Root Access

marcusdoesstuff

14:03

@Bob I got this to work in the end :)

@Bob I've now noticed that there might be a problem with the scraping script however. I see that it's set to look for "Contains" but that particular data is not in the resulting dictionary.

gist.github.com/marcusdoesstuff/b28d99e6a9ff486fa3ea

Boris_yo

Anyone familiar with this BangGood chinese store?

I wanted this Xiaomi and they sell authentic: banggood.com/…

Bob

@Boris_yo Slow shipping, dodgy returns, but genuine AFAIK.

@marcusdoesstuff Are you sure that's actually in your data source?

Open up the CSV in Notepad.

Journeyman Geek

@Bob: so... as good as you can hope for?

lol

@bertieb @Bob: My backups are daily, incremental and tested recently ;p

Bob

@JourneymanGeek My last backup 'test' involved data corruption from PrimoCache

marcusdoesstuff

@Bob I added the csv as plaintext as a comment on that gist now too. It does contain "Contains,,,"SODIUM HYPOCHLORITE SOLUTION, ... % Cl ACTIVE",,," though... I'll test on some other ones.

Journeyman Geek

14:14

@Bob: My last backup tests were migration to new drives

Bob

@marcusdoesstuff Ah, it got eaten up by the multilineendpattern :P

Journeyman Geek

actually I also did one where I tested a restore to a spare drive too

marcusdoesstuff

i wondered about that bit @bob

Journeyman Geek

My linux box on the other hand...

I haven't found a backup option I like yet

Bob

@marcusdoesstuff remove the matched = True from the multilineendmatcher loop, and remove the if matched: continue after it

marcusdoesstuff

14:16

@Bob Did it just find the start and end in the same place so return nothing?

Bob

since it's the first item, that should be fine

@marcusdoesstuff no, it ended a previous multilinematch and didn't process that line as a singlelinematch

albal

Hello. someone ( @barlop ) just told me network classes i.e. A, B, C are no longer in use for private address ranges. Is that true? Just we used them quite a bit where I last worked and now use them here because of that.

Bob

@albal ...they haven't been in use for any range for two decades

Journeyman Geek

No one uses classes now

barlop

They're not called classes anymore

but those IP ranges are still used

Bob

14:17

There is no such thing as a "class A" or "class B" or "class C" anymore.

albal

Wow I wonder how I missed that. So if someone wants to stand up a subnet with /24 they just choose any address range?

Bob

@albal Look up CIDR.

albal

I mean within the privates

Journeyman Geek

Pretty much

barlop

no

it's blocks now.. so you're assigned a block

and for private address ranges you're still using those ones like 10.x.x.x

listed in RFC 1918

Journeyman Geek

14:19

Most people use a 192.168.x.x subnet.

You don't typically go with a 10.x.x.x net unless you have a shitload of systems

barlop

yeah they can use /24 on 10/8

like 10.1.2.0/24

Journeyman Geek

Oh you could

I meant typically

albal

Hmm I understand you can. But for new networks using the simple class based stuff at least means you start from the same point as everyone else using classes

Journeyman Geek

Nope

barlop

the term classes isn't used anymore

that's the thing

albal

14:21

I was aware of CIDR, didn't understand that saying Class was so old skool to be misleading

Journeyman Geek

You pick an appropriate RFC 1918 block ;p

barlop

it's called blocks now

i'm not quite sure off hand how it differs from classes!

albal

it's not the same so can't be called the same

Journeyman Geek

(and for extra credit, find the equivilent, and the proper 'example' blocks for both ;p)

marcusdoesstuff

@bob That works. It drags three commas through though. "SODIUM HYDROXIDE,,,"

Bob

14:22

@barlop Classes don't just specify size. They are allocated a specific part of the global address space

A class C network must be within the 192.0.0.0-223.255.255.255 range.

You can have a CIDR /24 outside that range, but a class C network by definition cannot be.

@marcusdoesstuff You're going to have to figure out trimming or the regex yourself.

I suspect some of those might actually be lists, which would be more problematic.

albal

rfc1918 is still current, it refers to classes. I don't think it is misleading to talk about it for private networks

*it= classess

Bob

@albal It mentions classes.

As an aside.

The primary definition is in the realm of CIDR, and the 'class' definition no longer makes any sense today.

marcusdoesstuff

@Bob Okay :)

Bob

@marcusdoesstuff You said you know how to use grep, yea? Take a look at the regex.

albal

It states they are blocks and as an aside that they are the 3 classes

Bob

14:25

For the single line matchers, the first group defines the key, and the second group is the value

marcusdoesstuff

That'll be refreshing! Something I vaguely understand!

albal

I accept that in the realm of CIDR yes, but private LANs are a subset and I think class still rightly applies

barlop

maybe, i'm not sure in what way classes no longer apply besides the terminology not being used..

and that anybody will jump down your throat at the use of the term class

oh I know..

Bob

@albal Semantics aside, it makes absolutely no sense to keep both schemes.

barlop

and maybe this is still definitional

The C in CIDR means Classless

Bob

14:27

There is no good reason to refer to anything by classes outside of a historical review.

All it does is create confusion.

barlop

@Bob well, if you're still using 192.168/16 then why is it confusing to call it Class C?

Bob

Especially when people refer to a /24 as a "class C", which is also technically incorrect

barlop

(not that i'm promoting the use of the term class!)

Bob

@barlop because there is no need to, no good reason to, and no benefit to doing so?

albal

If it's referred to correctly it is not confusing but when you say /16 is class C it just lies :-)

barlop

14:29

@Bob RFC 1918 says "and third block is a set of 256 contiguous
class C network numbers."

albal

I'm not promoting it I'm just trying to fully understand why it is now outdated

Bob

@albal it is confusing because people see it and either come away with incorrect assumptions about the meaning of a class, or assume you're talking about some legacy scheme.

Pretty sure it violates a BCP somewhere.

albal

IPV4 is still current

barlop

@Bob what incorrect assumptions?

Journeyman Geek

unfortunately ;p

Bob

14:30

@barlop context: "Note that (in pre-CIDR notation)"

That's a side note.

The RFC defines the segments in CIDR.

barlop

ok.. And what incorrect assumptions are you referring to that people could make if the term class is used?

Bob

And the only reason it even bothers with the pre-CIDR notation is because it was written ~3 years after it was deprecated.

albal

A dozen is a way to refer to 12, it doesn't make the word Dozen redundant

Bob

@albal Except "class C" is not the same as /24

barlop

@albal Maybe pre CIDR was VLSM and FLSM and the term class suggests use of VLSM or FLSM?

Bob

14:31

It's a different thing

albal

Yes sorry that was a poor straw argument :-/

barlop

@albal no it's a good point.

Journeyman Geek

@albal You wouldn't call it XII tho

Bob

@JourneymanGeek Again, different.

albal

No its nearly as bad as calling CIDR blocks by class :-)

Journeyman Geek

14:32

;p

barlop

@albal are you familiar with VLSM and FLSM?

Bob

@barlop It's becoming less common now, but it was quite common to see people automatically assume any /24 was a "class C" network.

albal

okay so it is not cool anymore I will try and refrain from it around here

yes I used to look at a CLASS B

I mean /16

barlop

@albal you quit before finding your answer

Bob

Because some people referred to one /24 as a class C, whether correct or not, and others just thought that was the definition.

albal

14:33

*after

Bob

It happens a lot.

barlop

@Bob the question here though is not over definition

Bob

@albal Point is, they're different things. There is no such thing as a class B outside the 128.0.0.0-191.255.255.255 range, while /16 is valid outside that range.

barlop

we can agree that the definition now is that the term class isn't used anymore

Bob

@barlop You asked why it causes confusion.

barlop

14:34

ah ok.

Bob

It causes confusion because when you start using it, people read it and think it means something else.

barlop

I see 192.168 is /16

albal

I think the point was my reference was to private address space

barlop

err

Bob

If you use it in the context of 192.168.1.0-192.168.1.255, people will read it and think it can be applied to any /24, and will start doing so elsewhere where it's wrong.

albal

14:35

Personally I will have to go with the flow - as most dead fish do

barlop

I see 192.168 is /16

Bob

Most laypeople won't bother to look up the definition.

albal

So how will they be better informed if they are given 192.168.1.0/24?

Bob

@albal Neither term will make much sense. But one is more easily applied incorrectly.

'nother point, routing protocols do not deal with classes anymore.

barlop

I see even that CIDR document shows the classful system to not apply. When it says "Note that (in
pre-CIDR notation) the first block is nothing but a single class A
network number, while the second block is a set of 16 contiguous
class B network numbers, and third block is a set of 256 contiguous
class C network numbers."

'cos in classful addressing, class C is /24 Whereas that second block is 192.168/16 though just after the class B

Bob

14:38

@barlop 256 contiguous /24s make a /16.

albal

Yeah okay I see the point now

Bob

You can't even call it a single "class B" because class B by definition cannot be in that range.

> There is no such thing as a class B outside the 128.0.0.0-191.255.255.255 range

It is 256 contiguous class Cs, by the old definitions.

albal

In my mind: CIDR is the current way to express IP address ranges. Classes provide a superfluous finer definition of blocks that are not necessary.

Bob

Bit of a mouthful, eh?

Another point: once you make a 10.x.x.0/24 subnet, it's not a class A anymore either!

barlop

good point

Bob

14:41

In fact, subnetting like that isn't possible with classful addressing.

barlop

There was FLSM and VLSM with classful wasn't there?

they preceded CIDR

albal

@barlop should I leave my comments as they are, or remove the Class reference - which would then make your comment seem out of context

barlop

leaving it is better.

Bob

@barlop Not exactly sure where those terms are defined, but CIDR is (a form of?) VLSM, and classful networking applies FLSM

barlop

maybe an early RFC pre CIDR

Bob

14:44

yea, trying to find any references

barlop

I don't know if they said classes didn't exist, in the days of VLSM

@Bob tools.ietf.org/html/rfc1878

Bob

@barlop Doesn't use the term 'VLSM', though

barlop

"his table includes subnetting for Class A, B, and C networks"

Bob

and that was two years after CIDR

barlop

doh

Bob

14:46

I don't think it's really defined (or even used) in an RFC

the concepts are definitely described, but not the term

albal

Any reference now appears historical

Bob

in fact, there aren't many search results for it overall... maybe some old courses? or some networking company's docs?

I guess FLSM and VLSM are more general, if anything.

albal

Pretty sure it was in my CCNAX 3 years ago

but they moved to version 2.0 shortly after we did the course

Bob

@albal They still talked about classful networking outside a historical context?

Good lord. In 2012?!

(that's 19 years after CIDR)

albal

Yeah it is Cisco - they think they are it.

Bob

14:49

classful networking existed 1981-1993, CIDR 1993-

12 years vs 22 years

barlop

only cisco students spoke of classful and they got jumped on for it

Bob

CIDR has existed longer by now

barlop

wikipedia says "Classless Inter-Domain Routing is based on variable-length subnet masking (VLSM), which allows a network to be divided into variously sized subnets, providing the opportunity to size a network more appropriately for local needs. Variable-length subnet masks are mentioned in RFC 950"

albal

hahahah like now

barlop

though that RFC doesn't use the term VLSM

maybe cisco invented the term VLSM!

'cos I read of it in a cisco book

ebook-shopitctw.blogspot.co.uk/2011/06/…

I don't know if that's a download link

but it shows the cover of the book and says VLSM on that webpage

Bob

14:52

> a protocol for the ARPA-Internet community

:P

barlop

they kept changing between ARPA and DARPA I don't even know what came first anymore!

DARPA->ARPA->DARPA

Awal Garg

15:14

Heyo

barlop

@albal and from now, one day, you too will see somebody use the term classes, and jump on them!

Bob

15:40

@JourneymanGeek massdrop.com/buy/lg-31-digital-cinema-4k-display :P

that pricetag, ouch

bertieb

16:20

@Bob You brought Massdrop to my attention. MY POOR WALLET

"...<something something about you killing my father>, prepare to die."

That Brazilian Guy

17:13

Hello ~~humans~~.

0

Q: tag [remote-access] vs. [remote-control]

As of the writing of this, there's 225 questions tagged remote-control and 537 tagged remote-access. The description on the "tag info" page for the latter seems much more complete than the info for the former tag. Specifically what puzzles me is the snippet I bolded out: Gaining control to...

Boris_yo

18:02

@Psycogeek How about these to hold flash drive?

http://www.banggood.com/3-Size-Stainless-Steel-Wire-Ring-Keychain-Ring-Quickdraw-EDC-Tool-p-942843.html

http://www.banggood.com/New-Durable-Set-of-6-Steel-Nylon-Strings-Acoustic-Classical-Guitar-p-933557.html

Oliver Salzburg

18:13

Back to 3 displays \o/

Boris_yo

@OliverSalzburg You are spider?

Oliver Salzburg

@Boris_yo Nope, still human

Boris_yo

Actually I think you have your 3rd eye activated.

Oliver Salzburg

Possibly ;)

Boris_yo

19:59

@Psycogeek I found mirrorless Panasonic G5 but on second-hand sale. After talking to the guy he said that he used it for 8 months and it's in great condition. He purchased it from second-hand user too and that original user purchased it on Amazon US. A lot of hands for camera to change?

Journeyman Geek

23:07

@Bob: sounds arrround right

1) wierd aspect ratio 2) Its higher res than most regular 4K displays.

Journeyman Geek

23:23

arstechnica.com/gadgets/2015/08/…

whaaaaat?

DragonLord

23:53

@Boris_yo The G5 is a bit old, you might want to look at the G7, which is capable of recording 4K.

It's also reasonably priced.

If the G7 is too expensive, you could look at the less expensive GF7.

Transcript for