.merge+ 7292 7285

Added

.merges

Snipa: luigi1111w: please do merges and tag v0.17.1.9 afterwards

.merges

All clear for all merges, or do I need to watch my step again selsta?

all good

Tag is hot and spicy.

And properly signed this time. :)

^ for those doing reproducible builds

I just had this paste.debian.net/hidden/cca52451

Maybe a fluke, restarting.

hmm weird

depends CI did build fine for all OS as far as I can see

as far as I can see none of the code in make_new_connection_from_peerlist has recently changed

It built on the second run.

Copyright (c) 2014-2020 The Monero Project year is wrong

When starting up monerod, I got the idea of using two flags--ban block_tor.txt and --enable-dns-blocklist at the same time. Good idea or not?

Thought it would make a more profound filter of bad i.p.'s out there.

one-horse-wagon: you can use both

Terrific!

i got "failed to find tx meta: <2cc1e94f " on .9

yes, this exists for a while

ok, thanks. just confirming its something familiar.

yea it is harmless and we should fix it in the future

ok. xmrchain is on self compiled .9

Franck[m]: daemon has "print_tx" function, not sure about RPC

Thanks selsta , good to know. At this stage I assume that my software only has access to `monero-wallet-rpc` and was hoping to find a way so I can keep this infra requirement.

all right 1:18 of uptime

My reproducible build hashes: paste.debian.net/1180151

just started building here

running .9 with block.txt (no tor blocking), so far so good

sech1: Does it also run properly without the block.txt file?

I can try

running without the block list

I would guess it's pretty effective, but the auto-bans only last 24 hours

I see Host unblocked and then a minute later Host blocked

same hosts keep coming back

2021-01-08 07:34:07.789 I Host 49.51.155.129 unblocked.

2021-01-08 07:40:57.758 I Host 49.51.155.129 blocked.

Chinese IP, could be that they have horrible latency.

I also constantly see that IP, it is not an attacker.

oh well

I see one Chinese IP too, constantly getting unblocked and then blocked

on the other hand, it happens only once a day, so I don't have much action in the logs anyway

how does it get unblocked so quickly?

?

it's unblocked after 24 hours, but blocked quickly

ah ok

speaking of blocks: paste.debian.net/hidden/fbf7b394

FUK's nodes are getting blocked

sech1: thanks

ok my linux builds just finished, matches sech1 hashes

android building now

there are also iDunk's hashes: paste.debian.net/hidden/d4dd48e2

mine match 100%

cool

if one of ours didn't match at this point I suppose it would mean one of our machines was compromised

Or unstable overclock :D

heh

ok I just started it up on my VPS, without ban list

My PC is overclocked btw, but it's stable enough to run xmrig 10M benchmark correctly several times and pass Testmem5 anta777 extreme preset

insta reboot with Prime95 though :D

Sashed from my build: paste.debian.net/1180164

*Hashes.

so far so good. android & freebsd matched too. windows building now

Please sign your hashes when you paste them, otherwise it's only as solid as your IRC account security

PR up, includes my key, same as the one used to sign the main Monero repos.

How to sign? I don't even have keys generated

sign with a key that is in the gitian repo

if you have one

I probably don't

useful without anyway no biggy, just better for those who can

Resubbed w/ signature and XMR block hash for time/date verification.

I'll try to add my keys for the next release

Is it done here github.com/settings/keys ???

Fork the main gitian.sigs repo, PR w/ your key

Ala: monero-project/gitian.sigs #130/files

monero-project/gitian.sigs #131/files is mine that includes my key with this build.

I'll try later today

I only started doing these builds because I didn't want to wait for the release :D

Trying to be more active and not work 24 hours a day at the $dayjob.

I wonder what version of MacOS SDK we'll need to build Apple ARM64 binaries

all my builds done, hashes match

bitcoin/bitcoin #20880 <-- this is also something we can look into for monero

though looks quite complicated, not sure if worth the effort, we already sign the GUI anyway

Binaries for new version 0.17.19 are available on getmonero.org, great job everyone!

*0.17.1.9

xmrchain uptime 0d 8h 46m 53s

on .9

gingeropolous: "The most up to date software for Monero are version"

looks like this can also be updated :D

yeah yeah yeah

there we go :)

Thanks for all the hard work on .9, everyone! Running great on all my nodes without the Tor banlist so far.

I'd say, that the parallel tests are ready from my side. Everybody's points have been addressed one way or another.

mj-xmr_: dumb question maybe (because I haven't looked at it yet) but it doesn't parallelise performance tests right?

uhm... good point

it does.

But they can be launched sequentially as well. The framework allows this.

for CI performance tests are not that interesting

at least I don’t look at them

I mean, for CI we can skip performance tests for sure

(unless we want to have a performance pipeline to pick up regressions, but we're not exactly ingesting test data in graylog or anything)

I see, that the master doesn't call performance tests in the CI either (not through ctest at least)

22 tests in both cases, and none of them is performance test, or is it a range of tests with some name prefix?

might be that make release-test does not run them

monero-project/monero #7289/checks?check_run_id=1663049254 Click on Tests, and page down

To see the test logs

"make test" (which is an alias for ctest), only calls these executables, which are tagged as add_test(TARGET)

If the performance tests are python scripts, or sth, they have no chance of being picked up by ctest.

ok, then we can ignore it anyway

the server setup / hardening guide is up

would love any additional input from infosec people in the room

cc SerHack

I'm giving up on the proposed ML based ban solution for now. Firstly I see, that you're doing a great job on fighting off the attack, and secondly there's still a lot to do on the CI part, and this is my bag of tea.

fluffypony: That's very useful. Thanks!

alll right 12 hrs of uptime on xmrchain.net

without banlist?

with banlist

i might as well fully test in production eh?

all right, flyin hot

i've been running without banlist for 6 hours. so far onlay banned 1 IP, the one from CHina

My node keeps banning FUK's nodes, 71 so far after 7 hours

oh. time for .9?

it's out already :-P

M5M400 ^

want me to spin it up on pubnode without banlist?

updated

fluffypony: whoa! it's well-written

hay could the link to the binaries get fixed so that it is not cached please

mnt_grrrl: I believe they are fixed now, but you may still be affected by old ones that your browser has cached

oh ok thank you!

mm good thing I checked, it seems to work for gui links but not cli

oh

fixed now hopefully

If you do curl -I downloads.getmonero.org/cli/mac64 or another one, you should see an "Expires:" field that should show the current time

Hello, do you believe that the new law for crypto financial regulation will push more users to monero?

not a development related question. please try #monero or #monero-policy

/j #monero-police

ouch