22400+ entries in 0.046s
mod6: And as far as the gentoo stuff goes, I kinda had to put that on hold for a minute. I'm going to finish the rest of that up on real hardware. But to do so, I gotta drive across town and buy a new box. I might wait until after the 1st to continue this front.
mod6: s/regression/performance/
mod6: aside from keeping up with the unfolding nosuchlabs work, within 7000 blocks of my next regression benchmark being finished. should have some new charts to look at in 24 hours.
mod6: oh yeah, i saw that on outside sites.
mod6: how do you know they're subkeys? did I miss this in the log?
mod6: 8 broken keys now eh?
mod6: proof in the pudding :]
mod6 falls out of his chair
mod6: im stunned. very impressive.
mod6: <+mircea_popescu> yes. once keys are factored we have the privkey. << holy shit o.0
mod6: lol, it was hard not to comment on that
mod6: danielpbarron: what block are you on currently?
mod6: ok, it's in that same directory: 'nmon_bitcoin-v0_5_3_1-RELEASE_plus_OrphanageThermonukePatch.txt'
mod6: im throwing that raw nmon file up there for you now, incase you wanna see it.
mod6: ah, ok. i'll try to figure out how to capture that.
mod6: it should be /fairly/ accurate as there isn't anything else running on this box. just the usual: sshd/syslog/screen/cron ..
mod6: it's a work-in-progress
mod6: yeah, i'll have to figure out how to use either nmon or another tool to capture /just/ the bitcoind process.
mod6: we're getting closer, maybe 24 hours away; it's on block: 322683
mod6: if you want it, just holler.
mod6: well, actually, i'd post the orphanage-thermonuke patch test data now, but it's 113 mb of raw nmon captures.
mod6: make sense? i'll post all the data with a posting about the findings.
mod6: Now, currently, I'm running a very similar test with v0.5.3.1-RELEASE (as a baseline) without your OrphanageThermonuke patch included... it did oomkill once, yesterday.
mod6: so yes, the first performance test was me running v0.5.3.1-RELEASE with Orphanage_Thermonuke patched in. Performance test conducted with `vmstat 1` & `nmon -f s3 -c1000000`. During this test the entire sync process completed without any oomkill.
mod6: <+danielpbarron> the node to which i'm referring runs 0.7.2 though, so might not be the same issue << as ascii said, this is apples to oranges, but I'm curious anyway; where abouts are you at in the sync process on this particular device? (what block height range? 250`000 - 290`000) ?
☟︎ mod6: now, it'll be interesting to see the numbers i.e. network usage (among other utilization metrics) between full sync of v0.5.3.1-RELEASE+{Orphanage_Thermonuke} & v0.5.3.1-RELEASE (currently running -- which also, as i was saying lastnight, already oom killed once)
mod6: <+asciilifeform> ;;later tell mod6 so orphanage-thermonuke has never oom-crashed in your tests ? << nope! it was pretty exciting to see it get all the way through full sync without oomkill.
mod6: Once I have all the baseline perf metrics collected, I'll be making a posting about it and will publish the data.
mod6: Addtionally, I 100% agree, if anyone else should test either of these and capture anything surrounding the OOMKILL either in the v0.5.3.1-RELEASE or a patched version of v0.5.3.1-RELEASE, please let us know.
mod6: <+asciilifeform> mod6: please post any data you may have collected at the moment of the oomkill << so just to reiterate here, the current perf test I'm running is with the v0.5.3.1-RELEASE which oomkill'd (as it's known to do). I'll post the nmon charts, log, and vmstat log. no core file was created. However, as a reminder, the previous perf test that I ran with v0.5.3.1-RELEASE+{asciilifeform_orphanage_thermonuke.patch}(
http://thebitcoin.foun mod6: +1 for hanbot's "Narf" - pinky & the brain ref.
mod6: insane dipshit madlibs is right.
mod6: Speaking of Blazing Saddles... that movie is so damn funny. I must have watched it 20x as a kid. But what's interesting was recently I heard/read about these "Animal Trials" in Europe during the middle-ages. Makes me think of that part where they're hanging that guy AND his horse. lol
mod6: they have McBucks now 'eh? During february, clown-burger up here actually let you pay in ``hugs''.
mod6: ``That's one small step for man, one giant leap for every, complaining, son of a bitch on the face of the Earth.'' << lol!
mod6: Then I can compare and contrast and work with ben on a writeup for qntra
mod6: After this one is done, I was going to run a vanilla v0.5.3 to get a baseline from that as well. And I will do that, but did remember that it /does/ at least require the db patch to achieve full sync.
mod6: there will be some flat lines in the charts, but should still be able to get some sort of picture showing us the diff in network traf & other utilization metrics.
mod6: im getting closer to my v0.5.3.1-RELEASE perf full bc sync baseline being finished; currently on block 304969 -- it did oomkill today though, and didn't see it for like 8-9 hours :/
☟︎ mod6: ah. well it got bored maybe.
mod6: thing was cranking through the keys nicely.
mod6: was just gonna check on the phuctor stats
mod6: ah, ok. thought /maybe/ i saw that in the log.
mod6: speaking of which... is nosuchlabs.com down?
mod6: asciilifeform: ahh, cool, well, hopefully you'll get to play 'em again
mod6: remember kings quest? that was a 486 game iirc
mod6: thats when they call in the MOAR.jpg cake eating chick to sit on his face.
mod6: thought it was just like some occult dining ritual
mod6: ``Strong and Tough Sentence''
mod6: haha, what's on his chest? sushi rolls?
mod6: just as a pre-lizard-hitler fabrication that still works kinda thing?
mod6: asciilifeform is gonna be like the guy in cryptonomicon who has the technology museum
mod6: oh no. yeah, it was probably thrown up and down, left and right. gotta have a solid box & styrofoam enclosures.
mod6: ah, yea. send it back, get yourself a working relic.
mod6: huh. yeah, maybe you got scammed?
mod6: well.. the disks didn't usually fair so well, but. yeah.
mod6: back then they built that stuff to last.
mod6: hmm, i had a PB back in like 94 that was like a P1 -- had heatsink only. might have had a case fan tho.
mod6: you're not just going to re solder all the outputs?
mod6: so, basically, no. i'd like to keep our current ramshackle.
mod6: haha, agree. and like i said `if we can figure out a scheme inwhich it would work', and that would imply that it would have to somehow work better than what currently ``works''.
mod6: yeah, im not sure how we'd block posts from people like we do with mailman
mod6: <+jurov> mod6 ben_vulpes asciilifeform why don't we throw out the mailing list and move to 8chan? <+asciilifeform> jurov: move to 8chan? << share the lsd ? << as much as the turdolator has been a pain for jurov, I've kinda grown fond of it. however, I'm open to discuss it a bit if we can figure out a scheme inwhich it would work. mainly, ive just never been very into bbs.
mod6: they grow a lot of cotton out there.
mod6: aight well, have to check later. bbs.
mod6: <+jurov> can't we just use namecoin? << yeah, I thought maybe this was addressed in the logs before...
mod6: <+jurov> but gpg pubkeys are not practical for replacing DNS. what if you want to tranfer the name? << then there would be a deed detailing the transfer of the name to a new key fingerprint?
mod6: In this case, wouldn't you just be called "Juraj Variny (jurov) <rini17@gmail.com>"
mod6: decimation: hrm. could be i guess. i'll check into that.
mod6: this is bizzare. i can only get the ssh keys to work /after/ I set a password on the user acount on the instnace (in /etc/shadow). if the user doesn't have a password set, just asks for a pw when connecting. (gentoo uclibc hardened) am i missing something obvious?