| 00:01.24 | *** join/#devuan sh4 (~sh4@unaffiliated/sh4) |
| 00:13.56 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-78-12-58-52.clienti.tiscali.it) |
| 00:16.34 | *** join/#devuan underd0g (~underd0g@93.140.28.15) |
| 00:26.32 | *** join/#devuan reetp (~john@239.red-80-59-216.staticip.rima-tde.net) |
| 00:29.44 | *** join/#devuan IoFran (~Icedove@189.231.21.160) |
| 00:34.33 | DocScrutinizer05 | timeless: odds are the HDD will re-allocate spare blocks to the blocks with read errors during that procedure |
| 00:35.30 | DocScrutinizer05 | a smartctl -a will show the result, once md finished syncing |
| 00:36.43 | DocScrutinizer05 | just runs a `e2fsck -tvc /dev/sdb3` on that borked unused disk |
| 00:37.27 | DocScrutinizer05 | will take an hour or so |
| 00:37.55 | DocScrutinizer05 | on a raid md should probably do that for you |
| 00:41.41 | DocScrutinizer05 | idly wonders if enterprise HDD couldn't do a transparent write_verify-read_fix-rewrite |
| 00:42.47 | DocScrutinizer05 | unless under very heavy sustained write load, in which case there's no sufficient time to process the buffered verify-reads before buffer overruns |
| 00:43.14 | furrywolf | you can wonder that, but I'll wonder why anyone is still using spinning rust for anything other than throw-away archival data. |
| 00:44.44 | DocScrutinizer05 | sure sure |
| 00:45.06 | DocScrutinizer05 | for now it's a matter of bits per bang (literally) |
| 00:45.55 | DocScrutinizer05 | and it's not exactly like lifespan (aka write cycles until EOL) of a SDD were really comparable to HDD |
| 00:46.51 | DocScrutinizer05 | so the medium for R/O "archive data" is more the SSD than a HDD |
| 00:47.01 | furrywolf | SSDs have a write limit that's far in excess of what will ever be written to them under most workloads... and hdds are so much slower than you'll hit the mechanical life limit around the same amount of data anyway. :P |
| 00:47.34 | DocScrutinizer05 | hmm, citation needed :-) |
| 00:47.55 | DocScrutinizer05 | at least for the latter half |
| 00:49.45 | DocScrutinizer05 | I'd think on a server HDD you could write a F*H*S blocks err sectors 10000 times per second and the mechanical wear is minimal |
| 00:49.59 | DocScrutinizer05 | sorry, per minute |
| 00:50.15 | DocScrutinizer05 | according to RPM, not RPS |
| 00:52.30 | DocScrutinizer05 | though, prolly most HDD only have one read/write electronic and a mux to heads, so only 1*S, not H*S (forget about the F, it's nonsense anyway) |
| 00:52.52 | DocScrutinizer05 | should have been C and still was wrong |
| 00:52.57 | *** join/#devuan tsadok (~nathan@cpe-75-185-35-110.columbus.res.rr.com) |
| 00:53.09 | furrywolf | a mid-grade ssd might be rated for 15-30 drive writes per day for 5 years. on a 1tb drive, that's 15-30tb/day * 5 years = 27-55 petabytes endurance. a random hard drive might have a write speed of 150mb/sec. writing full speed non-stop would take 6-12 years to write that same amount of data. |
| 00:53.39 | furrywolf | your spinning rust is not likely to last 12 years of 24/7 full speed writes. |
| 00:53.59 | DocScrutinizer05 | ack |
| 00:54.19 | DocScrutinizer05 | though I'm a tad puzzled about the values and the math |
| 00:56.05 | DocScrutinizer05 | !150*60*60*24 |
| 00:56.05 | infobot | 12960000 |
| 00:56.48 | DocScrutinizer05 | aka ~13TB / day, aiui |
| 00:57.16 | furrywolf | 15-30/day * 1tb = 15-30tb/day. 15-30tb/day * 365 days/year * 5 years = 27375-54750tb. 27375-54750tb / 150mb/sec = 182500000-365000000 seconds. 182500000-365000000s / 60 sec/minute / 60 minute/hr / 24 hours/day / 365 days/year = 6-12 years. |
| 00:57.56 | furrywolf | ok, so writing full speed, this hdd writes 13TB/day. but a 30 DWPD SSD is rated to write 30TB/day for 5 years. |
| 00:58.08 | furrywolf | a 1TB 30DWPD SSD, that is |
| 00:58.53 | DocScrutinizer05 | so their performace is about same, aiui |
| 00:59.08 | furrywolf | SSD write endurance is not that bad if you get a decent model, and hdds are so much slower that the write endurance is limited by the mechanical life expectancy, to about the same range. |
| 00:59.23 | DocScrutinizer05 | except that the SSD could possibly do faster and lofe less than 5 years |
| 00:59.36 | DocScrutinizer05 | live* |
| 00:59.59 | DocScrutinizer05 | yes, seems about right |
| 01:00.16 | furrywolf | yes, it could. but a hdd wouldn't be an option for that much data anyway, given how slow it is. |
| 01:00.29 | DocScrutinizer05 | of course |
| 01:00.55 | furrywolf | compared to a drive can only write 13TB/day, you'd be using the SSD at less than its rated drive writes per day, so it'd last longer than 5 years, not shorter. :) |
| 01:01.43 | furrywolf | the only reason to use spinning drives is cost, and I suspect that reason will go away soon enough, and they'll be dead and gone. |
| 01:02.40 | DocScrutinizer05 | yep |
| 01:03.11 | DocScrutinizer05 | didn't know SSD were that tough on writing meanwhile |
| 01:03.48 | DocScrutinizer05 | amazing |
| 01:05.21 | DocScrutinizer05 | I recall times where flash was rated at a few k write cycles max |
| 01:05.26 | furrywolf | I just glanced at the top of the list at http://www.storagesearch.com/dwpd.html , figuring that if you're doing an application where write endurance is an important factor, you could select a model from the top half of the list, and not from the bottom half of the list. |
| 01:07.15 | *** join/#devuan cyteen (~cyteen@249.158.93.209.dyn.plus.net) |
| 01:29.14 | *** join/#devuan rrq (~ralph@60-242-139-200.static.tpgi.com.au) |
| 01:29.14 | *** join/#devuan OmegaPhil (~kvirc@fsf/member/OmegaPhil) |
| 01:39.37 | timeless | hey! |
| 01:39.42 | timeless | i got an email from peach! |
| 01:39.55 | timeless | <PROTECTED> |
| 01:40.20 | timeless | um |
| 01:40.23 | timeless | not happy |
| 01:40.34 | timeless | it's still using the wrong stupid disk https://www.irccloud.com/pastebin/j4OT781e/ |
| 01:41.18 | timeless | again, the lsscsi output is not remotely correct https://www.irccloud.com/pastebin/lmhOPcGY/ |
| 01:41.50 | timeless | frowns |
| 01:41.58 | timeless | anyway, let's ignore lsscsi |
| 01:42.10 | timeless | afaict i lost ~4 hours letting this stupid thing try to rebuild |
| 01:42.14 | timeless | and i'm back where i started |
| 01:42.20 | timeless | with the stupid wrong disk alive |
| 01:42.28 | timeless | DocScrutinizer05: ? |
| 01:44.36 | Centurion_Dan | that would becuase when you did the --add instead of --re-add it rewrote the metadata on that disk so the existing data is considered invalid. |
| 01:44.48 | Centurion_Dan | where did it rebuild to/from? |
| 01:44.50 | DocScrutinizer05 | you removed a disk and added a spare, so that's the result |
| 01:44.57 | timeless | ... |
| 01:46.23 | timeless | ok. forget what i did. |
| 01:46.31 | timeless | i'm open to precise commands that are likely to do something remotely useful |
| 01:46.42 | timeless | otherwise, i'm going to shut down the stupid box, and pull the plug on sda |
| 01:46.54 | timeless | if i can figure out which stupid volume has the serial matching the one that linux claims sda has |
| 01:47.01 | DocScrutinizer05 | that won't do anything useful, so much is certain |
| 01:47.11 | timeless | why not? |
| 01:47.22 | timeless | clearly sda is more of a problem than sdf |
| 01:47.32 | DocScrutinizer05 | why would pulling a cable fix anything? |
| 01:47.41 | timeless | because i won't have sda trying to be the master? |
| 01:48.13 | timeless | and worst case i can pull up a keyboard and do some stupid mdadm command to build md1 out of sdf and missing |
| 01:48.30 | timeless | grr |
| 01:48.43 | timeless | worse the stupid volume /backups is mounted rw |
| 01:49.14 | timeless | which means that technically /backups and the whole volume can get more out of sync |
| 01:49.18 | timeless | mounts it ro |
| 01:49.35 | DocScrutinizer05 | huh? |
| 01:49.38 | timeless | if sda is in bad shape and can't properly mirror in sdf |
| 01:50.15 | Centurion_Dan | DocScrutinizer05: /dev/disk/by-path doesn't give you any detail |
| 01:50.23 | timeless | then can't i make md1 out of sdf, and later ask it to add sda and let it poke sda until sda gives it available blocks? |
| 01:50.24 | DocScrutinizer05 | indeed |
| 01:50.51 | timeless | yeah, i had already looked at /dev/disk/by-path -- it's useless |
| 01:51.16 | timeless | really there doesn't seem to be a way in linux to ask/answer that question, which seems bizarre |
| 01:51.22 | timeless | since at some level the system has to know |
| 01:51.22 | DocScrutinizer05 | timeless: I'm not sure your concept of how a raid works is correct |
| 01:51.30 | timeless | i'm sure it isn't |
| 01:51.58 | DocScrutinizer05 | I'm totally lost on making sense of your last question |
| 01:52.31 | Centurion_Dan | timeless: you have one of those mobo's with a 4port + 2port sata chip - probably one of the 2ports exposed as an e-sata port. |
| 01:52.31 | timeless | hrm... ok, dumb question |
| 01:52.57 | timeless | in raid1, are identical sectors used for mirroring? |
| 01:53.07 | Centurion_Dan | if that is the case ATA8 is likely port 5 |
| 01:53.08 | DocScrutinizer05 | how would md poke sda when that is affed later? |
| 01:53.21 | Centurion_Dan | timeless: yes |
| 01:53.28 | Centurion_Dan | raid1 is mirror |
| 01:53.33 | DocScrutinizer05 | please forget about all that cable topic, it's really not what you need |
| 01:53.57 | timeless | ok... so effectively even if a given sector is good in sdf, the raid needs to treat it as unusable if it isn't usable in sda |
| 01:53.58 | DocScrutinizer05 | aka "totally irrelevant right now" |
| 01:54.20 | timeless | is really considering <drop-file-system> |
| 01:54.53 | DocScrutinizer05 | ((treat as unusable)) yes, for writing |
| 01:55.01 | DocScrutinizer05 | NOT for reading |
| 01:55.32 | DocScrutinizer05 | master vs slave is only a question of which drive gets written to *first*, afaik |
| 01:55.33 | Centurion_Dan | but the problem is that when you did the --add command right at the start of the process, that tells mdadm to over-write it's superblock and say that the data on /dev/sdf is crap and it all needs to be copied back from /dev/sda |
| 01:55.58 | timeless | Centurion_Dan: ok, so, how do i undo that? |
| 01:56.11 | timeless | i want it to think that sda is crap and that sdf is charming |
| 01:56.32 | DocScrutinizer05 | you should build a new clean md4 with sda = 'master' and sdf secondary |
| 01:56.45 | Centurion_Dan | if you had done --re-add, it would most likely have re-added /dev/sdf7 no problem |
| 01:57.37 | DocScrutinizer05 | however not that sdf is pretty much nukes aiui, it's prolly not wise to try "undo" anything |
| 01:57.49 | DocScrutinizer05 | s/ not / now / |
| 01:58.30 | Centurion_Dan | to recover from this you need to unmount the filesystem drop /dev/sda and then re-create the md volume using /dev/sdf7 and "missing" and --assume-clean |
| 01:58.57 | DocScrutinizer05 | wait |
| 01:59.32 | DocScrutinizer05 | aiui now sda is the only drive that has (minimally corrupted due to bad blocks) genuine data |
| 02:01.35 | DocScrutinizer05 | you should have just shut down the system when cable came off, then on reboot bot disks get synced again, and then you shut down again and swap sda for a good drive, and disks again get synced on boot |
| 02:01.44 | timeless | ok |
| 02:01.45 | timeless | stop |
| 02:01.51 | Centurion_Dan | DocScrutinizer05: not so. Only the md superblock on /dev/sdf7 is stuffed. Assuming he's using a journalling filesystem, it's likely to be recoverable or clean provided unless it was writing at the time it got unplugged. |
| 02:01.52 | timeless | i do not give a **** about what i should have done |
| 02:01.55 | timeless | i can't go back in time |
| 02:02.15 | timeless | the only thing remotely helpful is a list of steps from here forward |
| 02:02.29 | Centurion_Dan | timeless: I know it's moot at this point in time.. |
| 02:02.40 | timeless | doc seems to be missing that |
| 02:02.46 | DocScrutinizer05 | sorry, I can't suggest steps forward without clear idea of the past and the 'now' |
| 02:02.52 | timeless | Centurion_Dan: the next layers on top of this are VG/LV |
| 02:03.01 | timeless | the layer on top of that is ext3 |
| 02:03.17 | timeless | DocScrutinizer05: i have no clue what the system was doing when the cable went away |
| 02:03.17 | Centurion_Dan | does it have the root filesystem on that? |
| 02:03.30 | DocScrutinizer05 | timeless: that's not relevant |
| 02:03.47 | timeless | Centurion_Dan: it's actually hard to tell? |
| 02:03.50 | DocScrutinizer05 | the md anyway was either syncing or writing or just reading |
| 02:04.03 | timeless | https://www.irccloud.com/pastebin/JkHPrPNt/df |
| 02:04.09 | DocScrutinizer05 | most likely the latter |
| 02:04.25 | timeless | https://www.irccloud.com/pastebin/zumSZykv/lvdisplay |
| 02:05.03 | timeless | afaict the areas that are unhappy are in the backups portion of the LV, but technically both Backups and Root are all backed by the same Volume Group |
| 02:05.14 | DocScrutinizer05 | Centurion_Dan: when he removes sdf and then re-adds it, what would be the result? |
| 02:05.30 | timeless | nothing |
| 02:05.39 | timeless | it tries to sync, fails, makes it a spare |
| 02:05.42 | timeless | i've done that 4 times now |
| 02:05.46 | timeless | total waste of my time |
| 02:05.59 | timeless | probably the only redeeming quality here is that the root file system changes much less frequently than the backups filesystem |
| 02:06.12 | timeless | and thus there shouldn't have been any change to the portion that holds the rootfs |
| 02:06.41 | DocScrutinizer05 | sorry i'm out. I think you need better advice than I can provide, by somebody more willing to deal with this |
| 02:08.01 | timeless | sorry... i appreciate your effort |
| 02:08.11 | timeless | i do... it's just, being told "you shouldn't have done x" doesn't help |
| 02:08.25 | timeless | especially when someone is frustrated, and starving, and hates touching the thing in the first place... |
| 02:08.54 | timeless | i know that the system shouldn't be using raid the way it does, that's pretty obvious .... |
| 02:09.19 | timeless | i'm still not sure if i should be wasting a single minute trying to fix it, or if i should have just procured an AWS instance and provisioned it and moved my backups over |
| 02:11.32 | Centurion_Dan | timeless: moving to AWS depends on how pedantic you are and whether the performance characteristics you need are within the bounds of what your willing to pay. |
| 02:12.03 | timeless | well, this system specifically is a backup system, ideally it's write once, read never |
| 02:13.30 | Centurion_Dan | What I would do is see if the state of sdf7 is good. You can do this by removing it from /dev/md1 for a start and then spin up another md volume to have a look |
| 02:13.57 | timeless | mdadm: hot removed /dev/sdf7 from /dev/md1 |
| 02:14.24 | timeless | um, keeping in mind that it doesn't have a real file system on it... |
| 02:15.36 | Centurion_Dan | Now before you try this lets plan it first: |
| 02:15.36 | Centurion_Dan | 1) `mdadm --zero-superblock /dev/sdf7` this wipes the superblock |
| 02:16.07 | timeless | mdadm --create /dev/md29 --level 1 --raid-devices=2 /dev/sdf7 missing |
| 02:16.49 | timeless | wouldn't zeroing the superblock make the partition effectively useless? |
| 02:17.11 | Centurion_Dan | but you need to add the flag --assume-clean otherwise it will zero it out the volume |
| 02:18.14 | Centurion_Dan | no - because it's only zero'ing the metadata |
| 02:18.53 | Centurion_Dan | in the superblock |
| 02:19.02 | timeless | am i relying on a backup superblock to get me out of hell, or is something going to rebuild the metadata from the rest of the data? |
| 02:20.42 | Centurion_Dan | we're effectively building a new md volume without disturbing the data you want to keep - a new superblock, same data. |
| 02:20.54 | timeless | ok |
| 02:21.02 | timeless | talk me through the steps :) |
| 02:21.26 | DocScrutinizer05 | finally |
| 02:22.29 | Centurion_Dan | so step 2 is `mdadm --create /dev/md29 --level 1 --assume-clean --raid-devices=2 /dev/sdf7 missing` |
| 02:22.57 | timeless | ok |
| 02:23.40 | Centurion_Dan | step 3 is to power off, identify (by serial number) and unplug /dev/sda - this is our fall back so we don't mess with it. |
| 02:23.56 | timeless | ok |
| 02:23.58 | timeless | that step is painful |
| 02:24.05 | timeless | and requires a screw driver (I have) |
| 02:24.18 | timeless | since i will be undocking each drive until i find the right stupid drive |
| 02:24.29 | timeless | but, that step had to happen anyway |
| 02:24.31 | timeless | next? |
| 02:26.44 | Centurion_Dan | actually before step3 that we need to change /etc/mdadm.conf to add the volume with the new identifier and run update-initramfs ... otherwise it won't assemble the volume. |
| 02:27.26 | timeless | hrm |
| 02:27.43 | timeless | is the uuid going to be the same for 29? |
| 02:28.09 | timeless | ARRAY /dev/md/1 metadata=1.2 UUID=44d6595a:a8a78dc5:ff572b88:b6e7067b name=peach:1 |
| 02:28.44 | timeless | really expects this adventure to result in a shiny brick |
| 02:28.52 | Centurion_Dan | If the hdd is in a caddy with a drive activity light per disk you can run hdparm -t /dev/sda (assuming you have hdparm installed) and see what light goes on solid for a couple of seconds |
| 02:29.15 | timeless | i know about hdparm (it's installed, i've used it) |
| 02:29.24 | timeless | i don't think anything has activity lights |
| 02:29.39 | timeless | so, i think dd if=/dev/sda of=/dev/null is probably my friend |
| 02:30.03 | Centurion_Dan | it will be a different UUID. easiest way is to do: |
| 02:30.21 | DocScrutinizer05 | pr you use a cable stub with jack and LED and plug it to HDDs one at a time, and then use same blink command |
| 02:30.37 | timeless | "cable stub"?? |
| 02:31.00 | DocScrutinizer05 | post jack, 19cm 2 wire cable, LED |
| 02:31.11 | DocScrutinizer05 | 10cm, but whatever |
| 02:31.35 | DocScrutinizer05 | each drive has a few post connectors to e.g. connect LEDs |
| 02:31.49 | DocScrutinizer05 | virtually each one |
| 02:32.02 | Centurion_Dan | nah, I use backplanes that have a led per drive ;-) |
| 02:32.08 | DocScrutinizer05 | often even labeled "LED" |
| 02:32.23 | DocScrutinizer05 | Centurion_Dan: those are connected exactly same way ;-) |
| 02:32.44 | DocScrutinizer05 | unless they are connected to MoBo/controller |
| 02:34.11 | Centurion_Dan | yup.. |
| 02:34.20 | DocScrutinizer05 | dd if=/dev/sda of=/dev/null is a nice idea but fauils due to sequential read, so not many head positioning actions |
| 02:34.32 | Centurion_Dan | re mdadm.conf: |
| 02:36.03 | Centurion_Dan | mdadm -E -s /dev/sdf7` >> /etc/mdadm/mdadm.conf |
| 02:36.29 | DocScrutinizer05 | hdparm -Y Put drive to sleep might prove useful, particularly when accessing drive and watching for spin-up vibration |
| 02:36.33 | *** join/#devuan Humpelstilzchen (erik@x4e366d70.dyn.telefonica.de) |
| 02:37.13 | *** join/#devuan tsadok (~nathan@cpe-75-185-35-110.columbus.res.rr.com) |
| 02:37.23 | Centurion_Dan | or just run `mdadm -E -s /dev/sdf7` and paste that line at the bottom of /etc/mdadm/mdadm.conf |
| 02:39.13 | timeless | Centurion_Dan: so, how does this interact w/ the lvm stuff |
| 02:39.14 | Centurion_Dan | DocScrutinizer05: can we stay on task please ;-) |
| 02:39.24 | timeless | because there isn't a file system directly on top of this junk |
| 02:40.08 | timeless | if i take away sda (and thus md1), the volume group will be unhappy |
| 02:40.18 | timeless | https://www.irccloud.com/pastebin/xw1qpoxQ/pvdisplay |
| 02:40.59 | Centurion_Dan | so the steps are: |
| 02:41.18 | Centurion_Dan | 1) `mdadm --zero-superblock /dev/sdf7` this wipes the superblock |
| 02:41.23 | Centurion_Dan | 2) `mdadm --create /dev/md29 --level 1 --assume-clean --raid-devices=2 /dev/sdf7 missing` |
| 02:42.07 | Centurion_Dan | 3) `mdadm -E -s /dev/sdf7` >> /etc/mdadm/mdadm.conf` or `mdadm -E -s /dev/sdf7` and paste that line at the bottom of /etc/mdadm/mdadm.conf |
| 02:42.42 | Centurion_Dan | 4) power off and unplug /dev/sda |
| 02:43.01 | Centurion_Dan | 5) power up. |
| 02:43.12 | timeless | you left off the command to update some boot process thing |
| 02:43.40 | timeless | and it won't boot because the VG will be missing md1 which is part of what is needed for the LV which is needed for everything |
| 02:44.02 | timeless | https://www.irccloud.com/pastebin/8rjVHvm0/lvdisplay |
| 02:44.20 | Centurion_Dan | because /dev/sdf7 was a mirror of the physical volume that is used for the volume group, LVM won't notice. |
| 02:44.22 | timeless | (well, i could be wrong about whether or not it will boot, i'm speculating) |
| 02:44.29 | timeless | really? |
| 02:44.42 | timeless | so it doesn't care about the mdadm uuids? those are just for mdadm? cute |
| 02:44.43 | timeless | ok |
| 02:44.53 | Centurion_Dan | but correct step: 3b) |
| 02:44.54 | timeless | what about update-initramfs? |
| 02:45.01 | Centurion_Dan | update-initramfs |
| 02:45.25 | timeless | alright, so, let's figure out the best way for me to do step 4b (unplug /dev/sda) |
| 02:46.41 | Centurion_Dan | it might complain bitterly about /dev/md1 not being able to be assembled, but I don't think it will halt the boot unless the root fs can't be found. |
| 02:46.54 | DocScrutinizer05 | if I may dare: you can get serial number of sda easily, you prolly also can see serial number in BIOS setup |
| 02:47.25 | DocScrutinizer05 | unless you unplugged the HDD in which case you won't see it in BIOS |
| 02:47.37 | Centurion_Dan | and it will be printed on the drives label |
| 02:47.49 | timeless | yeah, the problem is that they're in a bay |
| 02:48.01 | Centurion_Dan | smartctl -i /dev/sda |
| 02:48.03 | DocScrutinizer05 | alas label is often hard to read unless you unmount the drive |
| 02:48.10 | timeless | effectively something like http://usb.brando.com/prod_img/zoom/UHDCL001700_01_L.jpg |
| 02:48.40 | timeless | (not actually, it's just a bay in a normal tower case, but the point is that they're stacked one on top of the next, and each is screwed iin) |
| 02:49.12 | *** join/#devuan Hoshpak (~Hoshpak@p2003005B4B1A7F0098A424399D4D9150.dip0.t-ipconnect.de) |
| 02:49.48 | DocScrutinizer05 | http://wstaw.org/m/2017/01/18/plasma-desktopr17764.png LED posts |
| 02:50.10 | timeless | So, bios is tempting, but... |
| 02:50.45 | timeless | could i use a multimeter on the pins? |
| 02:50.53 | DocScrutinizer05 | sure |
| 02:51.25 | timeless | ok, that sounds workable |
| 02:51.31 | DocScrutinizer05 | just check the label or drive manual first, to actually use the right pins |
| 02:51.42 | DocScrutinizer05 | silkscreen label |
| 02:51.59 | DocScrutinizer05 | or whatever, usually there's something |
| 02:52.31 | DocScrutinizer05 | out again |
| 02:52.38 | Centurion_Dan | based on your output of lsscsi -v I'd say it will be the lowest numbered sata port. |
| 02:53.21 | Centurion_Dan | another question too is do you have a backup boot drive setup - assuming your using grub... |
| 02:54.25 | timeless | grew up w/ lilo :( |
| 02:54.31 | timeless | so, i have /etc/grub.d/ |
| 02:54.49 | timeless | i suppose i can look in /boot/grub |
| 02:55.39 | Centurion_Dan | is your /boot on raid? |
| 02:56.03 | Centurion_Dan | and what's the output of `df` |
| 02:56.23 | timeless | menuentry 'Debian GNU/Linux, with Linux 3.2.0-4-686-pae' --class debian --class gnu-linux --class gnu --class os { |
| 02:56.23 | timeless | menuentry 'Debian GNU/Linux, with Linux 3.2.0-4-686-pae (recovery mode)' --class debian --class gnu-linux --class gnu --class os { |
| 02:56.35 | timeless | um |
| 02:56.56 | DocScrutinizer05 | boot CD with a live system? |
| 02:58.18 | DocScrutinizer05 | would be so much easier |
| 03:00.20 | timeless | <PROTECTED> |
| 03:00.20 | timeless | frowns |
| 03:00.20 | timeless | the dvd drive isn't plugged in |
| 03:00.20 | timeless | and i don't really want to figure out how to burn something for it |
| 03:00.20 | timeless | bad news |
| 03:00.22 | timeless | sdf1 doesn't have an ext3 volume |
| 03:00.22 | timeless | shooting this computer would be easier |
| 03:01.04 | timeless | so, i don't think i have a /boot mirror of sda1 https://www.irccloud.com/pastebin/ZiRZe2dY/ |
| 03:01.22 | timeless | i.e. if i take sda out, the system won't boot... |
| 03:02.50 | *** join/#devuan tsadok (~nathan@cpe-75-185-35-110.columbus.res.rr.com) |
| 03:03.42 | Centurion_Dan | what is the output of `df` |
| 03:04.01 | timeless | https://www.irccloud.com/pastebin/EMPwSWBk/df |
| 03:04.26 | timeless | looking at volumes in sda/sdf by paired https://www.irccloud.com/pastebin/xLRYmIA5/ |
| 03:05.54 | timeless | https://www.irccloud.com/pastebin/clKqOsIT/fdisk |
| 03:06.09 | Centurion_Dan | can you do the outputs of `fdisk -l /dev/sda` and `fdisk -l /dev/sdf` |
| 03:06.14 | timeless | ^^ |
| 03:06.53 | timeless | i think i can do a dd if=/dev/sda1 of=/dev/sdf1 to get somewhere |
| 03:07.13 | timeless | or i could format sdf1 and rsync or similar |
| 03:07.42 | *** join/#devuan tsadok (~nathan@cpe-75-185-35-110.columbus.res.rr.com) |
| 03:08.08 | Centurion_Dan | anything in /dev/sdf1? |
| 03:08.55 | timeless | looks like lots of \0's |
| 03:09.20 | timeless | (which is different from what /dev/sda1 looks like) |
| 03:09.30 | Centurion_Dan | I'd shrink the /boot filesystem a little first as /dev/sdf1 is marginally smaller. |
| 03:09.56 | *** join/#devuan kelsoo1 (~kelsoo@dragora/developer/kelsoo) |
| 03:09.58 | timeless | indeed |
| 03:09.59 | timeless | sighs |
| 03:10.17 | timeless | gparted? |
| 03:11.03 | Centurion_Dan | resize2fs |
| 03:11.20 | Centurion_Dan | you can safely unmount it |
| 03:11.38 | Centurion_Dan | are there other volumes on /dev/sda you need to keep? |
| 03:11.48 | Centurion_Dan | partitions... |
| 03:11.59 | timeless | obviously i need everything to survive :) |
| 03:12.08 | timeless | but the system only has 3 things at the end of the day afaict |
| 03:12.41 | Centurion_Dan | Are they mirrored elsewhere? |
| 03:12.43 | timeless | presumably i have to use tune2fs to disable the ext3 journal? |
| 03:12.54 | timeless | there are 6 physical disks |
| 03:13.02 | timeless | generally each disk has a pair |
| 03:13.25 | timeless | each disk also has lots of stupid partitions which are generally mirrored to the equivalent partition in the paired drive |
| 03:13.41 | timeless | there are 3 mount points (/boot, /, /backups) |
| 03:15.20 | timeless | umount; fsck; tune - remove journal; fsck https://www.irccloud.com/pastebin/UM8HtFpW/ |
| 03:16.29 | timeless | ok, so, next... doing math to get the right size? |
| 03:18.06 | *** join/#devuan bpmedley (~bpm@rrcs-24-199-194-123.midsouth.biz.rr.com) |
| 03:18.28 | timeless | -> resize2fs /dev/sda1 2103491s |
| 03:18.32 | timeless | does that seem right? |
| 03:19.00 | Centurion_Dan | timeless: nope. just umount /boot, do a fsck on it and then reduce it's size by like 10% |
| 03:19.50 | timeless | so 1.8G ? |
| 03:19.56 | Centurion_Dan | so /boot is 1GB, reduce it to 800MB. |
| 03:20.15 | timeless | <PROTECTED> |
| 03:20.30 | timeless | admittedly, it's only using ~130mb... |
| 03:20.49 | timeless | I could reduce it to 512M ... |
| 03:20.52 | Centurion_Dan | don't be fooled by fdisk, as it reports in 512k blocks |
| 03:21.04 | Centurion_Dan | that would do too. |
| 03:21.05 | timeless | this was from df https://www.irccloud.com/pastebin/BUU6bBW6/ |
| 03:21.23 | timeless | ok, so... |
| 03:21.32 | timeless | -> resize2fs /dev/sda1 512M |
| 03:22.00 | timeless | waits for confirm |
| 03:22.11 | Centurion_Dan | yup |
| 03:22.32 | timeless | ok https://www.irccloud.com/pastebin/tMCR7c9N/ |
| 03:22.56 | timeless | hrm, that of course doesn't impact fdisk |
| 03:24.03 | timeless | ok, it mounts as ext2 |
| 03:24.05 | Centurion_Dan | then we might as well make /dev/sdf1 a raid volume as well while we're at it |
| 03:24.25 | timeless | given how unfriendly raid is, i'd like to hear justification before i do that... |
| 03:25.00 | timeless | i mean, i understand the idea of wanting to be able to swap out sda and swap in a replacement w/o having to manually mirror |
| 03:25.06 | Centurion_Dan | it's not unfriendly, just needs appropriate care and attention ;-) |
| 03:25.12 | timeless | but... tbh, manually mirroring is easier! |
| 03:25.43 | timeless | ok... |
| 03:26.44 | Centurion_Dan | mdadm --create /dev/md30 --level 1 --raid-devices=2 /dev/sdf1 missing |
| 03:27.12 | timeless | sdf1 being the unformatted thing? |
| 03:27.25 | Centurion_Dan | mdadm -E -s /dev/sdf1 >> /etc/mdadm/mdadm.conf |
| 03:27.31 | Centurion_Dan | yup... |
| 03:27.54 | timeless | wow https://www.irccloud.com/pastebin/xUIpO7tU/ |
| 03:27.54 | Centurion_Dan | actually take a backup of /etc/mdadm/mdadm.conf before doing that: |
| 03:28.22 | timeless | is amused |
| 03:28.36 | timeless | this is the first sign of intelligence i've seen from mdadm :o |
| 03:28.58 | *** join/#devuan underd0g (~underd0g@93.140.28.15) |
| 03:29.09 | Centurion_Dan | your running grub2 right? |
| 03:29.26 | Centurion_Dan | what release is this but the way? wheezy, jessie? |
| 03:29.36 | timeless | ii grub 0.97-64 i386 GRand Unified Bootloader (dummy package) |
| 03:29.46 | timeless | root@peach:/# cat /etc/debian_version |
| 03:29.46 | timeless | 7.6 |
| 03:30.20 | timeless | decides to be terrified about the idea of updating grub from 0.97 to 2 |
| 03:30.33 | timeless | ii grub2-common 1.99-27+deb7u2 i386 GRand Unified Bootloader (common files for version 2) |
| 03:30.39 | timeless | ok, so i guess i have grub2 |
| 03:30.45 | timeless | and a thing that lies about grub |
| 03:30.59 | timeless | ii grub-pc 1.99-27+deb7u2 i386 GRand Unified Bootloader, version 2 (PC/BIOS version) |
| 03:30.59 | timeless | ii grub-pc-bin 1.99-27+deb7u2 i386 GRand Unified Bootloader, version 2 (PC/BIOS binaries) |
| 03:31.31 | Centurion_Dan | that should be fine then. |
| 03:31.33 | timeless | ok, so it's actually grub2 |
| 03:32.01 | timeless | mdadm: Defaulting to version 1.2 metadata |
| 03:32.01 | timeless | mdadm: array /dev/md30 started. |
| 03:32.27 | timeless | am i backing up the mdadm conf to / /boot or another computer? |
| 03:32.58 | timeless | starts with "cloud backup" https://www.irccloud.com/pastebin/2Nbfj7f8/peach-mdadm.conf |
| 03:33.51 | timeless | turns out, there's already a backup from july |
| 03:35.03 | Centurion_Dan | ok.. you'll need a copy of the new one as these changes with the new volumes won't be there on the other side. |
| 03:35.13 | timeless | ok, added /dev/md/30 to mdadm.conf https://www.irccloud.com/pastebin/rVQkKkeR/ |
| 03:36.16 | timeless | cloud backup :) |
| 03:36.58 | timeless | ok, but seriously... i can copy it into /boot if that makes sense |
| 03:37.43 | timeless | alright, so, presumably the next step is to try to convince mdadm to take sda1 and incorporate it into md30 |
| 03:37.56 | Centurion_Dan | yeah do that too, but after steps 1 & 2. |
| 03:40.25 | timeless | ok... https://www.irccloud.com/pastebin/PKwTZ1Z6/ |
| 03:40.26 | Centurion_Dan | just umount /boot and `dd if=/dev/sda1 of=/dev/md30 bs=1M count=515` |
| 03:41.21 | timeless | md29 created https://www.irccloud.com/pastebin/2Bv9fduE/ |
| 03:42.22 | timeless | am i doing `update-initramfs` before or after the umount /boot ? |
| 03:43.42 | Centurion_Dan | doesn't matter we'll check and mount the new /boot (/dev/md30 and do it there |
| 03:43.49 | timeless | ok |
| 03:44.07 | Centurion_Dan | do it after, that way /dev/sda1 is less messed up too. |
| 03:44.23 | timeless | md30 initialized https://www.irccloud.com/pastebin/ZwBGvtms/ |
| 03:44.41 | timeless | so, now, presumably i plead w/ md to let me mount /dev/md30 as /boot ? |
| 03:46.21 | Centurion_Dan | you don't need to plead with md at all. `mount /dev/md30 /boot` should do it. perhaps fsck /dev/md30 first to be sure. ;-) |
| 03:46.37 | timeless | clean https://www.irccloud.com/pastebin/iTKX26JB/ |
| 03:46.54 | timeless | root@peach:/etc/mdadm# mount /dev/md30 /boot/ |
| 03:47.09 | timeless | next presumably i change /etc/fstab |
| 03:47.16 | Centurion_Dan | yup. |
| 03:47.40 | timeless | is the uuid from mdadm.conf compatible w/ the uuid that fstab wants? |
| 03:47.43 | timeless | it doesn't look like it |
| 03:48.27 | timeless | ooh |
| 03:48.32 | timeless | i don't need to do anything here |
| 03:48.47 | timeless | because /dev/md30 is a mirror of /dev/sda1 so it has the same uuid |
| 03:49.45 | timeless | unchanged https://www.irccloud.com/pastebin/oMg4WH3c/fstab |
| 03:50.03 | timeless | ok, so... next.... update grub? |
| 03:50.10 | timeless | err `update-initramfs` |
| 03:50.12 | Centurion_Dan | no you want the UUID from `e2fstune -l | grep UUID` |
| 03:50.40 | timeless | not this one? https://www.irccloud.com/pastebin/dZGDJh6a/ |
| 03:50.41 | Centurion_Dan | yup |
| 03:50.54 | timeless | doesn't have e2fstune |
| 03:50.59 | timeless | has tune2fs |
| 03:51.19 | Centurion_Dan | yeah that one... :-] |
| 03:51.26 | furrywolf | has tunafish |
| 03:51.40 | timeless | still the same as the other one https://www.irccloud.com/pastebin/VMcx7tj8/ |
| 03:51.57 | Centurion_Dan | yup so it should be |
| 03:52.00 | timeless | perfect mirrors are perfect |
| 03:52.13 | Centurion_Dan | ;-) |
| 03:52.22 | timeless | ok, so, fstab unchanged |
| 03:52.29 | timeless | now... update-initramfs? |
| 03:52.48 | Centurion_Dan | so now we need to also setup grub |
| 03:53.45 | Centurion_Dan | `update-grub` and then `grub-install /dev/sdf` |
| 03:54.27 | timeless | update-grub is not terribly happy https://www.irccloud.com/pastebin/fSzgJnXe/ |
| 03:55.58 | timeless | ok... so... |
| 03:56.03 | timeless | <PROTECTED> |
| 03:56.05 | Centurion_Dan | just a bit pissed cause it can see 2 versions of the same pv ... not an issue. Whats in /boot/grub/devices? |
| 03:56.37 | timeless | there's no devices, but there is: https://www.irccloud.com/pastebin/ANLGdwJo/device.map |
| 03:57.55 | Centurion_Dan | ah, you need to do grub-mkdevicemap |
| 03:57.57 | timeless | for references my available devices: https://www.irccloud.com/pastebin/0LMXIU4I/ |
| 03:58.38 | timeless | ok, that now more or less matches available (order differs, but hey) https://www.irccloud.com/pastebin/An1gXHQn/ |
| 03:59.04 | timeless | update-grub is slightly less unhappy https://www.irccloud.com/pastebin/f4dtwNla/ |
| 04:00.58 | Centurion_Dan | what does /boot/grub/grub.cfg look like now? |
| 04:01.24 | timeless | "stuff" |
| 04:01.47 | timeless | https://www.irccloud.com/pastebin/czRKTHi1/grub.cfg |
| 04:02.53 | timeless | by which i mean, it's more or less templaty goop that we haven't impacted afaict |
| 04:03.25 | Centurion_Dan | looks great, it sees that it needs mdraid1x module installed |
| 04:04.01 | *** join/#devuan rleigh (~rleigh@unaffiliated/rleigh) |
| 04:04.14 | Centurion_Dan | now you can do grub-install /dev/sdf |
| 04:04.36 | Centurion_Dan | shouldn't say much.. |
| 04:04.57 | timeless | not much... https://www.irccloud.com/pastebin/h29lQjkg/ |
| 04:05.31 | timeless | on the bright side... at least this system has grub2 |
| 04:05.53 | timeless | is pretty sure one of these other systems isn't using it |
| 04:06.07 | timeless | hrm |
| 04:06.18 | timeless | or maybe grub changed how it organized itself later? |
| 04:08.04 | Centurion_Dan | so we're ready for the power off and pull /dev/sda1. I'd say it's in port 0 (or 1 - whatever the lowest number is |
| 04:08.24 | timeless | alright... this should be exciting |
| 04:08.34 | timeless | says goodbye to peach |
| 04:08.52 | Centurion_Dan | you probably want to swap in the cable for dev sdf anyway so you don't have to piss with the bios to change the first boot disk. |
| 04:13.20 | *** join/#devuan ppisati (~ppisati@2-235-152-247.ip228.fastwebnet.it) |
| 04:13.40 | timeless | no boot devices available :o |
| 04:14.22 | Centurion_Dan | so not even grub comes up?? |
| 04:14.52 | timeless | picked the wrong drive |
| 04:15.13 | timeless | tried again |
| 04:15.15 | timeless | system booted |
| 04:15.39 | Centurion_Dan | and ?? |
| 04:15.43 | timeless | mdstat https://www.irccloud.com/pastebin/HOan6tyv/ |
| 04:16.58 | Centurion_Dan | looks good?? |
| 04:17.03 | timeless | yeah |
| 04:17.09 | timeless | ok.... so... |
| 04:17.26 | Centurion_Dan | hi fives timeless |
| 04:17.31 | timeless | thanks Centurion_Dan ! |
| 04:17.43 | timeless | but... what about the drive that used to be known as sda1 |
| 04:17.45 | timeless | err sda |
| 04:17.58 | timeless | can i put it back in and try to get something out of it, or is it trash? |
| 04:18.33 | Centurion_Dan | was there anything there that you need to recover still or was it all mirrored elsewhere? |
| 04:18.52 | timeless | in theory sda and sdf were perfectly mirrored until the mirror broke |
| 04:19.04 | timeless | the data will start pouring in once i mark the /backups volume rw |
| 04:19.16 | Centurion_Dan | and all the volumes are there? |
| 04:19.36 | golinux | I'm exhausted just following this marathon! |
| 04:19.38 | timeless | there are only 3 volumes /boot, /, /backups |
| 04:19.53 | timeless | and /backups has stuff, which will be replaced once it's allowed to |
| 04:19.54 | Centurion_Dan | ok and /backups are back? |
| 04:20.28 | Centurion_Dan | golinux: I know... there goes an afternoon hacking due to fixing this :-/ |
| 04:20.34 | timeless | yeah, there's a backup of one of the other fruits from the 12th |
| 04:20.57 | timeless | i owe you at least a beer... |
| 04:21.11 | timeless | but i'm still trying to figure out what to do about this system |
| 04:21.44 | timeless | do i buy a 5/6TB drive and try to copy all the data over and decomission all the drives i have, and then mirror that onto a second 5/6TB drive? |
| 04:21.51 | timeless | do I try to reintroduce the failing drive? |
| 04:21.57 | Centurion_Dan | get a new drive and rebuild the mirror while you decide. |
| 04:22.07 | Centurion_Dan | bin that dying drive!!! |
| 04:22.08 | timeless | do I just leave it and start working on getting an AWS replacement |
| 04:22.35 | timeless | ok, so... i guess i need to deal w/ `auto-read-only` |
| 04:22.48 | Centurion_Dan | that depends on your use-case |
| 04:23.22 | Centurion_Dan | just touch a file in that volume an it will go to readwrite ;-) |
| 04:23.37 | timeless | eh?? what-the-heck? |
| 04:24.19 | Centurion_Dan | auto-read-only is the state md will leave volumes in until something writes to them. |
| 04:25.02 | timeless | is it documented anywhere? and is there a reason for it? |
| 04:25.07 | timeless | (obviously there's a reason) |
| 04:25.38 | Centurion_Dan | it's quite normal behaviour by md |
| 04:26.53 | Centurion_Dan | anyway I have to go and cook dinner for the family ;-) |
| 04:27.07 | timeless | it's nearly midnight |
| 04:27.14 | timeless | thanks for the help |
| 04:27.24 | Centurion_Dan | your welcome... |
| 05:02.04 | *** join/#devuan DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 05:37.21 | *** join/#devuan rrq (~ralph@60-242-139-200.static.tpgi.com.au) |
| 05:59.56 | *** join/#devuan DocScrutinizer05 (~saturn@openmoko/engineers/joerg) |
| 06:08.48 | *** join/#devuan hightower2 (~hightower@192-234.dsl.iskon.hr) |
| 06:14.53 | *** join/#devuan lantw44 (~Ting-Wei_@linux5.csie.ntu.edu.tw) |
| 06:22.11 | *** join/#devuan Katnija (~KittenGNU@unaffiliated/kittengnu) |
| 06:45.54 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-78-12-58-52.clienti.tiscali.it) |
| 07:29.32 | *** join/#devuan davec_ (~davec@host-69-145-45-32.shr-wy.client.bresnan.net) |
| 07:29.32 | *** join/#devuan edbarx (~edbarx@78.133.22.227) |
| 07:30.53 | *** join/#devuan edbarx (~edbarx@78.133.22.227) |
| 07:31.30 | *** join/#devuan giricz81 (~giricz81@ip4-83-240-38-90.cust.nbox.cz) |
| 07:31.34 | *** join/#devuan Levure (~quassel@91.179.237.111) |
| 07:35.37 | *** join/#devuan davec_ (~davec@host-69-145-45-32.shr-wy.client.bresnan.net) |
| 08:09.49 | *** join/#devuan bpmedley (~bpm@rrcs-24-199-194-123.midsouth.biz.rr.com) |
| 08:28.02 | *** join/#devuan JohnnyRun (~gianni@net94-124-67-2.static.bb.mainsoft.it) |
| 08:46.03 | *** join/#devuan peetaur (~peter@p57AAAF06.dip0.t-ipconnect.de) |
| 08:48.16 | *** join/#devuan Besnik_b (~Besnik@athedsl-221717.home.otenet.gr) |
| 09:00.46 | *** join/#devuan oel (~me@p5B37789C.dip0.t-ipconnect.de) |
| 09:05.15 | *** join/#devuan antenagora (~antenagor@147.162.124.215) |
| 09:09.56 | *** join/#devuan b0stik (~b0stik@net-2-42-70-204.cust.vodafonedsl.it) |
| 09:23.59 | *** join/#devuan Katnija (~KittenGNU@unaffiliated/kittengnu) |
| 09:33.15 | *** join/#devuan engideavr (~damiano@host10-38-static.197-31-b.business.telecomitalia.it) |
| 09:45.01 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-84-220-218-163.clienti.tiscali.it) |
| 09:47.14 | *** join/#devuan AntoFox1 (~Thunderbi@dynamic-adsl-78-12-46-146.clienti.tiscali.it) |
| 09:53.43 | *** join/#devuan engideavr (~damiano@host10-38-static.197-31-b.business.telecomitalia.it) |
| 10:04.51 | *** join/#devuan Spagno (~Spagno@2.229.195.21) |
| 10:12.22 | Hestben | Big ups for Centurion Dan sorting out that failing drive issue. It was an interesting adventure to read through it ;) |
| 10:17.07 | *** join/#devuan djph (~dpurgert@104-57-151-177.lightspeed.bcvloh.sbcglobal.net) |
| 10:18.16 | *** join/#devuan ferdy- (~ferdy@funtoo/contrib/ferdy-) |
| 10:25.47 | Centurion_Dan1 | Hestben: It's experience and taking a methodical approach to solving the problem... took up way to much of my time tho... |
| 10:26.23 | Centurion_Dan1 | should really have pushed him on to the linux-raid mailing list. |
| 10:28.08 | *** join/#devuan AntoFox1 (~Thunderbi@dynamic-adsl-78-12-44-25.clienti.tiscali.it) |
| 10:35.22 | Hestben | Centurion_Dan1: Yeah, methodology is key when doing stuff. And there are some concepts you need to grasp for using mdadm |
| 10:36.12 | Hestben | I have had my fair share of desparate moments where I am cursing, and not just calmly go to the bottom of things |
| 10:36.29 | *** join/#devuan AntoFox1 (~Thunderbi@dynamic-adsl-78-12-39-120.clienti.tiscali.it) |
| 10:36.47 | Hestben | Including time spent figuring out mdadm stuff |
| 10:38.57 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-78-12-52-14.clienti.tiscali.it) |
| 10:47.17 | *** join/#devuan blueness (~blueness@gentoo/developer/blueness) |
| 10:51.14 | Centurion_Dan1 | I'm a problem solver, and taking time and maintaining your cool is critical. |
| 10:57.34 | *** join/#devuan AntoFox1 (~Thunderbi@dynamic-adsl-78-12-50-186.clienti.tiscali.it) |
| 11:07.08 | Centurion_Dan1 | drools over https://globalscaletechnologies.com/p-72-marvell-espressobin.aspx |
| 11:07.27 | Centurion_Dan1 | bugger it'll be 3 months wait.... |
| 11:25.06 | *** join/#devuan SylvieLorxu (~TheLastPr@541B7AAC.cm-5-4b.dynamic.ziggo.nl) |
| 11:37.11 | *** join/#devuan mystified (~whatif@110-174-9-2.static.tpgi.com.au) |
| 11:39.38 | DocScrutinizer05 | Centurion_Dan: yep, sorry it took so long |
| 11:41.56 | *** join/#devuan chomwitt1 (~chomwitt@ppp-94-66-61-113.home.otenet.gr) |
| 11:42.06 | DocScrutinizer05 | >>Marvell 64bit Dual Core ARM A53 Armada 3700 SOC<< hmmmm, are there manuals for that thing? |
| 11:44.14 | DocScrutinizer05 | >>optional 4GB EMMC<< WTF? aren't they JEDEC and thus "arbitrary size"? |
| 11:45.17 | DocScrutinizer05 | also haha at soldering a JEDEC eMMC in there, ideally with a 5⬠soldering iron |
| 11:50.18 | *** join/#devuan blueness (~blueness@gentoo/developer/blueness) |
| 11:52.00 | Centurion_Dan | It's marvells latest chip targeted at their soho NAS products. |
| 11:52.33 | DocScrutinizer05 | yeah, but does it have public docs? |
| 11:52.59 | DocScrutinizer05 | I'm not exactly a fan of Marvel already |
| 11:53.42 | DocScrutinizer05 | and are there schematics of the board itself? |
| 11:55.09 | DocScrutinizer05 | I mean, was it RasPi which needs a GFX binary blob to *boot* the whole thing? |
| 11:55.53 | Centurion_Dan | apparently it has mainline kernel support since v4.6 |
| 11:56.01 | DocScrutinizer05 | good |
| 11:56.50 | DocScrutinizer05 | and docs, so I know the GPIO numbers of the different signals? |
| 11:58.05 | DocScrutinizer05 | look, N900 had FOSS 'kernel support' (though not all was upstreamed) since very beginning |
| 11:59.39 | DocScrutinizer05 | still it's demanding even after 7 yeahs to develop low level software for it, you need a schematics, the TI OMAP3430 TRM, and quite some understanding of how stuff works, to do it |
| 11:59.59 | DocScrutinizer05 | years* |
| 12:00.40 | DocScrutinizer05 | ask freemangordon about e.g. camera |
| 12:01.41 | DocScrutinizer05 | or OpenGL-ES for 3D on the Imagination PVR530 |
| 12:03.27 | DocScrutinizer05 | not about the hardware, but the general analysis of industry and hw situation at large, is excellent in https://www.crowdsupply.com/raptor-computing-systems/talos-secure-workstation/updates/the-state-of-owner-controlled-computing-as-talos-winds-down?utm_source=Talos+Supporters&utm_campaign=19545976d2-talos_update_winds_down_2017_01_09&utm_medium=email&utm_term=0_451951d5fa-19545976d2-112395773 |
| 12:03.53 | DocScrutinizer05 | I still want to *own* the hardware I buy |
| 12:04.25 | DocScrutinizer05 | while we live in the age of Tivoization |
| 12:05.17 | DocScrutinizer05 | each shitty phone needs a jailbreak or CM, so you at least could do a *few* maintenance level things |
| 12:06.29 | DocScrutinizer05 | and "but this is Android (BSD based iOS), so Linux (BSD), so FOSS" is a lie then |
| 12:08.12 | DocScrutinizer05 | as long as you can't build *everything* (particularly Bootloader) from source, you don't own the thing |
| 12:09.15 | DocScrutinizer05 | I don't mind blob firmware for WiFi or BT, but the core system needs to be all FOSS, and have full hw docs |
| 12:09.56 | DocScrutinizer05 | meanwhile I even think the same about PC BIOS |
| 12:11.56 | DocScrutinizer05 | soon we will need a manufacturer signed Linux variant to install on our laptops/tablets/PCs (in that chronological order) |
| 12:14.41 | DocScrutinizer05 | and in my book those who are working to support secureboot in linux are actually "doing the enemy's job" |
| 12:16.06 | DocScrutinizer05 | http://en.qi-hardware.com/w/images/1/10/ME_382_LockedUpTechnology2.gif |
| 12:17.12 | MinceR | :) |
| 12:24.06 | DocScrutinizer05 | oops, the chronological order above swapped tablets and laptops |
| 12:37.49 | *** join/#devuan snux (~snux@net-188-217-95-145.cust.vodafonedsl.it) |
| 12:49.15 | *** join/#devuan davec__ (~davec@host-69-145-45-32.shr-wy.client.bresnan.net) |
| 12:51.28 | *** join/#devuan Xenguy (~Xenguy@204.101.156.79) |
| 12:56.55 | *** join/#devuan blueness (~blueness@gentoo/developer/blueness) |
| 12:57.49 | *** join/#devuan Centurion_Dan1 (~Icedove@office.centurion.net.nz) |
| 12:58.39 | *** join/#devuan chomwitt2 (~chomwitt@ppp-94-67-200-3.home.otenet.gr) |
| 12:59.52 | Centurion_Dan | alright, bed time... |
| 13:01.53 | *** join/#devuan Guest27061 (~Guest2706@83.220.238.135) |
| 13:02.33 | *** part/#devuan Guest270614 (~Guest2706@83.220.238.135) |
| 13:06.00 | *** join/#devuan cyteen (~cyteen@249.158.93.209.dyn.plus.net) |
| 13:12.48 | *** join/#devuan ccrhy (~ccrhy3@24.216.245.27) |
| 13:35.09 | *** join/#devuan thehornet (~a0225280@192.91.101.32) |
| 13:38.03 | *** join/#devuan Spagno (~Spagno@2.229.195.21) |
| 13:55.30 | *** join/#devuan hightower3 (~hightower@141-136-235-65.dsl.iskon.hr) |
| 14:01.45 | *** join/#devuan hightower2 (~hightower@141-136-203-47.dsl.iskon.hr) |
| 14:09.01 | *** join/#devuan FlibberTGibbet (~David@cust241-dsl93-89-129.idnet.net) |
| 14:09.11 | *** join/#devuan rafalcpp (~racalcppp@84-10-11-234.static.chello.pl) |
| 14:27.49 | *** join/#devuan rleigh__ (~rleigh@182.155.187.81.in-addr.arpa) |
| 14:28.02 | gnu_srs1 | Hi, seems like build-rdeps fails not finding Devuan names: my $release_pattern = '(.*_dists_(sid|unstable))_(?:In)*Release$'; and |
| 14:28.03 | gnu_srs1 | find(sub { readrelease($_, $1) if /$release_pattern/ }, $sources_path); |
| 14:28.03 | gnu_srs1 | if (!@source_files) {die "$progname: unable to find sources files.\nDid you forget to run apt-get update (or add --update to this command)?";} |
| 14:32.51 | *** join/#devuan cirucito (~edu@230.104.218.87.dynamic.jazztel.es) |
| 14:36.16 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-78-12-50-186.clienti.tiscali.it) |
| 14:40.59 | *** join/#devuan hightower2 (~hightower@247-234.dsl.iskon.hr) |
| 14:41.21 | *** join/#devuan Spagno (~Spagno@2.229.195.21) |
| 14:47.37 | *** join/#devuan hightower3 (~hightower@89-164-111-51.dsl.iskon.hr) |
| 14:51.04 | *** join/#devuan hightower2 (~hightower@141-138-23-177.dsl.iskon.hr) |
| 14:52.21 | *** join/#devuan Spagno (~Spagno@2.229.195.21) |
| 14:54.17 | *** join/#devuan hightower3 (~hightower@89-164-111-109.dsl.iskon.hr) |
| 15:03.33 | *** join/#devuan Spagno (~Spagno@2.229.195.21) |
| 15:09.47 | *** join/#devuan IoFran (~Icedove@189.231.21.160) |
| 15:12.15 | *** join/#devuan TemporalBeing1 (~Ben_Meyer@72.32.180.176) |
| 15:16.47 | *** join/#devuan engidea-vr (~damiano@host10-38-static.197-31-b.business.telecomitalia.it) |
| 15:17.54 | *** join/#devuan DeFender1031 (~DeFender1@89-139-120-54.bb.netvision.net.il) |
| 15:18.03 | *** join/#devuan zdzichu_ (~zdzichu@2002:b906:1d37::137) |
| 15:18.03 | *** join/#devuan koollman_ (samson_t@sp1.kooll.info) |
| 15:18.07 | *** join/#devuan berndj (~berndj@mail.azna.co.za) |
| 15:18.16 | *** join/#devuan djph_ (~dpurgert@104-57-151-177.lightspeed.bcvloh.sbcglobal.net) |
| 15:18.19 | *** join/#devuan zocker_ (~tobias@2a00:1158:2:9b20::666) |
| 15:18.34 | *** join/#devuan joostvb_ (~joostvb@beskar.soleus.nu) |
| 15:18.41 | *** join/#devuan _spOOn__ (~adam@nemesis.spoontech.biz) |
| 15:18.52 | *** join/#devuan doppo_ (doppo@2604:180::e0fc:a07f) |
| 15:18.52 | *** join/#devuan rleigh (~rleigh@134.36.162.100) |
| 15:18.53 | *** join/#devuan rleigh (~rleigh@unaffiliated/rleigh) |
| 15:19.19 | *** join/#devuan Katnija (~KittenGNU@unaffiliated/kittengnu) |
| 15:19.38 | *** join/#devuan uidzer0 (~uidzer0@168.235.65.160) |
| 15:19.38 | *** join/#devuan detha (~detha@unaffiliated/detha) |
| 15:19.39 | *** join/#devuan jotik (~7f000001@unaffiliated/jotik) |
| 15:19.39 | *** join/#devuan technoid_ (~technoid@3jane.sbce.org) |
| 15:20.12 | *** join/#devuan drawkula (~username@p4FC10C9B.dip0.t-ipconnect.de) |
| 15:21.07 | *** join/#devuan gattuso (~gattuso@gruppoculturale.it) |
| 15:21.53 | *** join/#devuan hanx_ (~hanx@unaffiliated/hanx/x-2769695) |
| 15:22.45 | *** join/#devuan sirix (Elite16648@gateway/shell/elitebnc/session) |
| 15:22.45 | *** join/#devuan jotun (~marvin@2001:41d0:e:275::1) |
| 15:22.45 | *** join/#devuan davec__ (~davec@host-69-145-45-32.shr-wy.client.bresnan.net) |
| 15:22.45 | *** join/#devuan cdanderson42 (~nuhrin@67.170.89.70) |
| 15:22.45 | *** join/#devuan zjason (~user@36-229-108-188.dynamic-ip.hinet.net) |
| 15:22.50 | *** join/#devuan sirix (Elite16648@gateway/shell/elitebnc/x-rhlufwfovyxnuras) |
| 15:23.07 | *** join/#devuan Bryanstein (~Bryanstei@shellium/admin/bryanstein) |
| 15:24.14 | *** join/#devuan rakiru (~rakiru@hi.i.wanted.to.let.you.all.know.that.i.think.incest.is.wince.st) |
| 15:24.27 | *** join/#devuan TwistedFate (~TwistedFa@unaffiliated/twistedfate) |
| 15:46.15 | *** join/#devuan Madda (~textual@hq.m3team.it) |
| 15:50.31 | *** join/#devuan SylvieLorxu (~TheLastPr@541B7AAC.cm-5-4b.dynamic.ziggo.nl) |
| 15:52.11 | *** join/#devuan chomwitt3 (~chomwitt@ppp-94-69-222-21.home.otenet.gr) |
| 15:57.20 | *** join/#devuan rypervenche (~rypervenc@unaffiliated/rypervenche) |
| 15:57.42 | *** join/#devuan furrywolf (~randyg@70-6-249-192.pools.spcsdns.net) |
| 16:04.42 | *** join/#devuan Humpelstilzchen (erik@x4e366d70.dyn.telefonica.de) |
| 16:08.48 | gnu_srs1 | Solved by: --distribution jessie --origin Devuan. Maybe the Devuan version of that perl script should s/Debian/Devuan/g to avoid using --origin? |
| 16:14.11 | *** join/#devuan wigbold (~me@p3E9C28E7.dip0.t-ipconnect.de) |
| 16:16.02 | Hestben | gnu_srs1: That seems logical. |
| 16:17.01 | Hestben | I'm not involved in anything of that. But I guess any maintainer would be interested in which version you are running (beta2 with the newest upgrades of today?) |
| 16:20.07 | *** join/#devuan golinux (~golinux@unaffiliated/golinux) |
| 16:22.53 | *** join/#devuan Madda (~Madda@hq.m3team.it) |
| 16:24.56 | gnu_srs1 | Hestben: beta2 with the newest upgrades of today: yes |
| 16:35.47 | *** join/#devuan AntoFox (~Thunderbi@dynamic-adsl-78-12-50-186.clienti.tiscali.it) |
| 16:41.49 | golinux | A question asked on https://dev1galaxy.org/viewtopic.php?pid=495#p495 |
| 16:42.22 | golinux | ". . . in my poking around, that there is a lot of systemd stuff in /etc/systemd/system, /lib/systemd, and /var/lib/systemd. What's that all about? Or did it get sucked in by something I installed maybe?" |
| 16:42.51 | golinux | ping Centurion_Dan , jaromil, nextime ^^^ |
| 16:47.43 | *** join/#devuan giricz81 (~giricz81@ip4-83-240-38-90.cust.nbox.cz) |
| 16:50.57 | *** join/#devuan Pali (~pali@Maemo/community/contributor/Pali) |
| 16:51.28 | *** join/#devuan panta_ (~panta@h081217057026.dyn.cm.kabsi.at) |
| 17:03.27 | KatolaZ | golinux: those should be systemd-related scripts installed by other packages |
| 17:04.19 | muep | it is usually harmless for a package to ship a config file for something that is not installed |
| 17:05.54 | *** join/#devuan cdanderson42 (~nuhrin@c-50-170-120-18.hsd1.wa.comcast.net) |
| 17:08.57 | golinux | muep: If a package is not installed, why would related config files be sucked in for a non--existent purpose? |
| 17:09.53 | muep | golinux: because packages include those files for the cases where they are installed alongside the thing that does not happen to be present |
| 17:10.27 | muep | and because devuan is using debian packages as-is where they are suitable |
| 17:18.38 | KatolaZ | golinux: and because we should not make the same mistake that all the distros have done |
| 17:18.50 | KatolaZ | i.e., making it impossible to use anything else |
| 17:19.08 | KatolaZ | if a package has scripts for runit or openrc or systemd, why should we remove them? |
| 17:19.30 | KatolaZ | those script will never be executed if runit, openrc, or systemd is not installed |
| 17:22.34 | *** join/#devuan edbarx (~edbarx@78.133.22.227) |
| 17:31.51 | Leander256 | do you think that at some point, it will be a reasonnable (technically speaking) Devuan feature to be able to switch back to systemd? |
| 17:32.49 | MinceR | seems to be unreasonably difficult already |
| 17:32.52 | golinux | Leander256: Why would anyone want to do that? Easier to just install Debian |
| 17:33.05 | MinceR | and systemd people will probably make it more difficult in the future |
| 17:33.26 | MinceR | (that is, if you are to maintain compatibility with non-systemd systems) |
| 17:35.43 | Leander256 | golinux it's really just from a technical point of view, since ideally it should be easy to switch from one init system to another |
| 17:36.06 | golinux | systemd has made modularity impossible. |
| 17:38.11 | golinux | Before too long, there will be two Linuxes - the 'monolith' and all other modular init systems. |
| 17:38.30 | djph_ | golinux: so we'll have "Linux" and "systemd" |
| 17:38.34 | Leander256 | if that's the case, I could see it as an argument to get rid of all systemd related files |
| 17:38.45 | MinceR | we already have GNU/Linux and systemd/Linux |
| 17:38.58 | Leander256 | but then, that's probably too much effort for a very small net worth |
| 17:39.14 | djph_ | heh, sounds like RMS may get his way, and people will be calling it gnu/linux |
| 17:40.24 | golinux | Leander256: Perhaps eventually all those files could be cleaned out but atm, there are more important targets to de-systemdize |
| 17:40.53 | MinceR | RMS has already given up on it, he supports systemd |
| 17:41.59 | Leander256 | and what's your source about that/ |
| 17:42.01 | Leander256 | ? |
| 17:46.30 | MinceR | https://www.mail-archive.com/dng@lists.dyne.org/msg01743.html |
| 17:48.19 | *** join/#devuan minnesotags (~herbgarci@c-50-137-242-97.hsd1.mn.comcast.net) |
| 17:49.11 | KatolaZ | MinceR: RMS does not support systemd |
| 17:49.22 | KatolaZ | he has just said that he doesn't have an opinion on that |
| 17:49.31 | KatolaZ | and since it is free software, for him it is fine |
| 17:49.31 | MinceR | he claimed it was free software |
| 17:49.34 | djph_ | All he said was "meh, it's free software". That was a very safe position to take - neither for or against it |
| 17:49.37 | KatolaZ | and it is free software |
| 17:49.44 | MinceR | which means he believes it supports the freedoms of its users |
| 17:49.47 | MinceR | which is not true |
| 17:50.01 | KatolaZ | it just means that it is distributed under a free software license |
| 17:50.03 | KatolaZ | fullstop |
| 17:50.06 | MinceR | nope |
| 17:50.15 | MinceR | free software is not defined on basis of license |
| 17:50.16 | KatolaZ | then, if you hate RMS, go on with your tant |
| 17:50.19 | KatolaZ | ~rant |
| 17:50.20 | MinceR | RMS should know -- he wrote the FSD |
| 17:50.35 | MinceR | "open source" is the term that is based on license -- compare the OSD with the FSD |
| 17:50.44 | KatolaZ | MinceR: freedom to use, distribute, modify, and redistribute the changes |
| 17:50.45 | MinceR | systemd is open source, but it is not free software |
| 17:50.52 | KatolaZ | which of those is missing from systemd? |
| 17:51.00 | KatolaZ | MinceR: it is free software |
| 17:51.06 | MinceR | yeah, how free are you to use it if it comes tied with a whole lot of crap? |
| 17:51.07 | KatolaZ | because it respects the 4 freedoms |
| 17:51.18 | KatolaZ | MinceR: and then also KDE is not free software |
| 17:51.25 | KatolaZ | since it comes tied with a whole lot of crap |
| 17:51.34 | KatolaZ | and noone can easily modify it |
| 17:51.39 | MinceR | how free are you to modify it if they deliberately write unreadable, unusable crap code, maintain and use unstable, undocumented APIs? |
| 17:51.40 | KatolaZ | this is just nonsense :) |
| 17:51.45 | MinceR | kde5 is the same crap |
| 17:51.48 | MinceR | and so is gnome3 |
| 17:51.54 | KatolaZ | MinceR: have you read the code? :) |
| 17:51.59 | MinceR | i have |
| 17:52.09 | KatolaZ | but kde is free software, whatever your opinion on that |
| 17:52.17 | KatolaZ | and gnome 3 is free software as well |
| 17:52.22 | KatolaZ | and systemd, as well |
| 17:52.23 | MinceR | keep saying that |
| 17:52.41 | MinceR | maybe that will erase the gratuitous dependencies that are making our life difficult |
| 17:53.03 | KatolaZ | MinceR: subverting reality does not make it better, or worse |
| 17:53.11 | KatolaZ | being freesoftware is a fact |
| 17:53.13 | Leander256 | I honestly think that systemd violates the spirit of free software as defined by RMS, but then maybe he has more important things to do than to arbiter such things |
| 17:53.13 | MinceR | i'll just keep using software that is not designed with the intent to force my hand |
| 17:53.20 | MinceR | and to make me give up control over my systems |
| 17:53.24 | KatolaZ | than can be verified or falsified |
| 17:53.29 | KatolaZ | is not an opinion |
| 17:53.46 | MinceR | Leander256: i thought this was literally his job |
| 17:53.56 | MinceR | also of the FSF, which has also forsaken this responsibility |
| 17:54.21 | KatolaZ | MinceR: keep hating RMS and the FSF if you like :) |
| 17:54.28 | KatolaZ | but use facts |
| 17:54.44 | MinceR | i do |
| 17:54.44 | KatolaZ | otherwise you might be mistaken for a troll ;) |
| 17:55.00 | MinceR | yeah, i know, everyone who doesn't love systemd and give it all the resources they have is a "troll" |
| 17:55.11 | KatolaZ | :D |
| 17:55.12 | Leander256 | IIRC, it does not violate the 4 rules, but it does not respect the spirit |
| 17:55.15 | MinceR | like the quoted person claiming that asking RMS about systemd is "trolling" |
| 17:55.29 | MinceR | i think it violates at least 2 of those freedoms in spirit |
| 17:55.36 | KatolaZ | Leander256: I hate systemd, but this does not allow me to say that it is not free software |
| 17:56.18 | KatolaZ | MinceR: there is no violation "in spirit" |
| 17:56.24 | KatolaZ | :) |
| 17:56.31 | KatolaZ | either you violate it or you don't |
| 17:56.41 | *** join/#devuan ppisati_ (~ppisati@2-235-152-247.ip228.fastwebnet.it) |
| 17:56.42 | MinceR | why, is it legal code? |
| 17:56.49 | KatolaZ | if we speak of code clarity, gcc should beconsidered at the same level of systemd |
| 17:56.54 | MinceR | or a mathematical expression? |
| 17:57.09 | KatolaZ | since some parts of gcc are *intentionally* obfuscated |
| 17:57.23 | MinceR | afaik gcc is intended to be portable and can very well coexist with other compilers/linkers/preprocessors/etc |
| 17:57.32 | MinceR | but yeah, i've heard of the horrors of gcc |
| 17:57.50 | MinceR | if it's intentionally obfuscated, it violates at least one of the freedoms |
| 17:57.54 | KatolaZ | :D |
| 17:57.58 | KatolaZ | nope |
| 17:58.10 | MinceR | also, it may cause an interesting legal situation regarding its distribution |
| 17:58.19 | KatolaZ | it was intentionally obfuscated to avoid *stealing* |
| 17:58.21 | MinceR | as the "preferred form" for editing is not available to anyone |
| 17:58.21 | KatolaZ | literally |
| 17:58.28 | MinceR | how could it be "stolen"? |
| 18:01.06 | KatolaZ | MinceR: you have the source code, so why the "preferred form" for editing is not available to anyone? |
| 18:01.15 | KatolaZ | :) |
| 18:01.25 | MinceR | is obfuscated source code the "preferred form" then? |
| 18:01.30 | MinceR | that's bad news to copyleft |
| 18:01.33 | KatolaZ | it is source code |
| 18:01.41 | KatolaZ | and there is much worse code around |
| 18:01.51 | KatolaZ | and it's free software :) |
| 18:01.53 | KatolaZ | anyway |
| 18:02.08 | KatolaZ | you're free to hate RMS and the FSF |
| 18:02.25 | MinceR | i doubt you'll see much third-party contributions to obfuscated code |
| 18:02.33 | KatolaZ | but I remain convinced that they are the real target |
| 18:02.45 | MinceR | and i doubt you'll see anyone taking over the development of a dead project the code of which was always obfuscated |
| 18:02.50 | KatolaZ | they are *not* the real target |
| 18:02.54 | MinceR | oh, they are targets indeed |
| 18:02.57 | MinceR | targets of red hat |
| 18:03.00 | KatolaZ | :) |
| 18:03.06 | MinceR | and the bribes have hit home |
| 18:03.31 | KatolaZ | MinceR: you can say anything, but not that RMS has been bought by RedHat |
| 18:03.41 | KatolaZ | RMS has no price |
| 18:03.43 | KatolaZ | :) |
| 18:03.45 | muep | most of systemd code looks like plain boring C code |
| 18:03.48 | MinceR | if obfuscated source code is acceptable, is decompiled source code also acceptable? |
| 18:03.59 | KatolaZ | MinceR: you are missing the point |
| 18:03.59 | muep | it is not obfuscated |
| 18:04.03 | MinceR | yeah, plain boring undocumented uncommented shitty C code |
| 18:04.11 | KatolaZ | muep: I was talking of a small part of gcc |
| 18:04.12 | MinceR | with tons of null pointer dereferences |
| 18:04.20 | KatolaZ | which was *intentionally* obfucated |
| 18:04.21 | MinceR | KatolaZ: i think you are missing the point |
| 18:04.40 | KatolaZ | that one was related to the translation to the internal code format |
| 18:04.44 | MinceR | the whole point of having the source code available is that anyone can read it, understand it and meaningfully modify it |
| 18:04.59 | KatolaZ | and RMS has explained many times that he did that as a form of "guerrilla-coding" |
| 18:05.09 | KatolaZ | MinceR: you can understand the whole gcc |
| 18:05.11 | MinceR | do you have a link to him explaining that? |
| 18:05.16 | KatolaZ | if you know how a compiler works |
| 18:05.17 | KatolaZ | only |
| 18:05.26 | MinceR | you can understand the whole gcc except for the obfuscated parts |
| 18:05.27 | KatolaZ | it would be VERY DIFFICULT to copy it straight away |
| 18:05.30 | KatolaZ | and steal it |
| 18:05.37 | MinceR | the whole point of obfuscation is to make it difficult to understand |
| 18:05.47 | MinceR | actually, gcc is easy to copy |
| 18:05.53 | KatolaZ | ok |
| 18:05.58 | KatolaZ | you won |
| 18:06.00 | KatolaZ | go on :) |
| 18:06.04 | Leander256 | are you talking about the tight coupling between the front-end and the back-end? |
| 18:06.04 | MinceR | which, again, doesn't help someone who wants to understand/modify it |
| 18:06.21 | MinceR | but to steal it, you'd somehow have to destroy the original |
| 18:06.38 | KatolaZ | MinceR: we are talking of a software developed in 1985 |
| 18:06.43 | MinceR | btw, did rms literally call it "guerrilla coding"? |
| 18:06.46 | KatolaZ | when all the compilers were *proprietary* |
| 18:06.54 | KatolaZ | but I guess this is too complicated to explain |
| 18:06.56 | MinceR | the FSD doesn't apply to software written in 1985? |
| 18:07.04 | KatolaZ | and I am not entitled or interested to do that |
| 18:07.13 | KatolaZ | yes MinceR |
| 18:07.17 | MinceR | interesting |
| 18:07.21 | KatolaZ | he called it "guerrilla-coding" |
| 18:07.24 | MinceR | oh, i see |
| 18:07.29 | MinceR | that will help me find it :) |
| 18:07.41 | KatolaZ | I suppose :) |
| 18:08.21 | *** join/#devuan clopez (~tau@neutrino.es) |
| 18:08.24 | MinceR | makes me wonder if we have any fully free c compilers at all |
| 18:08.39 | *** join/#devuan ppisati (~ppisati@2-235-152-247.ip228.fastwebnet.it) |
| 18:08.46 | KatolaZ | MinceR: we have gcc |
| 18:08.52 | KatolaZ | which is free software |
| 18:09.07 | KatolaZ | and we have plenty of other compilers around |
| 18:09.10 | MinceR | i mean, in the sense of actually supporting the 4 freedoms |
| 18:09.11 | KatolaZ | which are free software |
| 18:09.23 | KatolaZ | MinceR: gcc *actually* supports the 4 freedoms |
| 18:09.31 | MinceR | not just "free if you can afford to reverse engineer it" |
| 18:09.38 | KatolaZ | ok gotta go! :) |
| 18:09.39 | MinceR | not if it's partially or fully obfuscated |
| 18:09.40 | KatolaZ | HND |
| 18:11.04 | muep | it would seem quite difficult to produce a practical C compiler that is not at least complex to some extent |
| 18:11.17 | MinceR | complex is one thing |
| 18:11.27 | MinceR | deliberately undocumented is another |
| 18:11.33 | MinceR | deliberately obfuscated is yet another |
| 18:12.00 | muep | few things are deliberately undocumented. most projects will happily accept docs if someone writes them |
| 18:12.23 | MinceR | most project happily accept improvements from outside |
| 18:12.29 | MinceR | systemd explicitly doesn't |
| 18:13.14 | muep | depends on what counts as outside. my impression is that it is not very uncommon to have drive-by patches in there |
| 18:13.40 | MinceR | iirc they specifically stated that they won't accept patches to make it work on other kernels, for example |
| 18:13.56 | muep | that is not the same thing at all |
| 18:14.05 | *** join/#devuan minnesotags (~herbgarci@c-50-137-242-97.hsd1.mn.comcast.net) |
| 18:14.15 | MinceR | right, because portability is not an improvement, it's sacrilege |
| 18:14.19 | muep | it is patches that do not align with views of the main contributors |
| 18:14.26 | MinceR | right |
| 18:14.32 | muep | whose rejection would again be quite common |
| 18:14.37 | KatolaZ | sorry, had to run to the baby :) |
| 18:14.39 | MinceR | and the views of the main contributors are opposed to sane software engineering |
| 18:14.55 | MinceR | and to freedom |
| 18:14.57 | KatolaZ | MinceR: in theory, you can fork systemd, and make it support FreeBSD, if you like |
| 18:15.05 | KatolaZ | nothing prevents you from doing that |
| 18:15.10 | muep | it is a fair argument that you would run a comparable project differently |
| 18:15.12 | MinceR | yes, but it will remain a fork |
| 18:15.26 | KatolaZ | so what? |
| 18:15.38 | KatolaZ | forking is part of the 4 feedoms |
| 18:15.45 | KatolaZ | namely 2 and 3 |
| 18:16.05 | KatolaZ | and freedom 0 and 1 are guaranteed as well |
| 18:16.12 | MinceR | the topic was a project accepting docs if someone wrote them |
| 18:16.12 | KatolaZ | because you can use systemd without restrictions |
| 18:16.16 | KatolaZ | and redistribute it |
| 18:16.28 | KatolaZ | MinceR: we are running in circles |
| 18:16.30 | MinceR | yeah, having to throw out half my OS is not a restriction |
| 18:16.38 | KatolaZ | you said that systemd is not free software |
| 18:16.38 | muep | nothing in software freedom sets an expectation that the original author is mandated to accept any changes back |
| 18:16.41 | MinceR | having to pipe all my logs through journald is not a restriction either |
| 18:16.43 | KatolaZ | I am telling you that it is |
| 18:16.53 | KatolaZ | you might not like it, but this does not change the reality |
| 18:17.01 | MinceR | KatolaZ: and so far all you could say that it's under a free software licence |
| 18:17.10 | KatolaZ | MinceR: no, it;s not a restriction, it's a *feature* of that softwaree |
| 18:17.11 | muep | users of the program are free to organise themselves to run the project better if the original maintenance is doing a bad job |
| 18:17.13 | MinceR | and that is also apparently what rms based his comment on |
| 18:17.23 | KatolaZ | yes |
| 18:17.25 | KatolaZ | exactly |
| 18:17.34 | MinceR | which means he forgot what he wrote in the FSD |
| 18:17.36 | KatolaZ | because rms is concerned with freedom |
| 18:17.38 | KatolaZ | yes |
| 18:17.40 | MinceR | no he is not |
| 18:17.40 | KatolaZ | exactly |
| 18:17.45 | MinceR | he is concerned with a few licenses |
| 18:17.52 | MinceR | which should not be confused with freedom |
| 18:17.53 | muep | this happened e.g. to openoffice.org with patches getting rejected for a while and users organising these in an organization that was separate from the main development effort |
| 18:17.58 | KatolaZ | he has never spoken against a software on the basis of technical arguments |
| 18:18.05 | KatolaZ | or drawbacks |
| 18:18.07 | MinceR | this is not only technical |
| 18:18.13 | KatolaZ | forget it |
| 18:18.22 | KatolaZ | it is distributed with a free software license? |
| 18:18.29 | KatolaZ | then it is free, as in the FSD |
| 18:18.32 | KatolaZ | fullstop |
| 18:18.32 | MinceR | it is, but that only makes it open source, not free software |
| 18:18.33 | muep | eventually the OO.o users just set up a competing project and out-developed the parent project |
| 18:18.37 | MinceR | no, the FSD doesn't mention licenses |
| 18:18.40 | KatolaZ | you are wrong MinceR |
| 18:18.42 | MinceR | it is in terms of freedoms |
| 18:18.56 | MinceR | freedoms, at least some of which are violated by systemd and the conduct of the people behind it |
| 18:18.58 | KatolaZ | MinceR: a free software licence is a licence which guarantees the four freedoms |
| 18:19.01 | KatolaZ | by definition |
| 18:19.07 | MinceR | it is impossible to guarantee those |
| 18:19.11 | KatolaZ | uh? |
| 18:19.17 | KatolaZ | the freedom to use the software |
| 18:19.18 | MinceR | licences can only work within the framework of law |
| 18:19.24 | KatolaZ | to redistribute it |
| 18:19.26 | KatolaZ | to modify it |
| 18:19.32 | KatolaZ | to redistribute the modifications? |
| 18:19.40 | MinceR | GPLv3 does as much as it can to guarantee it, but that goes only as far as a copyright/patent licence can go |
| 18:19.40 | KatolaZ | where is systemd falling short? |
| 18:19.41 | muep | public domain software can be free software, too, without laws enforcing it |
| 18:19.54 | KatolaZ | exactly muep |
| 18:20.06 | MinceR | and public domain does even less to guarantee freedoms |
| 18:20.17 | KatolaZ | but only because public domain allows you to do what is guaranteed by the four freedoms |
| 18:20.20 | muep | MinceR: it can guarantee you freedoms with regards to a copy you have |
| 18:20.40 | KatolaZ | MinceR: but free software is about law |
| 18:20.41 | MinceR | so, if i distributed binaries with no sources and no documentation as public domain, you'd insist it's free software? |
| 18:20.46 | muep | MinceR: it does not guarantee that all derivatives of the program are free software, but that is also not included in the FSD |
| 18:20.48 | MinceR | no, free software is about freedom |
| 18:20.49 | KatolaZ | nope |
| 18:21.11 | KatolaZ | in fact muep said that public domain *can* guarantee some of the freedoms,sometimes |
| 18:21.20 | MinceR | so how does public domain guarantee the freedoms in that case? |
| 18:21.46 | KatolaZ | MinceR: a software is free software if you can use, distribute, modify, and redistribute modifications |
| 18:21.49 | KatolaZ | fullstop |
| 18:21.50 | muep | MinceR: if you have a public-domain copy with sources it is is free software, even if you are able to make non-free copies of it |
| 18:21.56 | KatolaZ | you can add any other constraints that yoiu like |
| 18:22.02 | KatolaZ | but this is what is in the FSD |
| 18:22.04 | KatolaZ | nothing else |
| 18:22.08 | KatolaZ | nothing less |
| 18:22.10 | KatolaZ | nothing more |
| 18:22.33 | *** join/#devuan minnesotags (~herbgarci@c-50-137-242-97.hsd1.mn.comcast.net) |
| 18:22.58 | MinceR | so, if a software introduces gratuitous restriction on what i can do with my own computer, and what else i am allowed to run on it, does that support my freedoms? |
| 18:23.03 | MinceR | s/ion/ions/ |
| 18:23.24 | muep | MinceR: depends on how it introduces it. if it is by law, it is not free |
| 18:23.26 | KatolaZ | MinceR: a software is free software if it guarantees the four freedoms |
| 18:23.30 | *** join/#devuan ppisati_ (~ppisati@2-235-152-247.ip228.fastwebnet.it) |
| 18:23.35 | MinceR | if i distributed binaries with sources generated by decompiling the binaries and no documentation as public domain, you'd insist it's free software? |
| 18:23.42 | muep | if it is technical limitations that you are allowed to change, it is free software |
| 18:23.45 | KatolaZ | namely, the freedom to use it, modify it, distribute it, and redistribute the modified versions |
| 18:23.46 | MinceR | well, systemd does not guarantee the four freedoms |
| 18:23.48 | KatolaZ | fullstop |
| 18:23.55 | KatolaZ | and that's a lie, MinceR |
| 18:23.57 | MinceR | indeed, it tries to infringe on some of them |
| 18:24.16 | KatolaZ | but it is still distributed with a license which guarantees those freedoms |
| 18:24.24 | muep | it does not infringe any more than e.g. the iproute2 package that uses linux specific interfaces |
| 18:24.25 | KatolaZ | so it *is* free software |
| 18:24.29 | MinceR | it does not guarantee those freedoms, as it can't |
| 18:24.43 | KatolaZ | MinceR: please tell me how |
| 18:24.48 | muep | FSD is not a guarantee of compatibility with anything |
| 18:24.51 | KatolaZ | is systemd preventing you from using it? |
| 18:24.59 | KatolaZ | ofr to distroibute it? |
| 18:25.02 | KatolaZ | or to modify it? |
| 18:25.07 | KatolaZ | or to redistribute the modifications? |
| 18:25.20 | MinceR | if i tried to use systemd, it would prevent me from using a real init system, and it would also prevent me from using another syslog daemon |
| 18:25.30 | KatolaZ | MinceR: that is *another* story |
| 18:25.35 | KatolaZ | is systemd preventing you from using it? |
| 18:25.35 | MinceR | it would insist on mangling log entries before optionally passing them on to syslogd |
| 18:25.38 | MinceR | it's the same story |
| 18:25.39 | KatolaZ | ofr to distroibute it? |
| 18:25.42 | KatolaZ | or to modify it? |
| 18:25.47 | KatolaZ | or to redistribute the modifications? |
| 18:25.59 | MinceR | they're also trying to prevent me from using a growing set of other software without systemd |
| 18:25.59 | KatolaZ | these are the freedoms which define free software |
| 18:26.09 | MinceR | saddling me with additional gratuitous restrictions |
| 18:26.10 | KatolaZ | that's *other* software |
| 18:26.17 | MinceR | which would have been less work to add |
| 18:26.21 | KatolaZ | and free software does not have to allow you to use *other* software |
| 18:26.28 | KatolaZ | it is nowhere in the FSD |
| 18:26.30 | MinceR | doing all this properly would actually have been easier for them |
| 18:26.32 | KatolaZ | *nowhere* |
| 18:26.56 | MinceR | thus systemd (and all its components), gnome3, kde5, etc. include measures that infringe on my freedom to use them as i want to |
| 18:27.04 | MinceR | which are not technically justified |
| 18:27.19 | KatolaZ | MinceR: all free software comes with ABSOLUTELY NON WARRANTY |
| 18:27.23 | MinceR | thus, it seems preposterous to me to claim that these pieces of software support and enhance my freedoms |
| 18:27.26 | MinceR | irrelevant |
| 18:27.37 | KatolaZ | MinceR: it is not about *your* *freedom* |
| 18:27.42 | KatolaZ | that's *another* story |
| 18:27.55 | KatolaZ | free software is about *the* *four* *freedoms* defined in the FSD |
| 18:27.58 | KatolaZ | fullstop |
| 18:28.25 | KatolaZ | then, nothing avoids that free software can hinder *other* freedoms |
| 18:28.33 | *** join/#devuan fsmithred (~fsmithred@68-184-46-18.dhcp.oxfr.ma.charter.com) |
| 18:28.35 | MinceR | yes, including the freedom to use and to modify |
| 18:28.39 | KatolaZ | but this does not automatically make it non-free software |
| 18:28.45 | KatolaZ | that's a lie MinceR |
| 18:28.49 | KatolaZ | you can use systemd |
| 18:28.52 | MinceR | though i suppose you believe that source code obfuscation does not impede modification either |
| 18:28.53 | KatolaZ | whereven yu want |
| 18:29.02 | MinceR | yes, if i give up control on my computer, i can use systemd |
| 18:29.04 | KatolaZ | and you have the source code, so you can modify it |
| 18:29.08 | KatolaZ | yes MinceR |
| 18:29.09 | fsmithred | MinceR, if you don't like systemd, you can re-write it to your liking. You can't do that with Internet Explorer. |
| 18:29.14 | KatolaZ | if you decide to use it |
| 18:29.15 | MinceR | also, you keep saying i "lie" |
| 18:29.20 | MinceR | i suggest you look up what lying means |
| 18:29.21 | KatolaZ | yes |
| 18:29.33 | MinceR | fsmithred: sure i can |
| 18:29.45 | fsmithred | and redistribute it? |
| 18:29.53 | MinceR | fsmithred: i can decompile it, and the result of that is apparently as good as un-obfuscated, properly written source code |
| 18:30.01 | MinceR | s/written/documented/ |
| 18:30.03 | KatolaZ | MinceR: you have the code of systemd |
| 18:30.08 | MinceR | yes, and it's crap |
| 18:30.12 | KatolaZ | you don't need to de-compile nothing |
| 18:30.17 | KatolaZ | that's another storyu |
| 18:30.21 | MinceR | and i also have the code of gcc, which you said was partially obfuscated |
| 18:30.25 | KatolaZ | the FSD is not about code quality |
| 18:30.36 | MinceR | the FSD is about freedoms |
| 18:30.40 | KatolaZ | anyway, this is not leading anywhere :) |
| 18:30.42 | MinceR | which systemd is designed to encroach on |
| 18:30.51 | MinceR | oh, i think it's leading somewhere |
| 18:30.53 | KatolaZ | you hate systemd, and I totally agree on that |
| 18:31.10 | KatolaZ | but you can't say that it is not free software, because that is simply not true |
| 18:31.11 | MinceR | to the insistence that the presence of source code is immaterial |
| 18:31.15 | MinceR | and that copyleft is pointless |
| 18:31.30 | MinceR | (which it may be, as there are many ways to circumvent it, not just obfuscation...) |
| 18:31.33 | KatolaZ | it is a free software which can hinder some of your freedoms, still respecting the 4 freedoms defined in the FSD |
| 18:31.52 | MinceR | sorry, i can't truthfully state that systemd supports my freedoms |
| 18:32.01 | KatolaZ | I am not saying that |
| 18:32.02 | MinceR | or that it respects my freedoms |
| 18:32.11 | KatolaZ | it respects the 4 freedoms stated in the FSD |
| 18:32.16 | KatolaZ | which make it free software |
| 18:32.17 | KatolaZ | fullstop |
| 18:32.37 | KatolaZ | then I agree that systemd hinders *other* freedoms |
| 18:32.40 | KatolaZ | obviously |
| 18:32.46 | KatolaZ | but not the 4 ones defined in the FSD |
| 18:32.57 | KatolaZ | it has a lot of other problems |
| 18:33.11 | KatolaZ | techinical, phylosophical, and ethical |
| 18:33.19 | KatolaZ | but it is free softwaree |
| 18:33.23 | KatolaZ | adhering to the FSD |
| 18:33.32 | KatolaZ | this does not make it good |
| 18:33.35 | KatolaZ | or bad |
| 18:33.38 | KatolaZ | it's just a fact |
| 18:34.11 | MinceR | so a piece of software that was designed to limit my freedoms is "free software" |
| 18:34.22 | MinceR | makes the term "free software" pretty useless |
| 18:34.33 | KatolaZ | if it respects the 4 freedoms defined in the FSD, it is free software, yes |
| 18:34.42 | KatolaZ | the term free software is defined in the FSD |
| 18:34.45 | KatolaZ | as you said above |
| 18:35.30 | KatolaZ | free software is a softeware that you can use for any purpose, distribute, modify, and redistribute in the modified versions as you like |
| 18:35.35 | KatolaZ | this is what the FSD says |
| 18:35.39 | KatolaZ | nothing more |
| 18:35.41 | KatolaZ | nothing less |
| 18:35.51 | MinceR | even if we accept that respecting the freedom to use an alleged "init system" or "OS building block" is preserved even if it prevents proper usage of the computer we run it on... |
| 18:35.57 | MinceR | what about the freedom to modify systemd? |
| 18:36.13 | KatolaZ | you have the code |
| 18:36.18 | KatolaZ | you read and study it |
| 18:36.19 | MinceR | binary is also code |
| 18:36.23 | KatolaZ | and you modify |
| 18:36.31 | KatolaZ | MinceR: you have the source code of systemd |
| 18:36.36 | KatolaZ | it is under the GPL |
| 18:36.42 | muep | you have the same source code that the original authors use |
| 18:36.43 | MinceR | strange that the FSF (and even the OSI) would insist on source code if source code is after all meaningless |
| 18:37.02 | KatolaZ | MinceR: are we putting up a process to thre FSF here? |
| 18:37.11 | KatolaZ | :) |
| 18:37.20 | MinceR | i'm just trying to find out if these terms have meaning at all anymore |
| 18:37.21 | KatolaZ | you might dislike the FSD |
| 18:37.31 | KatolaZ | MinceR: free software is defined in the FSD |
| 18:37.33 | MinceR | so if the original authors have lost the source code in a HDD crash, does that make binary-only distribution acceptable? |
| 18:37.40 | MinceR | apparently poorly defined |
| 18:37.42 | KatolaZ | nope MinceR |
| 18:37.49 | muep | does devuan have its own definition, like debian has its DFSG that disagrees with the FSF FSD on some corner cases? |
| 18:38.01 | MinceR | afaik the DFSG is almost or entirely identical to the OSD |
| 18:38.12 | MinceR | including the dependency on the license |
| 18:38.19 | KatolaZ | muep: the DFSG is almost entirely identical to the FSF, actually |
| 18:38.29 | KatolaZ | ~FSD |
| 18:38.42 | muep | KatolaZ: at least Emacs documentation was treated differently when I last checked |
| 18:38.50 | KatolaZ | and gcc documentation as well |
| 18:39.06 | muep | but yes, almost identical except for the corner cases |
| 18:39.07 | KatolaZ | the corner cases regard mostly the use of the FDL |
| 18:39.23 | KatolaZ | which prevents modifications, in some cases |
| 18:39.38 | KatolaZ | actually, the DFSG is more radical than the FSD, in some respects |
| 18:39.41 | MinceR | compare https://www.debian.org/social_contract https://opensource.org/definition https://www.gnu.org/philosophy/free-sw.en.html |
| 18:39.53 | KatolaZ | MinceR: the social contract is not the DFSG |
| 18:40.02 | KatolaZ | you are mixing two different things |
| 18:40.06 | MinceR | the DFSG is under the social contract |
| 18:40.09 | MinceR | on the same page |
| 18:40.23 | KatolaZ | I am talking about the DFSG, not the social contract |
| 18:40.40 | KatolaZ | that's a different document |
| 18:40.55 | MinceR | the only difference between the DFSG and the OSD i've spotted so far is "Specific to Debian" vs "Specific to a Product" |
| 18:40.58 | KatolaZ | which regulates the relationship between Debian and its usaers |
| 18:41.18 | MinceR | https://www.debian.org/social_contract#guidelines |
| 18:41.22 | KatolaZ | ok MinceR, what is the point? |
| 18:41.29 | MinceR | literally the same page |
| 18:42.25 | MinceR | the point is that what the DFSG/OSD specify is about "open source", not "free software" |
| 18:42.34 | KatolaZ | point 1, 2, and 3 is the definition of free software |
| 18:42.36 | KatolaZ | as in the FSD |
| 18:43.18 | MinceR | which is apparently meaningless |
| 18:43.33 | KatolaZ | why? |
| 18:43.55 | KatolaZ | you see it as meaningless |
| 18:43.56 | MinceR | because apparently the presence of source code that is actually practically usable for development and modification is irrelevant |
| 18:44.05 | KatolaZ | ok |
| 18:44.05 | muep | being free software is just one criteria that you should apply when you decide if some software is desirable for you or not |
| 18:44.25 | KatolaZ | then if you refuse the definition of free software, what are we talking about? |
| 18:44.26 | MinceR | you could take a binary, decompile it, run an obfuscator on the result, and the result being free software (according to you) would be a mere question of licensing |
| 18:44.31 | MinceR | i refuse it? |
| 18:44.33 | MinceR | interesting |
| 18:44.39 | KatolaZ | you said it's useless |
| 18:44.53 | KatolaZ | I respectfully disagree |
| 18:44.55 | MinceR | well yes, you've managed to twist the definition into something useless |
| 18:45.00 | KatolaZ | uh? |
| 18:45.02 | KatolaZ | please |
| 18:45.08 | KatolaZ | go and read the original FSD |
| 18:45.14 | KatolaZ | I haven't twisted anything |
| 18:45.19 | MinceR | i've read it many times |
| 18:45.34 | muep | he's not twisting it at all. at least to me, it seems pretty well in line with how it is commonly understood |
| 18:45.41 | KatolaZ | https://www.gnu.org/philosophy/free-sw.en.html |
| 18:45.57 | MinceR | well, there are many wrong things being "commonly understood" |
| 18:46.01 | KatolaZ | I am not interpreting anything |
| 18:46.08 | MinceR | like backdoored, locked-down, closed platforms being "hip" and "cool" |
| 18:46.16 | KatolaZ | I am reporting what is written in the FSD |
| 18:46.41 | MinceR | The freedom to study how the program works, and change it so it does your computing as you wish (freedom 1). Access to the source code is a precondition for this. |
| 18:46.52 | MinceR | how does obfuscated source code help you study how the program works? |
| 18:46.54 | KatolaZ | âFree softwareâ means software that respects users' freedom and community. Roughly, it means that the users have the freedom to run, copy, distribute, study, change and improve the software. |
| 18:47.16 | muep | obfuscated code would not typically qualify for the condition of having source code |
| 18:47.24 | KatolaZ | MinceR: where is this obfuscated code? |
| 18:47.31 | MinceR | in gcc, according to you |
| 18:47.48 | KatolaZ | I used "obfuscated" within doouble qoutes above |
| 18:47.52 | KatolaZ | it is not obfuscated at all |
| 18:47.59 | KatolaZ | it is a pretty complicated piece of softweae |
| 18:48.01 | MinceR | and from what i've seen, systemd's source code isn't exactly friendly to a reader's attempts to understand it |
| 18:48.09 | MinceR | considering how it's an undocumented mess |
| 18:48.27 | KatolaZ | MinceR: the same is valid for 90% of the free software out there |
| 18:48.30 | MinceR | which i wouldn't particularly mind if they weren't trying to make everything depend on it |
| 18:48.31 | KatolaZ | not just for systemd |
| 18:48.42 | KatolaZ | there is plenty of code that is a mess |
| 18:48.46 | MinceR | 90% of the free software out there does not attempt to determine what you're allowed to run on your computer |
| 18:48.52 | KatolaZ | badly-documented |
| 18:49.03 | MinceR | 90% of the free software out there doesn't try to force you to give up control over your own computers |
| 18:49.06 | KatolaZ | MinceR: just don't use systemd |
| 18:49.07 | muep | systemd is a large C program so it is not especially easy to understand, but being especially easy is not a requirement for fitting the FSD |
| 18:49.17 | MinceR | not using systemd makes systemd free software? |
| 18:49.21 | KatolaZ | but again, you can't say that it is not free software |
| 18:49.27 | MinceR | i still don't know where this leaves gcc |
| 18:49.46 | muep | if you can find the obfuscated bit, you could point at that and ask FSF for a clarification |
| 18:49.56 | KatolaZ | because anybdy who have read the FSD will say the contrary, and will be right |
| 18:50.19 | muep | I'd expect that there currently is no code in GCC that is clearly obfuscated |
| 18:50.26 | MinceR | i hope so |
| 18:50.26 | KatolaZ | obviously MinceR |
| 18:50.29 | KatolaZ | ~muep |
| 18:50.42 | MinceR | i don't see what they'd expect to gain from it now |
| 18:50.43 | KatolaZ | there *used* to be a tiny portion of very complicated code |
| 18:50.55 | KatolaZ | which was complicated on itself |
| 18:51.05 | KatolaZ | and left complicated for a while |
| 18:51.21 | KatolaZ | to avoid those algorithms to be stolen and included in closed-source compilers |
| 18:51.26 | KatolaZ | being a complicated piece of code |
| 18:51.39 | KatolaZ | it would have been easier to recognise a simple copy-paste |
| 18:51.41 | KatolaZ | fullstop |
| 18:51.46 | MinceR | "stealing" algorithms was one important part of "studying how the program works" |
| 18:51.48 | KatolaZ | it happened 30 years abo |
| 18:52.18 | MinceR | so it seems that they did not only not care about the 4 freedoms when systemd was forced on debian, they actually never did |
| 18:52.47 | KatolaZ | MinceR: none of the 4 freedoms has beed infringed by systemd |
| 18:53.07 | KatolaZ | ~been |
| 18:53.19 | KatolaZ | you keep repeating that |
| 18:53.41 | KatolaZ | but you have provided no evidence that systemd has infringed any of the 4 freedoms specified in the FSD |
| 18:53.51 | MinceR | so you keep insisting |
| 18:54.03 | KatolaZ | (oh lord, *I* have to defend systemd? :D) |
| 18:54.10 | MinceR | no, you don't have to |
| 18:54.16 | KatolaZ | yes, I do |
| 18:54.24 | KatolaZ | because I hate false statements |
| 18:54.25 | MinceR | ratcrap has enough money to do so |
| 18:54.42 | KatolaZ | and if a statement is false about systemd, is that it is not free software |
| 18:54.44 | muep | well you can just defend the facts while saying at the same time that systemd has other undersirable aspects even if it is free software |
| 18:54.58 | KatolaZ | muep: and it is exactly what I have done :D |
| 18:55.02 | MinceR | well, i hate having malware designed to take over my computers paraded around as paragons of free software |
| 18:55.20 | KatolaZ | that's another story MinceR |
| 18:55.32 | KatolaZ | you hate software that does things that you don't like |
| 18:55.38 | KatolaZ | I hate GNOME for the same reason |
| 18:55.39 | KatolaZ | and KDE |
| 18:55.48 | KatolaZ | but this does not make them non-free softwaere |
| 18:55.51 | KatolaZ | :) |
| 18:55.54 | muep | you can point at those other bad aspects rather than using arguments that a typical free-software-aware person will just find false |
| 18:56.10 | KatolaZ | exactly muep |
| 18:56.45 | KatolaZ | GNOME is another enormous hairball |
| 18:56.53 | MinceR | well, it seems to me that gratuitous dependencies, undocumented code and secret APIs infringe on my freedoms to use and modify software |
| 18:57.03 | KatolaZ | which is quite hard to study, understand, and modify |
| 18:57.16 | KatolaZ | but I can't say "oh well, GNOME is not free software" |
| 18:57.24 | MinceR | and the rest is just twisting words to try and make systemd, gnome3 and kde5 look good |
| 18:57.46 | KatolaZ | MinceR: *me* wanting systemd to look good? :D |
| 18:57.53 | MinceR | ¯\_(ã)_/¯ |
| 18:57.54 | KatolaZ | you hit the wrong target, my friend :) |
| 18:58.31 | KatolaZ | I like to stick with reality, and reality tells that systemd is a ugly piece of bloated free software |
| 18:58.48 | KatolaZ | which goes against the basic principles of the unix philosophy |
| 18:59.04 | MinceR | reality also tells that the FSF is shying away from doing its job |
| 18:59.14 | KatolaZ | and whose community has very little to do with the spirit that has motivated the GNU/Linux comunity so far |
| 18:59.28 | KatolaZ | MinceR: what should be its job, then? |
| 18:59.52 | muep | FSF is not strictly about unix |
| 18:59.53 | KatolaZ | fighting GNOME because it is a bloated mess as well? |
| 19:00.10 | muep | or even technical software quality |
| 19:00.11 | MinceR | more because it is yet another mess that is designed to force the users' hands |
| 19:00.24 | MinceR | and more importantly, it depends on systemd |
| 19:00.29 | KatolaZ | MinceR: you can use GNOME, or you can use something else.... |
| 19:00.36 | MinceR | it's yet another tool to force systemd on users |
| 19:00.36 | KatolaZ | again |
| 19:00.47 | golinux | WHEW! i go to have some b'fast and the channel is clogged with this 'debate'. Shouldn't this have been on #debianfork? |
| 19:00.56 | KatolaZ | ooops! |
| 19:00.59 | MinceR | yeah, maybe |
| 19:01.09 | golinux | No kidding . . . |
| 19:01.14 | KatolaZ | runs behind MinceR |
| 19:01.21 | KatolaZ | you're right golinux |
| 19:01.23 | KatolaZ | sorry |
| 19:01.24 | KatolaZ | :( |
| 19:01.39 | golinux | Shit happens. :) |
| 19:01.45 | KatolaZ | btw, going back to testing setnet with EAP-TLS |
| 19:01.51 | MinceR | i'm not quite sure where the line was between 'making devuan eventually support systemd as an option' and 'systemd is [not] free software' :) |
| 19:02.14 | KatolaZ | MinceR: I don't think that Devuan should support systemd |
| 19:02.18 | golinux | Yeah . . . conversations morph . . . |
| 19:02.33 | MinceR | neither do i, but also, it doesn't even seem feasible to me |
| 19:02.38 | fsmithred | devuan with systemd is debian |
| 19:02.42 | KatolaZ | MinceR: you mentioned that according to you systemd is not free software |
| 19:02.42 | golinux | KatolaZ: +1 |
| 19:02.46 | MinceR | afaik the whole reason it didn't was the massive amount of extra work it would have entailed |
| 19:02.59 | MinceR | which is almost certainly not a coincidence |
| 19:03.17 | golinux | OMG! Here we go again . . . |
| 19:03.24 | KatolaZ | nope, I am leaving now |
| 19:03.31 | KatolaZ | it seems like I am here, but I am not! |
| 19:03.32 | KatolaZ | :D |
| 19:03.33 | KatolaZ | :P |
| 19:03.35 | MinceR | and i think i'm done |
| 19:03.48 | KatolaZ | peace, love, and free software MinceR |
| 19:03.51 | MinceR | yeah |
| 19:03.57 | golinux | Group hug? |
| 19:04.08 | KatolaZ | it's great that we can argue and disagree |
| 19:04.11 | KatolaZ | freely |
| 19:04.14 | KatolaZ | ;) |
| 19:05.28 | *** join/#devuan edbarx (~edbarx@78.133.22.227) |
| 19:46.21 | *** join/#devuan Weeezy (Elite18779@gateway/shell/elitebnc/x-hdqxnqmjluwlocmw) |
| 20:17.31 | *** join/#devuan Hoshpak (~Hoshpak@p2003005B4B1A7F00B53E5FFB5F225050.dip0.t-ipconnect.de) |
| 20:26.55 | *** join/#devuan cdanderson42 (~nuhrin@c-50-170-120-18.hsd1.wa.comcast.net) |
| 20:42.50 | *** join/#devuan cirucito (~Cirucito@230.104.218.87.dynamic.jazztel.es) |
| 20:49.09 | *** part/#devuan cirucito (~Cirucito@230.104.218.87.dynamic.jazztel.es) |
| 21:03.33 | *** join/#devuan Cirucito (~Cirucito@230.104.218.87.dynamic.jazztel.es) |
| 21:06.28 | *** join/#devuan windowsrefund (~windowsre@unaffiliated/windowsrefund) |
| 21:10.04 | *** part/#devuan Cirucito (~Cirucito@230.104.218.87.dynamic.jazztel.es) |
| 21:12.21 | *** join/#devuan markizano (markizano@2600:3c00::f03c:91ff:fec8:382d) |
| 21:12.31 | markizano | I need to get some bead eyes and a board to put outside my bedroom window |
| 21:12.42 | markizano | (o_o) "Big Brother is Watching Your Every Move" |
| 21:12.46 | markizano | lol |
| 21:12.49 | Martin89 | :D |
| 21:13.03 | markizano | I'm in the middle of my complex, and I can see everyone had their blinds and doors open yest |
| 21:13.10 | markizano | they realized someone moved into this unit, now everyone's shit is closed |
| 21:13.28 | markizano | lol, let's just add to the paranoia and remind them of just who is watching them... |
| 21:14.37 | markizano | wow.. all that totally went to the wrong channel >.< |
| 21:14.38 | furrywolf | lol |
| 21:14.44 | markizano | sorry ya'll. .. |
| 21:28.34 | *** join/#devuan Chanku (~Chanku@c-68-52-188-31.hsd1.tn.comcast.net) |
| 21:49.05 | *** join/#devuan ff_ (~ff_@dui58.neoplus.adsl.tpnet.pl) |
| 21:50.07 | *** join/#devuan Hoshpak_ (~Hoshpak@p2003005B4B1A7F00004109FFFE40D1E1.dip0.t-ipconnect.de) |
| 21:51.06 | *** join/#devuan chomwitt3 (~chomwitt@ppp-94-69-222-21.home.otenet.gr) |
| 21:51.12 | *** join/#devuan Fervi (~fervi@81.219.6.206) |
| 21:53.07 | Fervi | Hello, I think about writing an article about devu. I wonder if there are any projects yet but systemd remove and replace pulseaudio by Alsa (and various library emulation Pulseaudio) |
| 21:53.45 | *** join/#devuan chomwitt4 (~chomwitt@ppp-94-67-203-112.home.otenet.gr) |
| 21:56.35 | Weeezy | we'll take it markizano, it's too quiet right now. |
| 21:57.10 | markizano | lol, sure :) |
| 22:05.37 | *** join/#devuan thehornet (~a0225280@192.91.101.31) |
| 22:27.38 | *** join/#devuan Kruppt (~Kirk_Krup@104.169.24.171) |
| 22:31.59 | *** join/#devuan NewGnuGuy (~david@69.41.81.178) |
| 22:43.03 | *** join/#devuan blueness (~blueness@gentoo/developer/blueness) |
| 22:48.18 | *** join/#devuan pydsigner (~pydsigner@unaffiliated/pydsigner) |
| 22:52.06 | *** join/#devuan unknown__ (~chatzilla@193-80-227-39.adsl.highway.telekom.at) |
| 23:04.05 | *** join/#devuan zocker (~tobias@x86.li) |
| 23:11.02 | *** join/#devuan TemporalBeing1 (~Ben_Meyer@172-6-231-63.lightspeed.tukrga.sbcglobal.net) |
| 23:24.56 | *** join/#devuan clopez (~tau@neutrino.es) |
| 23:24.56 | *** join/#devuan Bryanstein (~Bryanstei@shellium/admin/bryanstein) |
| 23:24.56 | *** join/#devuan jotik (~7f000001@unaffiliated/jotik) |
| 23:24.56 | *** join/#devuan uidzer0 (~uidzer0@168.235.65.160) |
| 23:24.56 | *** join/#devuan rleigh (~rleigh@unaffiliated/rleigh) |
| 23:24.56 | *** join/#devuan Centurion_Dan (~Icedove@office.centurion.net.nz) |
| 23:24.56 | *** join/#devuan OmegaPhil (~kvirc@fsf/member/OmegaPhil) |
| 23:28.41 | *** join/#devuan clopez (~tau@neutrino.es) |
| 23:44.13 | *** join/#devuan Xenguy (~Xenguy@206-47-113-63.dsl.ncf.ca) |
| 23:52.14 | *** join/#devuan AlexLikeRock (~AlexLikeR@unaffiliated/alexlikerock) |
| 23:52.53 | AlexLikeRock | wohaaaa! we are 174 :-> |
| 23:53.06 | Fervi | and 170 afkers xd |
| 23:53.13 | AlexLikeRock | good afternoon |
| 23:53.22 | Fervi | hi |
