#asciilifeform | 2022-02-18

cement activity is piped to debug log.

00:35

e.g.: 'Proposed block 00000000059f962b47efef20b5038f79cecbf406a1ab6ff9accaefdbe72b815f (57259) ACCEPTED by cement.'

00:36

notably, disabled 'inv' processing when 'in cement', so receives ~0 extraneous blox.

00:36

(also ~no tx, but these aint usefully edible when yer syncing anyway)

00:36

gives pretty close to the fastest possible (w/out parallelisms) sync afaik.

00:43

scoopbot

New article on billymg: Straightening a Bent 1U Server Chassis

~ 34 minutes ~

01:17

... in 70k's, 0 pauses thus far...

~ 34 minutes ~

01:52

I must say, cement is an insightful dectriptor asciilifeform

~ 16 minutes ~

02:08

any allcaps hardware on the horizon?

02:12

vex: nope

02:20

... 90+k, no pauses. (these are the lightweight blox, tho)

02:23

!e height

02:23

trbexplorer

vex: my valid commands are: src, uptime, version, help, view-height, view-address, view-utxos, view-balance, view-block, view-raw-block, view-tx, view-raw-tx, push

02:23

!e view-height

02:23

trbexplorer

block_height: 723831

02:23

trbexplorer

mins_since_last_block: 19

02:27

what's that in 14 fours alfie?

02:27

?

02:28

in 1.44 inch floppies. big stack

02:30

a

02:30

~4e6

02:30

inches?

02:30

disks

02:30

ah

02:33

whaack: comment in yer spam queue

02:39

verisimilitude: red now, good shit

02:39

tho i fail to see why u didnt just do smth greek like http://zzz.st/up/NeHlDMls/

02:39

wb jonsykkel

02:39

ty

02:45

cunning linguists in full effect.

02:50

... 100k+, 0 pauses

03:02

get well soon signpost

03:10

cementing as we speak from blok235000

~ 1 hours 3 minutes ~

04:14

jonsykkel: a++. no pauses ?

04:14

... 125k+

04:15

interestingly, seems to 'automagically' sync from 'fastest' atm peer

04:15

(rather than one-from-1, one-from-anuther)

04:17

depends wat constitues a pause

04:17

(not sure wat trbsync normally looks like)

04:17

jonsykkel: let's say a full minute w/out an eat

04:17

Logged on 2022-02-17 19:35:15 asciilifeform: e.g.: 'Proposed block 00000000059f962b47efef20b5038f79cecbf406a1ab6ff9accaefdbe72b815f (57259) ACCEPTED by cement.'

04:18

jonsykkel: normally, w/out cement, the thing will stall, sometimes for hours, 'waiting with open mouth' for sumbody to feed it

04:18

(when 'behind', that is, or freshly syncing)

04:18

aha

04:18

dont have no timestamps in debuglog so hard to say

04:19

jonsykkel: i left mine w/ tail -f debug.log, so visually apparent

04:23

a nice, puted now tail -f debug.log | xargs -IL date +"%Y/%m/%d %H:%M:%S L" >> debug2.log

04:23

will chek bak tomorow

04:28

jonsykkel et al : can try 'unloadcement' to see the difference.

04:28

( then can load it again w/ 'loadcement' )

04:31

*

asciilifeform notes that it is possible w/ this patch to 'catch up' any 'fallen behind' noad via cement, generate start .. end on a synced one, loadcement on the sad noad

04:32

as i understand this is in fact what jonsykkel was doing

04:32

Logged on 2022-02-17 22:10:03 jonsykkel: cementing as we speak from blok235000

04:33

indeed, my sadnode was set up couple days ago

04:33

a

04:34

jonsykkel: what height nao?

04:35

asciilifeform: 235941

04:36

that "235000" was some 235xxx

04:36

jonsykkel: paste coupla tho. ln. of tail ?

04:36

http://paste.deedbot.org

04:37

sec

04:40

*

asciilifeform curious wai his seems to be going ~6x faster than jonsykkel's

04:41

http://zzz.st/up/nLyxv0Bz

04:42

its a 15yo laptop with mech hdd

04:42

aa

04:42

prolly explains

04:42

my testbed is a 'ryzen' on 1g pipe, lol

04:42

lol

04:43

this node a bit more sad yep

04:43

w/ nvme

04:43

jonsykkel's test is prolly moar typical tho

04:43

so a++

04:43

at any rate log loox visually correct.

04:44

i.e. dun see any stalls

04:44

ty jonsykkel

04:44

well see if can get syncd before newyear

04:44

*

asciilifeform must bbl

04:44

cool

04:44

bey

~ 3 hours 57 minutes ~

08:41

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-15#1079464 << Fugg. I can't tell if I'm being milked for lulz or you were serious.

08:41

Logged on 2022-02-15 14:32:08 asciilifeform: crtdaydreams: reiser

~ 28 minutes ~

09:10

The more I read the more I'm inclined to the former.

~ 58 minutes ~

10:09

https://www.youtube.com/watch?v=OqFCBarT-Q0 I checked your mum cdd. My forklift isn't big enough

10:17

protip. noone gives a fuck about you

10:19

That actually comes off harsh. you might be nice

10:25

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-18#1079793 << No. You're right.+

10:25

Logged on 2022-02-18 05:17:53 vex: protip. noone gives a fuck about you

10:26

Just another fucking cancer in this hellhole that will ultimately contribute nothing.

10:26

shit. you're gonna make me hang myself

10:27

I've fucked up, that's all. Big time.

10:28

dyareckon your mum'll let you go wardriving?

10:28

wut.

10:28

I could ask her very nicely

10:29

sure. go ahead.

10:30

got any sploits?

10:30

who am I kidding? I can't code ofc not.

10:31

I'm also not gonna try to sex your mom

10:31

k, she's a virgin anyway.

10:32

now i feel bad for chatting shit

10:32

I'm joking mate

10:33

me too. i'll totally fuck ur mum

10:34

swap numbers?

10:34

yeah nah you spooky little thing

10:34

deals a deal

10:36

ok then

10:36

that was a joke ftr, I don't wanna be doxxed.

10:37

i probabbly dont wanna orgasm ur mum

10:38

ok.

10:40

if she worked on her diet & fitness. mebbe

10:41

what % ya drinkin?

10:41

yeah. nah

10:41

ipa

10:42

nah fuck nah

10:43

is ur mum hot tho?

10:43

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-18#1079822

10:43

Logged on 2022-02-18 05:42:35 crtdaydreams: nah fuck nah

10:44

I ain't into that oedipal complex shit.

10:45

Wouldn't know.

10:45

totes innapropes

10:46

hydrate.

10:46

fuck off cunt

10:46

gonna trigger a winter soldier episode?

10:48

fuck me, that's enough shittalk, got any music?

10:48

yep. i'VE GOT A DANISH DOOSIE FROM 1992

10:48

whoops. hit capslock

10:49

aw yip?

10:50

Here's a banger for ya.

10:52

seriously. do you have a daddy?

10:52

No. I have a father.

10:53

fair cop

10:53

I might've undersestimated your age

10:54

What's that supposed to mean?

10:54

I thought you were like 20. ur prolly 29

10:55

I literally told you the other day.

10:55

I can't rmember shit

10:55

Doesn't matter.

10:55

right

10:57

You'll forget I even existed in a few weeks from now anyway.

10:59

if the courst ever ask. never heard of the cunt

10:59

time to zap you with one of those men in black forgettamathingies

~ 17 minutes ~

11:17

https://www.youtube.com/watch?v=hIw7oeZKpZc

~ 2 hours 29 minutes ~

13:47

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-17#1079773 << mech hdd with the bdbmay prove unfeasible, while still an interesting experiment

13:47

Logged on 2022-02-17 23:42:13 jonsykkel: its a 15yo laptop with mech hdd

~ 1 hours 7 minutes ~

14:54

height 216141 nao & no apparent stalls

14:55

at this height already heavier, some take 1-2s to verify

14:57

the experiment worx, to the extent asciilifeform expected it to -- i.e. noad syncs at ~same pace as it would have via 'eatblock'.

14:58

at some pt asciilifeform will cut some of the gnarl from the patch and bake a 'final' on top of the 'draft'.

14:59

*

asciilifeform encourages folx to test the current item

15:00

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-17#1079658 << does 'level-by-level' here also mean 'branch-by-branch'? ie. can validate transactions one by one, as they stream from peer, if provided with enough related higher-level merkle hashes

15:00

Logged on 2022-02-17 13:37:43 asciilifeform: would've been entirely possib. to send the header, then the 1st level of merkelisms, then n+1st, etc.

15:00

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-17#1079659 << asciilifeform prolly agrees that, strictly speaking, block hash is derived from the 80-byte block header, while the 'merkle root' hash is ~embedded~ in the same header. but i don't grasp how would merkleism allow tx reordering without the merkle root changing

15:00

Logged on 2022-02-17 13:38:52 asciilifeform: (for n00bz: block hash is a 'merkle tree', i.e. permits reordering tx and get same hash)

15:00

see CBlock::BuildMerkleTree() here, which returns the merkle root hash

15:00

cgra: level by level.

15:00

example: lv 3: tx hashes A B C D E; lv 2: merkle hashes AB CD EE; lv 1: merkle hashes ABCD EEEE; lv 0, merkle root: ABCDEEEE.

15:01

asciilifeform oughta have said concretely -- obv. cannot reorder arbitrarily; but possible in principle to send by level.

15:02

asciilifeform: how do txes divide into levels?

15:03

can only see branch division

15:07

come to think of it, cgra is right, can't send by level

15:09

*

asciilifeform not sure where got this notion

15:15

!w poll

15:15

Polling 14 nodes...

15:15

205.134.172.6:8333 : (172-6.core.ai.net) Alive: (0.081s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Return Addr=0.0.0.0:8333 Blocks=723899

15:15

205.134.172.4:8333 : (172-4.core.ai.net) Alive: (0.082s) V=70001 (/therealbitcoin.org:0.7.0.1/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:15

54.39.156.171:8333 : (ns562940.ip-54-39-156.net) Alive: (0.112s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:15

205.134.172.27:8333 : Alive: (0.145s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899 (Operator: asciilifeform)

15:15

71.191.220.241:8333 : (pool-71-191-220-241.washdc.fios.verizon.net) Alive: (0.155s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899 (Operator: asciilifeform)

15:15

205.134.172.28:8333 : Alive: (0.154s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Return Addr=0.0.0.0:8333 Blocks=723899 (Operator: whaack)

15:15

208.94.240.42:8333 : Alive: (0.204s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:15

54.38.94.63:8333 : (ns3140226.ip-54-38-94.eu) Alive: (0.259s) V=88888 (/therealbitcoin.org:0.8.88.88/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:15

94.176.238.102:8333 : (2ppf.s.time4vps.cloud) Alive: (0.308s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723044

15:15

82.79.58.192:8333 : (static-82-79-58-192.rdsnet.ro) Alive: (0.328s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:15

103.36.92.112:8333 : (terebe.ns01.net) Alive: (0.654s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:15

75.106.222.93:8333 : Alive: (0.477s) V=99999 (/therealbitcoin.org:0.9.99.99/) Jumpers=0x1 (TRB-Compat.) Blocks=723899

15:16

143.202.160.10:8333 : Busy? (No answer in 100 sec.)

15:16

205.134.172.26:8333 : Busy? (No answer in 100 sec.)

~ 44 minutes ~

16:01

asciilifeform: why are gapped cements supported in your draft patch?

16:02

a minor point: for consistent only-public-methods-with-lock, would move cs_cement from currentEntry() to getExpected()

16:07

cgra: for experiments where a .. b blox cemented, b+1 ... c not, c+1 .. d -- are

16:08

cgra: currententry aint even needed, lotsa gnarl in there, will clean up

16:08

not even convinced the lock is needed

16:09

asciilifeform: right (re lock likely unnecessary as cement now is)

16:10

cgra: other thing is, orig. asciilifeform wrote this where only supported cements of 0..n. then realized that often enuff will want to use cement to 'catch up' a mostly-synced noad, and 0 .. max is pretty heavy (>50MB atm) , would like to be able to cement a .. b for arbitrary a<b height.

16:11

cgra: hypothetically w/out lock could get inconsistent behaviour if loading cement takes place while cement's 'verify' or 'expectcement' happening

16:12

*

cgra refrains from further lock comments until has pressed item to look at

16:12

there's at least 1 place where asciilifeform forgot to remove a lock and it gets invoked from inside same. surprisingly worx.

16:12

(evidently trb's locks lib makes this a noop)

16:13

testbed at height 221646 just nao.

16:13

asciilifeform: cement segment makes sense for catch-up yeah, not yet come up with a case where gapped beneficial

16:14

cgra: prolly oughta eggog on multiple cements that aint consecutive.

16:14

(conceivably sumbody may want to load cement in batches, esp. if e.g. we post'em in batches on tbf. but these oughta be consecutive.)

16:15

asciilifeform: trb's locks are recursive, yeah. sätöshi's been doing nested same-locks too

16:16

("sätös", btw, in finnish, means "a haphazard, poorly designed or built creation")

16:16

ha.

16:17

cgra: was thinking re parallelism -- can't think of how to even attempt it on the current turd, what with its single db cursor, ocean of locks, and the fact that block n might contain tx where inputs in anywhere 0 .. n-1

16:19

asciilifeform: the only part i was thinking the parallelization for, was that now it alternates between downloading and verifying, these don't overlap other than perhaps worth a 64k tcp buffer

16:20

cgra: where decent (e.g. 1g/s) pipe on both sides, block downloads in 8msec. so not 100% certain this is worth the moving parts/complexity

16:21

asciilifeform: yeah, true. depends on whether can lock-in to a 100kb/sec node as well a 1gb/sec node, depending on the phase of the moon

16:21

cgra: interestingly, as-written it seems to gravitate to pulling blox from fastest (at given time) peer

16:21

given as how that tends to be the one who responds 1st

16:22

an entirely accidental win

16:22

w/out any purposeful coad to this end

16:23

asciilifeform: what signs of gravitation you've seen? not just the first node it started with, and still going with the same?

16:23

cgra: seems to switch erry coupla hrs, but at no point thus far 'stuck with slow'

16:24

granted this'll need plenty moar empirical test to say that in fact never happens

16:24

right

16:25

so far all asciilifeform knows is that the sync is carrying on at ~same pace as does w/ 'eatblock' from disk in earlier experiments.

16:25

however obv. these so far are the 'lighter' blox

16:34

*

asciilifeform does suspect that 'fastest' peer in fact delivers the asked block 1st, to the degree that the 'ask' in fact takes place in parallel

16:34

... and consequently moar likely to be the 1 who answers the ~next~ ask, etc

16:36

the peculiar thing is that asciilifeform sees virtually no dupes from the lagging peers

16:37

asks, more or less, schedule with 2min intervals

16:38

no dupes prolly cuz always download and verification from first giver succeeded under 2min

16:47

http://logs.nosuchlabs.com/log/asciilifeform/2022-02-18#1079851 << cud be. havent looked into trb and its diseases too much yet, maybe cementing changes the equation, dunno. if run into wall ill rip out either db or hdd

16:47

Logged on 2022-02-18 08:46:58 cgra: http://logs.nosuchlabs.com/log/asciilifeform/2022-02-17#1079773 << mech hdd with the bdbmay prove unfeasible, while still an interesting experiment

16:50

jonsykkel: cement dun affect the verification time of valid blox, but does cut down on the cycles wasted on unwanted ones during sync

16:51

from where does all the unwanted blox come from anyway?

16:52

jonsykkel: inv responses

16:53

normally you end up getdataing for whatever new items yer noad saw in incoming inv's

16:53

in hopes that 1 of'em is 'next' block

16:56

would think these invs are requested based on blokheight+1 or smth

16:56

barely looked at code tho so dunno how it works

16:56

jonsykkel: there's no involvement of blockheight in inv's or getdata's

16:57

i.e. you get sent a block , and noad attempts to fit it as the next one, and succeeds or not

16:57

how does peers know wat blok to send then?

16:58

(in ancient pre-trb, incoming blocks were buffered, asciilifeform abolished this, see the 'orphanage' vpatches / discussions)

16:58

jonsykkel: see the coad, there are 2 ways of requesting blox

16:58

*

asciilifeform must bbl

16:58

ye ill read relevant parts

~ 1 hours 19 minutes ~

18:18

... 230k.

18:22

cgra: was thinking, anuther (if ugly) form of 'parallel' would be to load'em 1st, ~then~ verify sequentially. imho not req'd tho, would likely save very little time on a reasonably-connected box

~ 52 minutes ~

19:14

asciilifeform: prolly just gotta take time to see experimentally/think through staring at the code, whether significantly still could lock-in to a slomo-peer

19:16

aha

19:16

hmm, i remember as if some trb peer used to be a slow-link one. could try making cement-mode start off with such a peer, and see if makes operator sad or not

19:16

cgra: i'ma leave mine going for at least coupla wks or until fully syncs

19:17

cgra: did try it w/ buncha randomly-picked peers, did not manage to get it stuck thus far

19:17

prolly worth moar deliberate efforts if folx have time ( asciilifeform sadly does not atm )

19:17