshireTimeout is now known as nshire [21:51] platvoeten [21:51] running grub-install by hand may break things, if you didn't give the correct parameters for your configuration [21:51] I'm a new user in linux and Ubuntu is my first SO [21:53] sarnold What I should to do? [21:54] What should I do?* [21:56] Fabgiese: I don't know; what error do you get when you try to boot your computer? [22:01] sarnold Failed to start File System Check on /dev/disk/by-uuid/2f90... [22:04] Fabgiese: is there anything more specific about why it couldn't start the check? [22:04] In journalctl -xb don't have any red trouble [22:05] I don't know, my computer stop running when my son tap shutdown button, and fsck don't running more. [22:05] Where should I see the problems? [22:06] journalctl -xd? [22:06] maybe; or it might be printed to the console [22:08] here [22:09] Have the same problem [22:09] fsck failed with exit status 4 [22:09] 4 Filesystem errors left uncorrected [22:09] that's not a good sign [22:10] maybe it just means you need to run fsck by hand and force it to try to correct errors? [22:10] yes [22:10] It's my first action, I use fsck sda3 [22:11] I don't know if it is right. [22:11] I have no idea what happened but my brother is using popos and i know this is ubuntu but im hoping somebody knows how to fix this [22:11] he inverted right and left click [22:11] and the home button dosent work [22:11] and i have no idea how to fix that ive never seen anything like that before [22:11] I don't know if it is right. [22:12] How I run fsck by hand and force it? [22:12] kenwoodfox: check xmodmap -pp output [22:13] sarnold: sure thing [22:13] ugh everything keeps slipping into like [22:13] idk what its called, the windows want to tile [22:14] but a reboot seems to have fixed the inverted mouse [22:14] xmodmap 1 error encountered, aborting [22:14] unable to open file output for reading [22:15] Fabgiese: maybe try fsck -y /dev/sda4 --- this is a dangerous operation, it'll try to repair what it can, and in doing so, it might make things worse [22:15] If two apt sources provide the same version of the same software, how do I tweak which one takes priority? [22:15] kenwoodfox: I wonder if your brother also decided to remap other keys to do window manager operations? heh [22:16] Bilge: iirc repository pinning can do that [22:16] `apt-cache policy` shows one has priority 100 and the other 500, but I want to switch them somehow [22:16] sarnold: I have no clue, i disabled tile windows, aparently that was turned on, Super / launches and switches applications [22:16] i thnk thats supposed to be the home key? [22:17] sarnold my system is in sda3, and when I tried this operation is aborted because sda3 is mounted [22:17] i reset them to default but the home button still does nothing [22:17] Fabgiese: oh, that makes sense. sigh. do you have a usb installer handy? [22:18] yes, I have [22:19] Fabgiese: I suggest booting into that, and then use fsck from that installer image to try to repair the filesystem [22:19] I need umount sda3? [22:21] Fabgiese: yeah; that can be difficult to do if you don't have another root filesystem already mounted that you can pivot_root into -- it's much easier to reboot into another filesystem like a USB installer [22:25] I open the usb installer, but what's the root password? It's not the same. [22:26] How do I pin this (default) package: 100 /var/lib/dpkg/status? [22:26] I never use usb installer for it [22:26] I don't know how to address it in the pin file line [22:28] `DEBIAN_FRONTEND=noninteractive; DEBIAN_PRIORITY=critical; sudo apt-get install ubuntu-mate-desktop -y` even though I'm specifying noninteractive why am still being asked for input? [22:31] when I try the console inform that I don't have permisson [22:37] Fabgiese: hmm, that's a surprise; I didn't think we had passwords on the installers.. the whole point is to use the thing to install an OS :) [22:37] Fabgiese: try adding init=/bin/bash to the kernel command line [22:38] Fabgiese: Ubuntu doesn't have a root password [22:38] Fabgiese: from the USB live image, you ought to be able to sudo without a password [22:52] sarnold I got it [22:53] thx for you help [22:53] Fabgiese: were you able to repair your filesystem? [22:53] I fsck in kive installer [22:53] but I need log root with sudo su [22:54] In second fixed it. [22:54] sweet [22:55] It can't use fsck in main system, you need log other system for it, and I use live cd confirm your orientation. [22:55] Need umount main system === cassandra is now known as cass [23:46] upgraded to 20.10. on xorg server works normally until the display sleeps, when it restores, windows have a series of short horizontal lines around areas of movement/scrolling. when windows are closed, a wide "static" pattern flashes on the screen, always in a shape that resembles a wide but short rectangle immediately above a right triangle where two points are on the right side. i tried to capture this in simplerecorder, but the art [23:46] acts do NOT show up on the capture [23:46] switching to wayland seems to bypass the issue