DocScrutinizer05 has quit [Disconnected by services]
DocScrutinizer06 has joined #qi-hardware
DocScrutinizer06 is now known as DocScrutinizer05
guanucoluis has joined #qi-hardware
guanucoluis has quit [Ping timeout: 245 seconds]
DocAvalanche has joined #qi-hardware
DocScrutinizer51 has quit [Ping timeout: 258 seconds]
DocAvalanche is now known as DocScrutinizer51
lkcl has quit [Ping timeout: 258 seconds]
lkcl has joined #qi-hardware
wolfspraul has quit [Ping timeout: 245 seconds]
wolfspraul has joined #qi-hardware
pcercuei has quit [Quit: dodo]
wolfspraul has quit [Ping timeout: 260 seconds]
wolfspraul has joined #qi-hardware
<wpwrak>
grmbl. one of these oscilloscope-instead-of-printk moments again ...
<wpwrak>
first hint of trouble: zero activity on the I/O pins. yes, that would explain a thing or two ...
<coyo>
question for the channel: what is the scope of this community? would a do-it-yourself molten thorium floride fast breeder reactor fall under this community's scope?
<coyo>
in other words, would schematics and instructions be welcome here?
<wpwrak>
we're more into electronics than making friends with the nuclear power authorities. but i have to admit that the idea is intriguing ;-)
<coyo>
XD
<wpwrak>
i could imagine that the raw materials are a bit hard to find ?
<coyo>
well, there's nothing illegal about thorium. it may be theoretically possible to construct thermonuclear warheads with thorium by breeding it into uranium, but it's such a pain in the butt, and i have no need for those anyway
<coyo>
i just want power for my tesla coils :D
<coyo>
without the power company yelling at me about blacking out half the state
<coyo>
what you do is "invest" in thorium ore, and smelt and refine it yourself, then induce the initiating of reaction with a beta particle accelerator to provide the initial high energy neutrons without needing a crumb of uranium
<coyo>
let me doublecheck that really fast, hold on
<wpwrak>
oh, i think you can already have a lot of fun with a dirty bomb. that shouldn't be all that hard to make.
<coyo>
whoops
<coyo>
alpha particles
<coyo>
i meant alpha particles
<wpwrak>
of course, if you have a few uranium bricks at hand, that'll help
<wpwrak>
seems that the best way to jump-start one of your reactors is if you have some decommissioned nuke to scavenge
roh_ has joined #qi-hardware
roh has quit [Ping timeout: 276 seconds]
kanzure has quit [Remote host closed the connection]
kanzure has joined #qi-hardware
<coyo>
wpwrak: i dont think so
<coyo>
besides
<coyo>
they arent me
<coyo>
:D
<coyo>
grabbing fissile material from a decommissioned nuke is a good way to get on every list out there
<coyo>
i'm sure there are other ways to do it successfully
<coyo>
without attracting undue attention
<coyo>
wpwrak: however, am i to take it that the schematics, how to videos, diagrams, and detailed instructions would not be unwelcome in this community?
<coyo>
i'm going to assume diy firearms are clearly unwelcome here. i cant find a tos on qi-hardware.com but it's pretty safe to assume.
<coyo>
:P
<wpwrak>
well, someone may find schematics entertaining :) but then, we're not really all that much into irradiating ourselves
<wpwrak>
maybe lekernel could steer you into a better direction once he's back. he's had some wacky talks at his conference in december
<coyo>
alright
jekhor has joined #qi-hardware
<coyo>
according to this energy company saleswoman, they dont carry micro nuclear reactors, however, they do have large scale diesel generators with evaporative cooling systems that could provide enough sustained power for a city block or two
<coyo>
i may go that route
<coyo>
diy diesel generator, and diy syndiesel (aka biodiesel)
<coyo>
so you never have to pay another cent for arabian dinosaur corpse juice
jluis has joined #qi-hardware
<wpwrak>
or just build a little rocket and shoot your experiment into the sun ? plenty of power there.
<wpwrak>
you'll also be able to put all your experience in radiation shielding to good use :)
<coyo>
pfft
xiangfu has joined #qi-hardware
<coyo>
y so srs?
<wpwrak>
xiangfu: heya ! a question: do you know how 0020-qi_lb60-NAND-add-data-partition.patch is doing on its way to upstream ?
xiangfu has quit [Ping timeout: 264 seconds]
xiangfu has joined #qi-hardware
Guest19240_ has quit [Ping timeout: 264 seconds]
<kyak>
wpwrak: thanks! i'll try it as soon as i can. Does anything need to be done on atben side? I figured you were using a newer kernel on ben
coyo has quit [Remote host closed the connection]
<kyak>
yeah, there is no good reason why the default partition layout is not upstream.. It's not like there are many modifications of Ben and the upstream holds the generic configuration..
<kyak>
brb
kyak has quit []
kyak has joined #qi-hardware
kyak has quit [Changing host]
kyak has joined #qi-hardware
<wpwrak>
kyak: atben doens't work yet with the new kernel. i get no gpio activity at all.
<kyak>
good :) can use the the default kernel then
<kyak>
wpwrak: out of your kernel tree, can i just extract drivers/net/ieee802154/atusb.c and compile it with 3.9 tree? Or did you do some other changes?
<kyak>
i see there are some changes to at86rf230.c.. So i'd need to recompile this module as well
Ornotermes has quit [Remote host closed the connection]
unclouded has joined #qi-hardware
jekhor has quit [Ping timeout: 240 seconds]
Ornotermes has joined #qi-hardware
<wpwrak>
oh, you don't need it for atusb
<wpwrak>
you need atusb.c, atusb.h, and at86rf230.h. that should be all.
<wpwrak>
btw, my kernel is 3.9.0-rc5+. so if your aim is 3.9, you may just use that one
<wpwrak>
(not the branch in qi-kernel. that one is obsolete.)
<kyak>
ok, i'll try with 3.8.5 which is my current kernel. If it doesn't work, i'll try with 3.9
<viric>
kyak: do you manage your workstation kernel manually?
<kyak>
viric: no, of course not.. It's arch linux who manages it for me
<kyak>
i have created the dkms-atusb package however
<viric>
ok
<kyak>
which is how i plan to work with atusb
<viric>
what is 'dkms'? :)
<kyak>
it's a special software that manages the list of external modules and recomiles it automatically on kernel version update
<viric>
wpwrak: do you make the atusb module build outside the kernel tree? or is it a patch to the kernel?
<viric>
kyak: ah ok
<kyak>
dynamic kernel management system, i guess
<wpwrak>
oh, i haven't even tried to build it as a module
<kyak>
wpwrak: ah! that's the key point :)
<wpwrak>
yeah, sorry. should have mentioned that :)
<wpwrak>
it wouldn't be surprised if it actually worked fine as a module. but then maybe it doesn't
<viric>
the linux trend is that things work as a module but not in-kernel :)
<wpwrak>
i try to stay ahead of the trend ;-)
<wpwrak>
besides, my test machine boots really fast
<wpwrak>
has a heavily gutted ubuntu that starts almost nothing
<kyak>
my laptop usually doesn't reboot, except for kernel updates and "reboot will fix it" issues
<wpwrak>
well, during krnel development, "reboot will fix it" happens rather often :)
<wpwrak>
at least to the point where the machine will talk to you again :)
<kyak>
sometimes my keyboard would just stop working. I then use mouse to reboot :)
<viric>
wpwrak: don't you develop on qemu or so?
<kyak>
this hasn't happened in a while though
<kyak>
guess qemu won't suffice for atusb.. Or maybe it will?
Calyp has joined #qi-hardware
<whitequark>
wpwrak: heh. not sure how someone who can't tell alpha from beta particles could build an LFTR
* whitequark
tries to guess the age range
viric has quit [Ping timeout: 245 seconds]
<whitequark>
1990-1995. "y so srs". I think
viric has joined #qi-hardware
wolfspraul has quit [Quit: leaving]
security has joined #qi-hardware
megha has quit [Ping timeout: 248 seconds]
Guest19240_ has joined #qi-hardware
wolfspraul has joined #qi-hardware
<larsc>
stupid doesn't need an age ;)
wolfspraul has quit [Quit: leaving]
<whitequark>
larsc: sure, but in my experience, vocabulary tends to reflect one
LunaVorax has joined #qi-hardware
LunaVorax has quit [Read error: Connection reset by peer]
LunaVorax_ has joined #qi-hardware
LunaVorax_ has quit [Remote host closed the connection]
LunaVorax has joined #qi-hardware
<LunaVorax>
Hi!
unclouded has quit [Ping timeout: 245 seconds]
wolfspraul has joined #qi-hardware
<DocScrutinizer05>
wpwrak: everything dry?
rz2k has joined #qi-hardware
rz2k has joined #qi-hardware
rzk has joined #qi-hardware
rz2k has quit [Ping timeout: 264 seconds]
rz2k has joined #qi-hardware
rzk has quit [Ping timeout: 264 seconds]
jekhor has joined #qi-hardware
<wpwrak>
DocScrutinizer05: at the 14th floor ? yeah pretty much :)
<DocScrutinizer05>
:-)
<DocScrutinizer05>
a question to the real sysop hackers: how would you manage to swap the /home HDD in a life system? As in: add a HDD, cp /home to new HDD, mount new HDD home on top of existing ext3 /home as unionfs, cp again...???
<DocScrutinizer05>
I gather there's NO way to accomplish that?
security is now known as fire
<DocScrutinizer05>
another alternative approach: fuser -k SIGSTOP /home; dd /dev/sdb3 /dev/sdd3; <swap HDD sdb for HDD formerly known as sdd>; killall SIGCONT
<hellekin>
DocScrutinizer05: you could mount your new disk on /mnt, and rsync -a /home /mnt
<DocScrutinizer05>
:nod:
<hellekin>
of course, stopping processes using /home would help :)
<DocScrutinizer05>
but what do I do for the files that changed after rsync synced them?
<DocScrutinizer05>
yeah :-)
<hellekin>
login as root, fuser -k /home, rsync again :)
<roh_>
DocScrutinizer05: disable regular user logins. unmount..
<roh_>
or copy in a ramdisk and pivot_root
<roh_>
usually.. boot rescue system and do it from there
<DocScrutinizer05>
the whole point is I don't want to logout, since that would e.g. stop this irc client and make me basically lose my queries
<DocScrutinizer05>
;-)
<DocScrutinizer05>
pondering suspend to disk
roh_ is now known as roh
<DocScrutinizer05>
but that's probably also not really working
<roh>
DocScrutinizer05: well. you lost. changing a fs cannot happen without getting rid of processes.
<DocScrutinizer05>
I guess I will have to kill my uptime and do a reboot, accept some downtime for replacing this SATA HDD
<DocScrutinizer05>
roh: :nod:
<roh>
disks are .. well.. temporary storage. if you need more uptime. go raid
<DocScrutinizer05>
I guess my question is: how do I minimize that downtime, by doing backups in a proper way during regular use of PC
<DocScrutinizer05>
my idea been to get a new HDD, plug in, init s, dd, power down and swap HDD, boot
<DocScrutinizer05>
alas the shop requires me to bring in the defect HDD to give me a new one, which would mean several hours of PC being down since no proper /home existing
<DocScrutinizer05>
I might opt for buying a spare, do as sketched above, then bring in the defect HDD *after* PC already up and running on new one
<DocScrutinizer05>
will require two visits at shop, plus a couple of bucks for the new HDD spare
<DocScrutinizer05>
amazingly the "defect" HDD still kinda "works" - as in "no errors/problems during normal operation of PC, except for the onnoying TOCKTOCKTOCK it did during last few days every once in a while"
<DocScrutinizer05>
annoying even
pcercuei has joined #qi-hardware
* DocScrutinizer05
glances at ddrescue output
<hellekin>
it's just warning you it's gonna die suddenly. You shouldn't wait for it :P
<DocScrutinizer05>
I'm not waiting as you can tell from this very conversation ;-)
<DocScrutinizer05>
also the drive is a mere 50 days up&running, manufactured(!) Nov-2012
<DocScrutinizer05>
I bought it 50 days ago
<DocScrutinizer05>
heck, with a 2nd HDD I could create a RAID :-D
<DocScrutinizer05>
so same problem won't bite me again
<roh>
well.. i always buy the same hdd twice (for backups) or 3-4 times (raid1
* DocScrutinizer05
checks how much they call for TOSHIBA DT01ACA100 today
<lindi->
DocScrutinizer05: hibernate should take less than 10 seconds, even tcp connections can survive that
<lindi->
DocScrutinizer05: but if you want to plan for this in the future then LVM with hot-pluggable disks should work
<DocScrutinizer05>
lindi-: my idea is not to hibernate but to suspend to disk, then dd-clone the whole friggin HDD, swap, resume
<lindi->
DocScrutinizer05: I thought "hibernate" and "suspend to disk" meant the same thing
<DocScrutinizer05>
but I guess suspend to disk won't work, as usual
<DocScrutinizer05>
aaah
<lindi->
I use hibernate every day
<DocScrutinizer05>
yeah, if it works once, it works every time (unless your swap already been used up so much that it can't take the RAM content anymore)
<DocScrutinizer05>
since all my PCs/laptops during last 4 years seen 10 boots each at most, I hardly have done much tweaking to make StR work
<DocScrutinizer05>
314GB ddrescued
<wpwrak>
phew. atben finally responds. it was something trivial: in the past, the gpios were set by default to GPIO, so my driver never bothered to change them. now they're set to function. hence it could no longer talk to the device.
<DocScrutinizer05>
600 to go
<wpwrak>
the little things that become so obvious after a few hours of sleep :)
<DocScrutinizer05>
hehe
<wpwrak>
hellekin: heya ! welcome back ! all services connected again ? :)
<wpwrak>
DocScrutinizer05: he's the one you should worry about. if i understand things right, he's some 13 floors closer to the water - and in an area that did get flooded this time.
<DocScrutinizer05>
eeew
<DocScrutinizer05>
hellekin: so everything dry (again) at your site?
<DocScrutinizer05>
after swapping the first defect one for a replacement the both of them will form a nice raid
<hellekin>
wpwrak, DocScrutinizer05: all back to normal, less candles, one less PC :(
xdpirate has quit [Ping timeout: 245 seconds]
<hellekin>
but the PC didn't die because of water: the dust was still intact.
<hellekin>
Instead it must have died of the recurring power outage
<wpwrak>
hellekin: losing a pc sucks :-(
<wpwrak>
hellekin: maybe it's just the power supply ?
<hellekin>
maybe
<hellekin>
not sure what part died, the motherboard lights up, but the CPU fan does not start
<hellekin>
if you have a spare power supply, might be useful to test
<DocScrutinizer05>
either cpu fan (they tend to not start again once worn out and stopped), or your 12V of PSU are dead
<DocScrutinizer05>
you tried helping fan to start up, manually pushing it a bit?
<wpwrak>
hmm, no new ones. but i think i have one that's been removed as preventive maintenance. may have small issues, though. that pc had been acting up on some occasions.
<hellekin>
DocScrutinizer05: nope, testing now
<wpwrak>
reminds me that i'm out of spare supplies. they've turned into consumables lately, much like disks :-(
<hellekin>
I don't have any electrical tools here, sso I cannot do any tests
<hellekin>
wpwrak: oh really? that sucks too
emeb has joined #qi-hardware
<hellekin>
nope, the fan plays dead
<DocScrutinizer05>
yeah ddrescue just hit the bad sector range of HDD, now it's spitting errors and drive starts woodpecking again
DocScrutinizer05 has quit [Remote host closed the connection]
<kristianpaul>
hmm how flexible and fast would be implement regular expresions on a fpga
DocScrutinizer05 has joined #qi-hardware
<DocScrutinizer05>
WOW, that been scary
<viric>
to beat a cpu?
<DocScrutinizer05>
nah, HDD went TOCKTOCKTOCK for 15min
<DocScrutinizer05>
luckily it stopped a few minutes after I managed to send a ^C to ddrescue
<DocScrutinizer05>
already been afraid my complete dsktop would freeze
<viric>
are you recovering a hdD?
<DocScrutinizer05>
yep
<DocScrutinizer05>
of course it's BS to do a ddrescue from a mounted /home
<DocScrutinizer05>
but I thought it's probably better than nuttin
<DocScrutinizer05>
1800 hard headpark events, that can kill the nerves of a techhnically sensible person
<DocScrutinizer05>
anyway afk for getting a new drive
jekhor has quit [Ping timeout: 264 seconds]
<kristianpaul>
DocScrutinizer05: why scary?
<viric>
BS?
<viric>
I guess DocScrutinizer05 is close to loosing data.
DocScrutinizer05 has quit [Ping timeout: 264 seconds]
<kristianpaul>
ahh
DocScrutinizer05 has joined #qi-hardware
guanucoluis has joined #qi-hardware
<DocScrutinizer05>
why scary? because it is freezing my desktop, spamming my syslog with tons of kernel error msgs
<DocScrutinizer05>
and the sound is curling my toenails
xiangfu has quit [Quit: leaving]
<kristianpaul>
i have to replace the OS hardisk recently, seems there is hdd failure fever..
<kristianpaul>
:/
jekhor has joined #qi-hardware
<wpwrak>
my system disk went early last month ...
<wpwrak>
still need to try to salvage some files that are in no backup.
<viric>
here one disk died, 1,5 months ago.
<viric>
(not the oldest in the system, by far)
<wpwrak>
i see a pattern :)
<wpwrak>
if you have an old maxtor, treat it well, it'll probably outlast you :)
<viric>
haha
<viric>
3 months ago, another one died at the office
<wpwrak>
i wonder if it's time yet to move to SSD ...
wolfspraul has quit [Ping timeout: 240 seconds]
<wpwrak>
oh, and i found out why my external WD USB disks didn't work: they have a little voltage regulator on the interface board that gets so hot that it no only burns the board but apparently also the disk's electronics underneath that board. at least, even after removing the disk, it wouldn't work in a pc.
<viric>
I recently found the idle3 timer thing
<viric>
on wd green
<viric>
(together with the start-stop count :)
<wpwrak>
do they have a start-stop limit ? like some ink printers ?
<kristianpaul>
wpwrak: mine was maxtor yeap
<viric>
here I bought a WD green, because seagate didn't work at all in my computer. it looked like a broken hard drive to the system
<kristianpaul>
wpwrak: i was thinking same about ssd as well
<viric>
wpwrak: no, wd green have a timer that stops spinning every 12,8 seconds of innactivity.
<viric>
wpwrak: very good for energy saving (in some usage patterns), but can be also very degrading for the disk life
<kristianpaul>
wpwrak: those external hardisk also have some sha or similar encryption on the fly i remenber..
<viric>
the 'respin' counter gets very high in those drives
<viric>
and idle3-tool is the tool to disable that 12,8s timer. :)
<wpwrak>
oh, an idle time beyond the regular mechanism. clever :)
<wpwrak>
kristianpaul: a dead maxtor ? nearly impossible. what did you do to it ?
<kristianpaul>
wpwrak: nothin
<kristianpaul>
but kernel was flooding me with ATA sync errors
<kristianpaul>
then i died on sunday
<wpwrak>
that's what they all say ;-)
<wpwrak>
oh, you're a ghost now ?
<kristianpaul>
s/i/it
<qi-bot>
kristianpaul meant: "then it dited on sunday"
<kristianpaul>
lol
jluis has quit [Ping timeout: 245 seconds]
jekhor has quit [Ping timeout: 246 seconds]
<larsc>
Did you mean "Then I dyed it on sunday"?
<wpwrak>
that would explain a lot of things :)
<kristianpaul>
lol
jekhor has joined #qi-hardware
baba has joined #qi-hardware
fire has quit [Ping timeout: 264 seconds]
<kyak>
wpwrak: the atusb module builds and loads fine on 3.8.5
<wpwrak>
case 1: insmod then rmmod. case 2: insmod, bring it up and communicate, rmmod
<kyak>
haven't tried communicating yet.. let me unpack atben :)
<wpwrak>
i tried unplugging it, which seems to be fine. still have to check whether i'm leaking memory, though. i don't trust those urb removal functions.
<kyak>
okay, the kernel just paniced :)
Calyp has quit [Ping timeout: 245 seconds]
<kyak>
i didn't do any special, just wanted to open the page with instructions for dirtpan
<wpwrak>
did it say where it panicked ?
<kyak>
yeah, i was looking for something usb-related, but there was more like "tcp-write"
<wpwrak>
you mean "at86rf230.h" ? no, that's a copy. it doesn't exist in the original code. at some point in time i'll try to delete the original definitions
<kyak>
argh.. sounds like it's mac802154 recompilation...
<wpwrak>
ah, someone asked about qemu. dunno how good it is with usb. a prefer to run things on the bare metal :) have to admit that some oopses can be a bit annoying, when things scroll off the console. but then, it's easy enough to add a loop that stops the system after the interesting part of its complaints.
<viric>
no kgdb?
<wpwrak>
kyak: we do indeed not need the full MAC address
<wpwrak>
viric: i developed my workflow when kgdk and such were still banned :)
<wpwrak>
i sometimes use gdb on the live kernel, though (with /proc/kcore)
<wpwrak>
most of the time, i find that debuggers don
<wpwrak>
a few printks/printf often give you that information more rapidly. of course, then retrieving details costs more time
<wpwrak>
't really make finding the bug faster. they give you a lot of control, but at the same time you have to spend time on a lot more details