23600+ entries in 0.05s
mod6: np, thanks for your help.
mod6: at the end of the static build there are 2 ugly warnings, but they're just warnings. so no worries.
mod6: (130 = "strip" command)
mod6: while you're at it danielpbarron comment out line 130 of "auto.sh" so I can have you run one other command if you get through the static build.
mod6: i've never tested this on 32bit so this is a first.
mod6: danielpbarron: ok, change line 4 to: "32", line 6 to: "x86_32" and line 8 to: "linux-x86_32"
mod6: so in the end the downloading/patching of the v0.5.3 base code won't be necessary since it'll already be complete.
mod6: asciilifeform: because how we'll roll out the release: 1) I'll patch the source myself and put it in a 'bitcoin' directory 2) i'll place the 'auto.sh' in the same parent directory as 'bitcoin' directory 3) tar those up 4) end user will unpack archive and execute `auto.sh'
mod6: danielpbarron: yeah, that'll be a problem. what os/arch are you on?
mod6: ok so your amazon linux is running gpg v 1.4.10?
mod6: for all: keep in mind if you're not running the test on a x64 machine, you'll probably have to edit the vars on lines 4, 6 & 8
mod6: no, the perl script is a separate thing that grabs the v0.5.3 base code and then patches it. then `auto.sh' pulls bdb,openssl,and boost and builds 'em and then builds the static bitcoind
mod6: did you get through the perl script ok alf?
mod6: thanks for testing trinque
mod6: ben addded those, not exactly sure what they do. lol.
☟︎ mod6: ahh, just take off those params then. it doesn't need 'em i don't think.
mod6: it's a bourne shell script /bin/sh
mod6: so centos6 might have issues if gpg 2.x, so maybe i just need to get my head out of my ass and add some code for that.
mod6: and yeah, im using gpg 1.4.10 there too
mod6: oh, i guess im running a deb6 ami up there.
mod6: decimation: anyway, it totally works fine for me on my aws machine.
mod6: asciilifeform: this is the correct key to pull for you right?
mod6: it seems, everyone has a mixed bag so it's hard to account for all of that i think.
mod6: everyone will start with an already patched codebase, unless they want to patch it manually or wrestle with one of the 2 scripts we have for this purpose
mod6: i'll use it to patch the source code, then i'll bundle up the release.
mod6: eh, all around it's not much of an issue with the perl script, because no one will run that ever anyway.
mod6: thanks for giving it a shot.
mod6: decimation: ok. that's fine. not sure what it's deal is.
mod6: ugh. anyway, i was hoping this would be the /easy/ part and we could actually test the part that needs testing lol.
mod6: if you run it again and you don't have $FULL_CLEAN set to "TRUE", you'll need to delete ~/build/.gnupg manually.
mod6: can you change expiry on that guy for like 1 year?
mod6: decimation: huh, just looked at that error. must be something with 2.x version i'll have to fix0r.
mod6: i use pastebin more often because it does forever-pastes .. guess.. i just have to set dpaste for a year
mod6: yeah, must be pastebin :/
mod6: root@debian-test:/home/mod6/build-test-3#
mod6: root@debian-test:/home/mod6/build-test-3# xxd build-v009.pl | grep "0d0a"
mod6: decimation: fair. i'll change that next time through.
mod6: oh we're all discussing them pulling down the patch files, verifing the patch sigs and patching the base code
mod6: asciilifeform: what script was that?
mod6: can you post those decimation?
mod6: 313 if($_ =~ /^.*Good signature.*$/) {
mod6: i suppose it could be some weird regex failure... although, what danielpbarron pasted as a manual verify should have matched this:
mod6: oh that wasn't from the log output. ok. i wonder why it would puke for you guys, but not for me... although, im running 1.4.10
mod6: what's weird to me... is that the log output that danielpbarron pasted, it says "good signature"
mod6: hmm, still, should be ok. i just re-ran on my side and got: >> Signature verified for bitcoin-asciilifeform.1.patch.sig
mod6: what does `gpg --version` say?
mod6: did you guys make new dirs to build this stuff in?
mod6: unless you ran it from ~/
mod6: it /shouldnt/ matter because it puts a .gnupg dir in the place where you run the script from. we didn't want it to diddle with your normal keyring in ~/
mod6: but i don't ever get that error... although, someone else has seen it before iirc.
mod6: ok... so. lets see here. i do think that ascii updated his key since then...
mod6: we'll keep the noise outta here, unless you want ... NOISE :)
mod6: ima invite you to #btcf
mod6: yeah, makefiles are touchy
mod6: yup, it just simply utilizes the makefile. but this way a person only has to run `auto.sh`
mod6: what's changed? 1) added some checks to see if boost,bdb,openssl are already built, if so, skip re-downloading & building 2) added a 'clean' command line parameter so you don't have to do this manually
mod6: so i made some additional changes to the modified pogo script. anyone wanna help me test quick? I tested it myself, but it'd be cool if someone else can verify.
mod6: <+mircea_popescu> now if ben_vulpes an' mod6 get this show off the ground, it'll be reheheheally fascinating in short order. << just wrapping up the regression of the static binary:
http://pastebin.com/raw.php?i=ny9m6Q3R and after this, we just need to bundle up this thing and sign patches & tar ball. gettin close!
mod6: mircea_popescu: is someone coming to pick me up from the aeropuerto tho?
mod6: i wont be there until like 9-10am friday morning. i leave on thursday afternoon.
mod6: jesus cristo, $160!?
mod6: mp: you run .ar now right? just tell 'em im comin and it's all good.
mod6: <+asciilifeform> the ar entry fee gizmo, that is << lol wtf, you have to pay a 'reciprocity fee'?
mod6: up to height=288287 (static build)
mod6: ok, im off for a bit. night #b-a!
mod6: then that'll be another version bump & release in itself.
mod6: i understand. i think the first objective will be to resolve that issue once we get the milestone "release" bundled up.
mod6: i'll check back in tomorrow.
mod6: let me know how it goes. my full-sync test with the static bin is up to height=273236 currently.
mod6: thanks asciilifeform :]
mod6: jurov: so as far as shorter release cycles, i think going forward they will be. the foundation will focus on say, one specific problem and then tackle it, patch it, and then regression test, release.
mod6: when we release there will be a formal annoucement to the mailing list, signed patches by ben & I, and an included signed tar of the patched source code with release notes.
mod6: haha. ok. but i don't see a problem with testing our own thing. even if no one else uses this, i'd like to know if it works or not.