| * Digit wonders about ungoogled-chromium in devuan... and if the chromium package can be ungoogled, or if need go off-repo to get | 07:58 | |
| * Digit new to chromium / ungoogled-chromium, just a curiosity | 07:59 | |
| xrogaan | which is better to backup a disk before resizing: rsync or dd? | 08:26 |
|---|---|---|
| xrogaan | also, is there a way to start user daemons at the start of a user session? | 08:27 |
| xrogaan | with fluxbox, you can just drop a line in the init file. | 08:27 |
| gnarface | xrogaan: rsync is faster but dd can also preserve partition structure | 08:31 |
| gnarface | and filesystems | 08:32 |
| gnarface | (whereas rsync will only preserve contents therein) | 08:32 |
| xrogaan | no the partition needs to go | 08:32 |
| gnarface | frankly though from years of experience i recommend tar because you don't want your backup system to be complicated. you want it linear and based on software that hasn't changed much in 30 years | 08:32 |
| xrogaan | I have a very old partition thingy, and I need to make / bigger. To do that I have to remove swap and reduce /home | 08:33 |
| gnarface | well, you might still want the disk image backup to contain the original partition structure just so you can go back to it in case you hose it somehow trying this | 08:33 |
| xrogaan | so, tar the content, resize, untar? | 08:33 |
| gnarface | tar content, resize. | 08:33 |
| gnarface | you shouldn't need the backup unless something went wrong | 08:34 |
| xrogaan | but isn't resizing destructive in nature? | 08:34 |
| xrogaan | I meant for the /home partition that will get reduced. | 08:34 |
| gnarface | no, only if you screw it up | 08:34 |
| gnarface | but there are important caveats you should know if you're asking this | 08:34 |
| xrogaan | oh, and I have a weird thing though. | 08:35 |
| xrogaan | i have an extended partition, containing both swap and /home | 08:35 |
| gnarface | you're gonna use gparted to do the resizing? | 08:35 |
| xrogaan | most probably yeah | 08:35 |
| gnarface | ok, it should protect you but had a early history of not doing so | 08:36 |
| gnarface | so let me tell you how to cover your ass | 08:36 |
| xrogaan | wait a second first :) | 08:36 |
| gnarface | ok | 08:36 |
| xrogaan | I believe I need to remove the extended partition in order to resize /, which is outside of it. | 08:36 |
| gnarface | maybe, i don't know for sure without seeing it | 08:36 |
| gnarface | you DO have to look in the menu of gparted to find the panel where it shows you filesystem support status | 08:37 |
| xrogaan | https://files.catbox.moe/w6ydim.png | 08:37 |
| gnarface | make sure it includes resize and shrink for the relevant filesystems | 08:37 |
| gnarface | (it wasn't always smart enough to stop you if you tried it on a filesystem that wasn't supported) | 08:37 |
| xrogaan | 12GiB as root isn't nearly enough nowadays. | 08:37 |
| xrogaan | Was plenty 10 or so years ago. | 08:37 |
| gnarface | the second thing you need to know is just to not shrink the partition to smaller than the data inside it. | 08:38 |
| gnarface | (it might succeed and truncate your data) | 08:38 |
| gnarface | the third thing you need to know is that if the partition is fragmented, it still might not be safe to shrink to the size of the contained data, becasue some of it may be near the end of the filesystem | 08:38 |
| gnarface | (that's where the tar then untar trick comes in handy; that's a quick&dirty way to defrag any filesystem from linux) | 08:39 |
| xrogaan | isn't ext4 fragmented by default? | 08:39 |
| gnarface | i think that's all you really need to know, but ask questions if anything is vague | 08:39 |
| xrogaan | by nature, the filesystem distribute the files all over the disk | 08:39 |
| gnarface | yes, journaling filesystems are "fragmented by nature" you would say | 08:39 |
| gnarface | as in, the journal mitigates the cost of fragmentation so it is not a concern. that's why disk defrag isn't a linux tool that exists | 08:40 |
| gnarface | but again - tar then untar will do the trick | 08:40 |
| gnarface | however | 08:40 |
| gnarface | gparted is *supposed* to be smart enough to just warn you and refuse, or relocate the data as necessary | 08:40 |
| gnarface | these days anyway | 08:40 |
| xrogaan | but I still need to remove the extended partition, as shown in the screen shot :) | 08:41 |
| gnarface | but it's one of those features it has had enough of a history of regressions on that you definitely want to back up your data first to a separate physical disk | 08:41 |
| gnarface | oh, can you put the screenshot in imgur? | 08:41 |
| gnarface | actually maybe i can simplify further without that: you don't need an extended partition at all on the MSDOS partition table format unless you have more than 4 partitions | 08:41 |
| xrogaan | i don't have a cli tool for imgur | 08:42 |
| xrogaan | maybe i do | 08:42 |
| xrogaan | no, I don't | 08:43 |
| gnarface | how about transfer.sh? | 08:44 |
| gnarface | supposedly it can work with curl | 08:44 |
| gnarface | some day i'm gonna have to put up my own drop box so that i don't have to have the burden of personally validating the domain of everyone else's favorite one | 08:46 |
| xrogaan | what's wrong with catbox? | 08:46 |
| gnarface | ... even though that is objectively making the problem worse | 08:46 |
| gnarface | i don't know what's wrong with catbox, maybe nothing. the point is just that i'm not in the mood to risk it right now. it's nothing personal; things have been weird. | 08:47 |
| xrogaan | want me to xdcc the file to you? | 08:47 |
| gnarface | maybe you can just copy&paste the fdisk view of the data to me in a /msg? i'd have to poke a hole in my firewall for xdcc | 08:47 |
| xrogaan | It's just a png, wget it and it should be fine. | 08:49 |
| gnarface | if you have 4 or fewer partitions, or 5 partitions but one of them is the extended partition, you can seriously just omit it though | 08:49 |
| gnarface | the only complication is then they'll renumber | 08:49 |
| gnarface | but if you're resizing them the UUIDs change anyway, so if you're mounting by UUID you have to edit your fstab either way to clean up after this | 08:49 |
| xrogaan | https://termbin.com/g4pc | 08:52 |
| xrogaan | termbin isn't too weird for you? | 08:52 |
| gnarface | come on why does everyone have to test my resolve on this? | 08:52 |
| gnarface | i'm really not trying to be difficult | 08:52 |
| gnarface | just paste me what fdisk says instead in /msg | 08:52 |
| gnarface | it's plain text and should still carry the necessary data | 08:53 |
| gnarface | open the disk with fdisk, (fdisk [device]) then just hit the "p" key to print the partition table | 08:53 |
| xrogaan | I'm still puzzled as to why you prefer imgur over anything else | 08:53 |
| gnarface | nothing specific other than that it's widely enough known that it is a bigger liability for them to hack me than for me to get hacked by them | 08:54 |
| xrogaan | what are you, the CIA? | 08:55 |
| gnarface | you know it's funny, i've been accused of that before | 08:55 |
| gnarface | usually by employers who are trying to pull a fast one | 08:55 |
| gnarface | but no, i just notice things | 08:55 |
| gnarface | patterns | 08:56 |
| gnarface | incidentally, the /msg you sent me stopped short of the actual partition table data | 08:56 |
| xrogaan | yeah, because flooding is bad | 08:56 |
| gnarface | the last line i got was "Device Boot Start End Sectors Size Id Type" | 08:56 |
| gnarface | you got a flood protect warning from sending me a /msg? | 08:57 |
| gnarface | i thought they weren't enforcing that in private communications, i've never had anyone complain about that before | 08:57 |
| gnarface | i don't mind if you flood my /msg | 08:57 |
| xrogaan | ah, no, because / is a command | 08:57 |
| gnarface | just add a whitespace before it | 08:57 |
| xrogaan | /dev < dev not recognized | 08:57 |
| gnarface | right | 08:58 |
| gnarface | it will choke on "/dev" but it will pass " /dev" | 08:58 |
| xrogaan | You know what, it's fine. I'll delete he extended partition and untar the backup. | 08:58 |
| gnarface | i *think* i've already told you everything you need to know | 08:59 |
| gnarface | i just don't have any idea how well you read it | 08:59 |
| gnarface | that's really why i asked to see the screenshot in the first place | 08:59 |
| gnarface | i was just trying to double-check your work | 08:59 |
| gnarface | you know, before you make a mess | 08:59 |
| gnarface | resizing partitions can get messy | 09:00 |
| xrogaan | you can still see it, you chose not to because of patterns. Thanks anyway :) | 09:00 |
| gnarface | hey you're the one who can't send a /msg ... | 09:00 |
| gnarface | there's only so far past the mid point i'm willing to go | 09:00 |
| xrogaan | I spent more time trying to send the list of partitions than reading your recommendations, which were useful btw. | 09:01 |
| gnarface | as long as you made the backup first then you'll be fine either way | 09:02 |
| gnarface | oh, and ext* filesystems resize really slow. don't panic if it is taking a long time (many hours) | 09:03 |
| xrogaan | well, I can bypass that with a delete and recreate | 09:08 |
| xrogaan | anyhow, for the next people trying to test your resolve, you might want to come armed of a list of pastebin/file hosts | 09:08 |
| xrogaan | And if you're willing, there are some software I'd like to try in order to get around the hosting issue. | 09:09 |
| xrogaan | Might not be to your taste because of your firewall though. | 09:09 |
| gnarface | what software? | 09:09 |
| xrogaan | https://magic-wormhole.readthedocs.io/ ? | 09:10 |
| gnarface | hmmm | 09:10 |
| gnarface | it's in the repo | 09:10 |
| gnarface | based on python apparently | 09:10 |
| gnarface | what's it got over ssh tunnels? | 09:10 |
| xrogaan | you setup a wormhole code, then send a file to that code. | 09:11 |
| gnarface | a GUI or something? | 09:11 |
| xrogaan | cli mostly | 09:11 |
| gnarface | does it rely on a hub server or is it p2p? | 09:11 |
| xrogaan | https://magic-wormhole.readthedocs.io/en/latest/welcome.html#relays | 09:12 |
| xrogaan | you might be able to setup your own relay though | 09:13 |
| gnarface | hmm, they're not using cookies, that's something to their credit... | 09:13 |
| xrogaan | or I could just mail you the damn png | 09:13 |
| gnarface | ugh, websockets | 09:13 |
| gnarface | i have distrust of that | 09:13 |
| xrogaan | there is also that ipfs thing | 09:13 |
| gnarface | heh, yea you could have just emailed it to me but oddly enough that's the part where everyone else tends to back out | 09:14 |
| xrogaan | interplanetary file system | 09:14 |
| xrogaan | No, I backed out when you asked me to reformat the output of fdisk to be irc friendly. That's just stupid. I could also use a pastebin of your choice, so long as it supports pastebinit | 09:15 |
| gnarface | i really didn't think it would choke on the paste of that | 09:15 |
| gnarface | lemme test something.... | 09:16 |
| gnarface | Disk /dev/sdb: 931.5 GiB, 1000204886016 bytes, 1953525168 sectors | 09:16 |
| gnarface | Units: sectors of 1 * 512 = 512 bytes | 09:16 |
| gnarface | Sector size (logical/physical): 512 bytes / 4096 bytes | 09:16 |
| gnarface | I/O size (minimum/optimal): 4096 bytes / 4096 bytes | 09:16 |
| gnarface | Disklabel type: dos | 09:16 |
| gnarface | Disk identifier: 0x0f56d2ac | 09:16 |
| gnarface | Device Boot Start End Sectors Size Id Type | 09:16 |
| gnarface | " | 09:17 |
| gnarface | hmmm, they do break lines i see | 09:17 |
| gnarface | yea, sorry about that | 09:17 |
| gnarface | for the benefit of anyone else who was wondering, i did verify that gparted can do that. it can copy a logical partition out of the extended partition | 10:29 |
| gnarface | you do have to make space first, but it works | 10:30 |
| nemo | what kernel is beowulf going to use on release? | 14:36 |
| gnarface | it'll be the same kernel as debian is using, i assume | 14:39 |
| ham5urg | I'm using devuan and trying to install an USV, APC Smart 750 (750smt) | 16:04 |
| ham5urg | lsusb is fine and showing Bus 001 Device 008: ID 051d:0003 American Power Conversion UPS | 16:05 |
| ham5urg | But I got an error from apcuspd like: | 16:05 |
| ham5urg | Communications with UPS lost | 16:05 |
| mason | ham5urg: I always have to change the config on every install. How are you specifying the line? | 16:05 |
| mason | ham5urg: In particular, what's your UPSCABLE set to? | 16:06 |
| mason | ham5urg: Here, I have UPSCABLE usb, then further down, UPSTYPE usb, and DEVICE without anything after it - empty. | 16:07 |
| ham5urg | I did not change a lot, as apcusvd.conf is preconfigured for USB. | 16:07 |
| ham5urg | UPSCABLE usb | 16:07 |
| ham5urg | ok, device has /dev/ttys0 | 16:08 |
| ham5urg | I will change that | 16:08 |
| mason | clear that out | 16:08 |
| mason | and UPSTYPE ? | 16:08 |
| ham5urg | it was DEVICE | 16:10 |
| ham5urg | mason, you are right | 16:10 |
| ham5urg | thanks | 16:10 |
| mason | ham5urg: Good. Glad it's working. | 16:11 |
| tuxd3v | fsmithred, hello, does you are experiencing any problems bootstraping in [ host ] Ascii to create [ target ] Beowulf rootfs ? | 16:59 |
| yann-kaelig | Hi. Anyone else has an issue with gpg --gen-key and devuan ascii 2.0.vdi . Look like it's stuck during the prime generation | 17:00 |
| tuxd3v | entropy problem? | 17:01 |
| yann-kaelig | no idea. I'm waiting but this is now more than 5 minutes | 17:04 |
| fsmithred | tuxd3v, I haven't done a manual debootstrap lately, but I have build beowulf isos with live-sdk, which uses debootstrap | 17:30 |
| fsmithred | s/build/built/ | 17:31 |
| fsmithred | in ascii host | 17:31 |
| tuxd3v | fsmithred, thanks a lot | 17:33 |
| fsmithred | what problems you having? | 17:33 |
| tuxd3v | Your isos are created in same host for same target arch, I believe? | 17:34 |
| tuxd3v | I having problems Cross-debootstraping | 17:34 |
| fsmithred | host is amd64, and I build that and i386 | 17:34 |
| tuxd3v | I am in host x86_64 | 17:34 |
| tuxd3v | and I build for severam ARM targets | 17:35 |
| tuxd3v | several | 17:35 |
| tuxd3v | but in ascii all plays nice | 17:35 |
| fsmithred | I think arm might be trickier. At least i386 will run on 64-bit hardware | 17:35 |
| tuxd3v | when in host ascii and trying to Cross-debootstrap to arm.. it gives me several errors related with qemu | 17:36 |
| tuxd3v | and Glibc | 17:36 |
| tuxd3v | I will have to finda option for it.. | 17:36 |
| tuxd3v | Anyway, thans a lot, with that, I already eliminated some variables... | 17:36 |
| fsmithred | I don't know enough to help | 17:36 |
| tuxd3v | I mean reduced the number of locations were problems could be | 17:37 |
| fsmithred | yup | 17:37 |
| tuxd3v | I already helped! thanks a lot! | 17:37 |
| tuxd3v | :) | 17:37 |
| tuxd3v | I -> you | 17:38 |
| amnt24 | Why does Beowulf have an older version of openvpn (2.3.10-1 vs 2.4.0-6) than Ascii? | 20:44 |
| golinux | amnt24: Because beowulf hasn't been officially released yet. Still much to do | 20:59 |
| nemo | https://seclists.org/oss-sec/2019/q4/122 | 21:57 |
| rhYtest | Has anybody here successfully run any version of ProxMox on Devuan? | 22:14 |
| rhYtest | There are instructions for building on Wheezy, would that theoretically work for Devuan? | 22:17 |
| rhYtest | And odd follow up question: Is there still development of refracta? And which version of that should I use atm, or should I not? | 22:19 |
| golinux | #devuab: | 22:33 |
| fsmithred | yes, refracta is still alive | 22:34 |
| fsmithred | there are 9.1 isos (ascii) | 22:35 |
| fsmithred | and there are some preliminary no-X beowulf isos | 22:35 |
| fsmithred | rhYtest, ^^^ | 22:35 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
