-
luigi1111w
branched. will tag tomorrow I suspect
-
selsta
thanks, yea tag tomorrow is important so that we can release on time
-
selsta
sarang:
trezor/trezor-firmware #1246 can you confirm that the HF date was communicated with ph4r05 (our Trezor contact)?
-
sarang
I had communicated the timeline to ph4r05 (and others, via GitHub):
trezor/trezor-firmware #1246#issuecomment-687156825
-
sarang
I also had separately emailed ph4r05 on July 19 with the same timeline
-
sarang
That email contained the expected dates, but of course not the specific fork height at that time
-
luigi1111w
Oxygen Orion?
-
nioc
luigi1111w: sadly yes
-
luigi1111w
-
luigi1111w
please check
-
sarang
Reminder that the weekly #monero-research-lab meeting starts at 17:00 UTC
-
sarang
.time
-
sarang
Didn't there used to be a bot in here?
-
sarang
.date
-
sarang
.time
-
camthegeek
no wowbux/monerobux in -dev
-
sarang
boo
-
sarang
Maybe it got annoying and someone booted it
-
camthegeek
he's in your lab though
-
sarang
ya
-
sarang
very helpful for datetimes
-
sarang
and s/one/two/ substitutions
-
iDunk
sarang: you probably misremembered.
-
sarang
Oh well!
-
dEBRUYNE
I will create a thread on Reddit for the deterministic builds results
-
selsta
I’m busy until tomorrow, maybe someone else can do repro builds + upload them
-
m2049r
i am having the issue that "sometimes" a transmitted tx is not seen in the pool and thus marked as failed - until the next refresh/update or it ends up in a block =>
gist.github.com/m2049r/8daf176822cf2d0763a403722d449330 (both client&daemon are 0.16.0.3)
-
m2049r
am i doing something wrong??
-
m2049r
is this an issue to report?
-
m2049r
this is stagenet, but i had it happen on mainnet as well
-
m2049r
with different daemons (mine & xmrto)
-
iDunk
-
selsta
m2049r: I have seen similar behavior but was not able to constantly reproduce it, I think it is Dandelion++ related
-
luigi1111w
one can always count on iDunk
-
m2049r
selsta: thanks. i hadn't seen this before upgrading to 0.16.0.3 daemon-side. how can i help in pinpointing the cause?
-
m2049r
it's annoying for the user to think their tx has failed (and maybe repeating it) only to find out it ended up in the next block a couple of minutes later
-
selsta
vtnerd: any idea if this is dandelion++ related?
-
selsta
^^
-
m2049r
it just happened again. i have the client log. the tx was not in the pool for a long time (>10s) and when it appeared, *that's* when it went to status failed. it stayed that way until it was mined.
-
vtnerd
how are you probing the pool for txes? in some cases the daemon should omit the tx from the pool list for privacy reasons
-
vtnerd
thats likely one source of funkiness
-
vtnerd
who was claiming it failed? the daemon or wallet? I don't recall a failed status in daemon although perhaps it gets reported that way in the UI
-
vtnerd
the ~10s blackout was almost certainly the dandelion++ stem phase
-
vtnerd
ah ok I see in the scrollback you mentioned "next refresh/update" - the algo seems to be working as expected but the "blackout" in the RPC might be causing issues
-
vtnerd
first question is this daemon RPC in restricted mode .. ? selsta m2049r <-
-
selsta
I have seen tx taking 2-3 minutes until they show in mempool
-
vtnerd
hmm, hit a blackhole (bug? someone on testnet probing things?) then tripped embargo timeout
-
vtnerd
one thing with testnet is there's a higher possibility of someone starting/stopping nodes, although I would expect that to be less likely if they accept inbound conns
-
vtnerd
if a node goes offline that was used in a d++ stem, its effectively a "blackhole" but not a bug or attack
-
selsta
was on mainnet
-
selsta
but like I said it was not reliably reproducible
-
m2049r
first saw it on mainnet and then checked stagenet where it also happens. but happened to me a lot on mainnet.
-
m2049r
using rpc-restricted-bind-port
-
Inge-
m2049r: slightly off-topic, I wonder if I hit sone kind of limit with Monerujo. one wallet when opened, scans and scans, but blocks to scan never counts down - but phone gets warm so is definitely doing something. all other wallets work fine. This one has a bunch of subaddresses (<100).
-
TheCharlatan
sigs looking good so far :)
-
iDunk
-
iDunk
^ that's a 404.
-
iDunk
5.9 seems to be the oldest release they now host.
-
iDunk
-
iDunk
Looks like that could work.
-
selsta
so.. retag?
-
selsta
we should set up our own mirror :P
-
TheCharlatan
there's still the bitcoin core mirror.
-
selsta
that only has some thing
-
selsta
things
-
selsta
we use the bitcoin core mirror
-
TheCharlatan
at least it has the qt archives.
-
selsta
oh it does?
-
selsta
so no retag?
-
selsta
good
-
iDunk
I'm looking at my builder/cache/common, and qt submodules are there. Idk from where, the download process gave 404s for qt stuff.
-
selsta
yea IIRC bitcoin core mirror has it
-
iDunk
Hmm, ok.
-
TheCharlatan
-
TheCharlatan
that said, I think I'll bump it 5.15.x, already have a patch for that more or less ready.