« first day (2932 days earlier)      last day (2384 days later) » 

01:50
Does anyone know how to determine where ElementaryOS is getting it's IP address from? I checked Network Manager, /etc/network/interfaces, and /etc/dhcp/dhclient.conf, but I'm still getting an old IP address
is it using network manager?
that has its own interface
02:54
hehe. I like cats :)
Goddammit, my Pentax DSLR indeed has a hardware problem: pentaxforums.com/forums/172-pentax-k-3/… (see latest reply to this thread)
> I have inspected the AF sub-mirror again and the spots have reappeared, exactly where they were before I tried to "clean" them". The camera has seen very little use over the last few months. At this point, I am absolutely convinced that this is delamination and not contamination.
!!/ultraheaddesk
so.. hardware eventually fails :(
I have serious doubts Precision Camera is able to fix this before my first photo assignment of the 2018-2019 season on September 1.
I'm going to give them a call tomorrow to try to get the case opened ASAP.
At this point, I am willing to spend the money on overnight shipping if it gets the camera back on time.
03:37
@JourneymanGeek Thanks, got that one
Now that I have the new modem/router combo installed, and reset all the IP addresses on my network... I can't SSH to my main server by IPv4, only IPv6
I feel I need a nap
@CanadianLuke ya, the commands/config files are listed there
always helps
I don't feel like learning IPv6 right now...
$ ssh [email protected] -vvv
OpenSSH_6.6.1, OpenSSL 1.0.1f 6 Jan 2014
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: /etc/ssh/ssh_config line 19: Applying options for *
debug2: ssh_connect: needpriv 0
debug1: Connecting to 10.0.0.4 [10.0.0.4] port 22.
debug1: Connection established.
debug1: identity file /home/luke/.ssh/id_rsa type -1
debug1: identity file /home/luke/.ssh/id_rsa-cert type -1
debug1: identity file /home/luke/.ssh/id_dsa type -1
debug1: identity file /home/luke/.ssh/id_dsa-cert type -1
I rebooted the server even, no change :(
try turning off key based auth to test
It smells almost like that's the issue
> debug3: load_hostkeys: loaded 0 keys
It's not loading any keys right now
Server: fatal: Read from socket failed: Connection reset by peer [preauth]
Phone: Key exchange was not finished, connection is closed
But I can connect from my phone when NOT on Wifi...
hm
the IP address is different then?
so wierdness is in the lan?
hm... got anything like fail2ban?
03:47
Yes. Externally, I connect to a hostname. Internally, using the IP address
I do, but it's not tripping up either
though its connecting, so not that
/etc/hosts.allow: ALL: LOCAL 10.0.0.
/etc/hosts.deny: ALL: 123.31. 69.197.
hm
that looks ok?
2018-08-15 20:30:00,375 fail2ban.filter [1091]: INFO Set maxRetry = 2
2018-08-15 20:30:00,383 fail2ban.filter [1091]: INFO Set findtime = 600
2018-08-15 20:30:00,388 fail2ban.actions[1091]: INFO Set banTime = 600
2018-08-15 20:30:00,446 fail2ban.jail [1091]: INFO Creating new jail 'apache'
2018-08-15 20:30:00,447 fail2ban.jail [1091]: INFO Jail 'apache' uses pyinotify
2018-08-15 20:30:00,473 fail2ban.jail [1091]: INFO Initiated 'pyinotify' backend
2018-08-15 20:30:00,534 fail2ban.filter [1091]: INFO Added logfile = /var/log/apache2/cloud.lukebarone.ca/error.log
That's my Fail2Ban log
Disabled Fail2ban, same issue. Restarted it. Restarted the server
can't be f2b, it wouldn't connect at all, since it sets firewall rules
03:51
iptables is virtually blank
# iptables -L -n
Chain INPUT (policy ACCEPT)
target prot opt source destination
fail2ban-apache-multiport tcp -- 0.0.0.0/0 0.0.0.0/0 multiport dports 80,443
fail2ban-apache tcp -- 0.0.0.0/0 0.0.0.0/0 multiport dports 80,443
fail2ban-ssh-ddos tcp -- 0.0.0.0/0 0.0.0.0/0 multiport dports 22
fail2ban-ssh tcp -- 0.0.0.0/0 0.0.0.0/0 multiport dports 22

Chain FORWARD (policy ACCEPT)
target prot opt source destination
ya, that's ruled out
I kinda think we can just focus on the ssh server ;)
Hmmm
# cat sshd_config
# Package generated configuration file
# See the sshd_config(5) manpage for details

# What ports, IPs and protocols we listen for
Port 22
# Use these options to restrict which interfaces/protocols sshd will bind to
#ListenAddress ::
#ListenAddress 0.0.0.0
Protocol 2
# HostKeys for protocol version 2
HostKey /etc/ssh/ssh_host_rsa_key
HostKey /etc/ssh/ssh_host_dsa_key
HostKey /etc/ssh/ssh_host_ecdsa_key
HostKey /etc/ssh/ssh_host_ed25519_key
#Privilege Separation is turned on for security
 
1 hour later…
05:05
FUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU‌​UUUUUUUUUUUUUUUUUUUUUUUUUSK
I am going to bed.... -.- Damn DHCP being out of my control tonight...
 
3 hours later…
07:47
morning
mronin @AndyK
08:06
morning
08:18
Hey so you know how sometimes you'll be chewing a piece of gum and then you come across a cube of cheese so then you want to eat cheese but cannot because there's gum in your mouth and then suddenly life just feels like such a meaningless burden...
lol
I want to say yes, but I can't say I've ever had that specific experience.
fo clippy.
08:43
Is there anonymous web-based online chat?
Where I can invite anyone to without it being required to register to have an account?
Bob
Bob
@Boris_yo @rahuldottech's time to shine? :P
also, just about every irc
few ircs require registered nicks
@Boris_yo h4cker.us
@Bob gossipmonger is offline atm
It'll take me just a moment to set up, but meh, too lazy
Sometime in the next few days, perhaps
@Dog @JourneymanGeek the fob thing from yesterday works
we just tested it in the car park at work
i can only guess that the fob RF signal is somehow piggybacked off the phone signal
one for Physics?
09:02
There's a CTF starting at 3:30pm UTC today
Anyone interested?
capture the flag?
Dog
Dog
09:57
@Burgi That's one awkward fob
@Dog how so?
Dog
Dog
@Burgi Unless the fob works merely by transmitting random bursts of signal in the same band as mobile phones, that can't be what's happening
i don't know what else to tell you, i saw it with my own eyes
Dog
Dog
If it's an acoustic fob as JMG suggested, you should be able to replicate it by recording the call or locally and playing it back on another phone next to the vehicle
we tested that the signal from the fob couldn't reach the car beforehand
Dog
Dog
10:02
Apparently combining MDMA and SSRIs is a really bad idea, which is obviously why I'm going to do it.
@Burgi Yeah, but try recording it on the phone, taking the phone to the car, and playing back the recording. If that works, it's a shitty fob system
If it requires a phone call i.e. two way communication/authentication then it's marginally better, but quite why they'd do that instead of RF is still beyond me
Apparently O2 plan on upgrading "1000 masts" to LTE B40 before the year's end. But I'm not aware of any device that can actually CA it, despite their claims. Still, there's at least a dozen in Edinburgh in the planning works, and one in progress
@Dog it was an audi
@Dog yes it needed an active phonecall
Dog
Dog
@Burgi I know nothing about cars, so that doesn't tell me much
audi's are quite premium
they are owned by the VW group which is the largest car producing group in the world so they have resources to throw at the system
10:23
@Dog not necessarily - rolling codes might possibly stop that.
Dog
Dog
@JourneymanGeek Apparently it's the same code on all cards, so no
@Dog then you could open any audi with any fob.
that dosen't sound right
@JourneymanGeek does your parents car have a key fob or is it keyless entry?
@Burgi keyless entry with a fob?
like you can keep it in your pocket and open the door
but there's no 'traditional' key
10:42
@rahuldottech It says channel is "Public". Means that big bro can come in and listen. I wanted to make private chatroom but there is no option.
10:56
@Burgi yep
Tech-oriented
Dog
Dog
@JourneymanGeek Only if it's got the optional acustic unlocking enabled
@rahuldottech Check this one: tlk.io
Dog
Dog
11:11
Hmm I need a cheap, 3CA dual-SIM phone
Preferably waterproof, unlocked bootloader, and Qualcomm chipset too. Easier said than found these days,,,
waterproof sounds like a tall order
Qualcomm rules out a lot of the chinese companies?
11:26
@rahuldottech in which game?
Dog
Dog
@JourneymanGeek All the latest flagships are waterproof, it can't be that hard
Even my Z5 Compact from a few years back was waterproof, except it had the option of 2xCA or 2xSIM, not both.
Being a 3 year old phone, I'd have assumed things have moved on since then
Sony have a thing for waterproofing their tech
12:03
@Burgi online CTF tech-based
Like a hunt thingy
 
1 hour later…
!! s/Z/{/
@allquixotic The {enphones look decent (source)
Hmm... My cat chose { as the replacement character
morning
@Bob how's your .NET Core impl of YASMA coming? :D (I figured more than likely you would actually work on it after joking about it)
Bob
Bob
@allquixotic lol. even if I wasn't busy as hell, messaging is a such a pain I'd be very hesitant to even start a project about it
just did another ad-hoc boot repair for a friend
@Bob I feel like Bitlbee is already YASMA. It just needs a proto for WeChat, then you run an IRC bouncer. Done.
Bob
Bob
13:30
who ran ubuntu/grub's boot repair and royally buggered up boot for both windows and ubuntu (which didn't work in the first place)
@Bob Brilliant!
Bob
Bob
it's an ivy bridge laptop that can't run efi shell v2
had to run v1 which didn't have bcfg -_-
I end up fixing Windows boot for people who don't dual boot all the time. It always amazes me how often people manage to brick boot without even trying. Does disk corruption intentionally target blocks related to boot or something?
Bob
Bob
@allquixotic the proto's the hard part, else we'd already have a matrix bridge
I mean, 2 TB spinning rust, what's the probability that a random bad block is in the critical boot path?
Bob
Bob
13:32
I've currently got it running on my old S7E, to be replaced with a Moto C next week.
@allquixotic Oh, maybe 4 months ago I helped this guy fix boot after the 1803 update stuffed it -_-
thankfully uefi boot is relatively straightfoward
it's so nice dealing with actual files rather than obscure binary blobs sitting in raw disk sectors
the only thing that gets me about UEFI boot is the Windows BCD. I rather like how Grub does it actually
some binaries in the ESP and a conf file... yum yum so simple
or should I say apt apt so simple
Bob
Bob
@allquixotic I mean, technically, that's what windows does too, except the conf file is a binary format (bcd)
remember when you did that weird BCD hacking for me to get Win10 installed on my server without IPMI? :D
Bob
Bob
problem I have with windows is it likes to steal boot if its fallback bootx64.efi gets invoked
but it looks like grub does the same so... :\
lol
I had trouble with that
13:35
I still think that was the most brilliant thing and I don't even understand what you did
Bob
Bob
@allquixotic don't worry, I don't understand what I did either :D
(read: I've actually completely forgotten and will have to make it up from scratch if it ever comes up again)
I think you did something with character encoding
UCS-16 or summat
Bob
Bob
@allquixotic oh, that wasn't the BCD, that was bcfg, which is a UEFI Shell thing
the BCD (bcdedit) is a Windows thing
wait, in your case it was efibootmgr, which is the Linux way of modifying the UEFI boot vars
maybe, but (1) I don't recall booting up UEFI shell, and (2) I recall distinctly you gave me an encoded string to put into the boot path to fix it
Bob
Bob
(bcfg is more reliable but you didn't have access to it cause no IPMI/Shell)
13:37
right, that.
so it wasn't BCD, it was modifying the NVRAM basically
Bob
Bob
yea, nice thing about bcfg is it's pretty easy to save the encoded text file to disk and tell it to read options from there
efibootmgr doesn't seem to do that, or maybe it's some arcane option combination
@allquixotic pretty much!
BCD, for all its obscurity, is still easy
it's a binary file but it's a file you can copy off and modify
that's the part that is slightly annoying to me when it comes to UEFI. Having readable text and PE binaries on disk as a clear and unambiguous way of booting? Awesome. Having random blubs in an NVRAM that's notoriously hard to poke (especially if you didn't boot up in UEFI mode)? Ick.
Bob
Bob
the EFI boot vars are a huge pain simply because they're localised to the machine hardware
Bob
Bob
there is no guaranteed way to modify them on another machine
dumping and loading with efibootmgr is about as close as you'll get but, as you saw, you run into encoding issues
13:39
whiiiich also makes dd or zfs send style server migration not foolproof if you needed something in NVRAM.
Bob
Bob
@allquixotic I think theoretically Windows works around it by making bootx64.efi work even without any vars, and it'll auto-discover the nearest BCD. Grub does something similar with its own bootx64.efi (which looks for a grubx64.efi)
But that assumes you can set the firmware to boot from the disk's fallback loader
which is usually the UEFI: <HDD name> option ... that you can't get to without physical (or OOBM) access
yeah, grub has worked fine for me without vars in the past
Bob
Bob
You could also manually create a bootx64.efi that's actually an EFI Shell implementation and script your own loader with a startup.nsh script
which can then call fs0:\EFI\Boot\blah.efi options
@Bob yeah I mean that style booting would be required in any case, because without it, you couldn't even insert a UEFI bootable media like a USB disk or DVD to boot up and modify the vars, and you'd be caught in a permanent catch-22 aka bricking :P
bricking of UEFI boot anyway. could still CSM
Bob
Bob
@allquixotic your biggest problem was OVH couldn't help/let you choose the right boot option :P
@allquixotic IIRC CSM won't let you modify EFI vars so you're stuck anyway
13:43
@Bob they probably could've chosen a different boot option for me at some ridiculous cost... their KVM over IP was IIRC $250
@Bob right, but even if you were unable to boot any UEFI OS you could still accept defeat and install an OS in BIOS mode if you have a system disk smaller than 4TB
Bob
Bob
instead of adding vars we probably could've just told efibootmgr to change the boot order so fallback hdd boot came first, now that I think about it
(...wait, didn't we accidentally nuke the hdd boot option? :D)
@Bob I think the only reason we were able to continue working on it is that OVH configures their servers with PXE boot as the highest priority, and I could go into the console and tell it to boot the rescue Debian over PXE, so even if we bricked networking you basically have a minimal IPMI (their rescue env, mercifully, could tickle NVRAM)
I might need to let go of that server, sadly. No more need for Conan Exiles atm
Bob
Bob
yea
Kas and I tried to start up a server but we didn't have the time to work on it, so I'm in ESO again
Bob
Bob
we don't seem to have enough time/will to put into such a big thing at the moment :\
13:47
I tried another Conan server but too much PDA/ERP for my tastes/mental state
ESO's turning out better
Bob
Bob
I might need to look at that sometime... time being the issue
it's easy to find 30 mins to play something
well, good news, you don't have to host a server.
Bob
Bob
but a nice block of several hours? not so much
we could just run dungeons in ESO if you wanted. no need to RP unless you want to
Bob
Bob
3
Q: SCP Changing text file contents

mzhaaseI am trying to copy a file from machine A: apt policy openssh-client openssh-client: Installiert: 1:7.2p2-4ubuntu2.4 to machine B: apt-cache policy openssh-client openssh-client: Installiert: 1:5.5p1-6+squeeze5 On A: scp myfile <server>:/myfile md5sum myfile 2ba67c5e816350d...

The way SCP was capitalised in HNQ made me think it was a SCP where a file would change itself
2
13:50
@Bob wow
somehow I remember reading that a long time ago, but hah. It would figure his problem was basically "wrong file dummy"
every time William Shatner releases a new album I go "Is he actually going to sing now, or will he just talk?" and every time I'm disappointed :P
@Bob Go hit the guy doing this to me, will you? I'm becoming more and more like you. I have over 50 tabs open (!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!) in my Firefox. How the hell have I gotten to this state?
What is even happening? I can feel my control over my tabs slipping away...
They proliferate on their own. The tabs are gaining the upper hand. I am getting carpal tunnel from Ctrl+W to close them and yet they still keep coming. Tabs on top of tabs. Stacked tabs. Tabs that hit my tab key when I don't mean to while coding Python. Tabs that are actually four spaces. Spaces in my tab bar! What...?
Now they're moving themselves all over my screen. I have a tab as my mouse cursor. I think they're gaining sentience.
14:10
0_0
you should switch to spaces
anyone know how to get back deleted git repos and medium posts? WBM hasn't archived them.
@Nick I assume you're talking about git repos stored on github or bitbucket or something? If they aren't cached anywhere and nobody else has a copy, there's no way. It's gone. Ditto for any generic website whether it's Medium or anything at all.
github would probably be able to restore a deleted repo for an owner for a limited time after it was deleted because I'm sure they keep offsite backups, but I can't imagine they would provide a copy to anyone else
@allquixotic dang. Is there atleast a statistic for the number of deleted stuff on github?
@Nick not that I know of
mhh..
@allquixotic Use Session Buddy, I had 140 tabs open a few days back and chrome's native session restore kept giving it back to me every time I opened the browser. With session buddy, I could save all the things and mindlessly close it all. No effort.
14:31
@Nick the problem isn't saving my tabs; the problem is getting rid of them :P
@rahuldottech this hasn't really explained it
@allquixotic ctrl+w feverishly
14:44
There should be some extension out there that plots your tabs open and closed activity
a graph like that would be useful to determine how many tabs on average you should open.
Bob
Bob
@Nick You can do all that natively in Firefox, sooo. Yeah.
i don't like the tab height on most browsers
also, the omnibar should disappear after page load... it's just a static waste of space
Starts in 11 minutes
@Bob interested?
Bob
Bob
uhhh
Lemme know, I'll share my ID
Bob
Bob
14:49
1am tho :P
@Bob aw
It's for 12 hours
So whenever it's convenient for you?
Bob
Bob
@rahuldottech a ctf?
Bob
Bob
do your teams have to be preregistered?
@Bob no, you can register now. I'm already registered tho, and you can join my "team"
(it's only me ATM)
Oh, I should ask @Ave
Bob
Bob
14:51
@rahuldottech I mean, do you need to tell them team members?
Ave
Ave
WHOMST DARETH PING ME
2
@Ave interested in a CTF?
Starts in 8 mins
Ave
Ave
no
Oh, okay
Ave
Ave
14:52
I had a drive failure yesterday
Ave
Ave
lost some prod data yes i had no backups
Bob
Bob
@rahuldottech I might poke you tomorrow
@Ave /me points at sidebar :P
Ave
Ave
spend 7 hours last night on (legally obtained, prescribed) speed just trying to recover one of the services
now one big service is back + there's backups
+ that service's files are open source now
Bob
Bob
@Ave How much space are you using?
14:54
pretty sure plenty of us have spare room...
you only have to ask
Ave
Ave
@Bob huh?
Bob
Bob
@Ave I mean, what's the size of the not-backed-up data?
Ave
Ave
uhh
@Bob Okay
Bob
Bob
(that's the biggest problem I've had... finding a place to put my backups without paying through the nose)
Ave
Ave
14:55
couple hundred gigabytes
Bob
Bob
Oh. That's ... moderately painful but not TB range at least
its only going to get worse as your economy freefalls
Ave
Ave
@Bob well most of it was backups of my nintendo switch
> /dev/mapper/vg-backup2 1.7T 345G 1.2T 23% /backups
Ave
Ave
I don't mind losing those
14:56
1.2T free...
Ave
Ave
but some of website data is lost
some historic metrics are lost (IDC)
Bob
Bob
ouch
@Ave any chance of recovering from google cache/archive.org?
Ave
Ave
and also the service I mentioned a bit ago, the one I worked 7 hours on recovering (tl;dr it's jenkins building many many projects) was lost
@Bob I'll have to do that on some sites, will just rewrite the rest
they were ugly anyways

« first day (2932 days earlier)      last day (2384 days later) »