<PaperSword>
Large flushes can be confused with hangs since bitcoind pretty much stops logging anything while the flush is invoked.
andrew_mo_ has joined #bitcoin-core-dev
ZeroMaster has quit [Remote host closed the connection]
<PaperSword>
threshold based logging could be another way to handle this better as well. Flushing 300MB to disk doesn't take long enough for anyone to notice but it does happen frequently. Large flushes 2GB+ even on SSD takes enough time to make users question if there client has crashed
flooded has joined #bitcoin-core-dev
szkl has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
test_ has quit [Ping timeout: 245 seconds]
Nekorand has joined #bitcoin-core-dev
ZeroMaster has joined #bitcoin-core-dev
ZeroMaster has quit [Remote host closed the connection]
andrew_mo_ has joined #bitcoin-core-dev
justache is now known as reddit-bot
reddit-bot is now known as justache
andrew_mo_ has quit [Ping timeout: 255 seconds]
Ara has quit [Ping timeout: 258 seconds]
Ara_ has joined #bitcoin-core-dev
Ara_ is now known as Ara
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 255 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
DarrylTheFiish has joined #bitcoin-core-dev
DarrylTheFish has quit [Ping timeout: 250 seconds]
greypw2546002161 has quit [Quit: I'll be back!]
greypw2546002161 has joined #bitcoin-core-dev
andrew_mo_ has joined #bitcoin-core-dev
<jamesob>
PaperSword: absolutely; I think provoostenator (Sjors) encountered this recently when testing assumeutxo. I hit the same thing yesterday; I assume it was hung during shutdown, but it was just flushing.
andrew_mo_ has quit [Ping timeout: 246 seconds]
<PaperSword>
It might be a bit noisy, is there way to log the flush as a % after a threshold just as validating the n blocks on startup does?
andrew_mo_ has joined #bitcoin-core-dev
<PaperSword>
But overall even a single log letting the user know that a flush is occurring?
andrew_mo_ has quit [Ping timeout: 255 seconds]
Guest97 has joined #bitcoin-core-dev
andrew_mo_ has joined #bitcoin-core-dev
Guest97 has quit [Quit: Client closed]
andrew_mo_ has quit [Ping timeout: 246 seconds]
DarrylTheFiiish has joined #bitcoin-core-dev
DarrylTheFiish has quit [Ping timeout: 240 seconds]
andrew_mo_ has joined #bitcoin-core-dev
AaronvanW has quit [Remote host closed the connection]
StayWoke_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 245 seconds]
danon has joined #bitcoin-core-dev
AaronvanW has joined #bitcoin-core-dev
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 255 seconds]
AaronvanW has quit [Quit: Leaving...]
andrew_mo_ has joined #bitcoin-core-dev
danon has quit [Ping timeout: 250 seconds]
andrew_mo_ has quit [Ping timeout: 246 seconds]
andrew_mo_ has joined #bitcoin-core-dev
Nekorand has quit [Ping timeout: 245 seconds]
andrew_mo_ has quit [Ping timeout: 244 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
test_ has joined #bitcoin-core-dev
flooded has quit [Ping timeout: 245 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
szkl has quit [Quit: Connection closed for inactivity]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
andrew_mo_ has joined #bitcoin-core-dev
jespada has quit [Ping timeout: 255 seconds]
boris has quit [Ping timeout: 246 seconds]
boris has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 255 seconds]
jespada has joined #bitcoin-core-dev
DarrylTheFiish has joined #bitcoin-core-dev
DarrylTheFiiish has quit [Ping timeout: 246 seconds]
andrew_mo_ has joined #bitcoin-core-dev
Nekorand has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 245 seconds]
andrew_mo_ has joined #bitcoin-core-dev
test_ is now known as _flood
DarrylTheFiiish has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
DarrylTheFiish has quit [Ping timeout: 245 seconds]
jespada has quit [Ping timeout: 255 seconds]
jespada has joined #bitcoin-core-dev
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 245 seconds]
<hebasto>
re "or figure out why the CI tasks keeps failing when it shouldnt be and fix it :)" -- fixed in master
Guest15 has joined #bitcoin-core-dev
<instagibbs>
PaperSword a single log saying it's happening plus size would be not very noisy and helpful for debugging it sounds like
Guest15 has quit [Client Quit]
Guest15 has joined #bitcoin-core-dev
Guest15 has quit [Client Quit]
<sipa>
with -debug=bench there is a log entry for every flush
<jamesob>
I think it's probably worth doing if it's flushing a nonzero number of coins
andrew_mo_ has joined #bitcoin-core-dev
<jamesob>
pretty big usability issue that shutdown can just hang for minutes with no explanatory log under normal circumstances and with default args
<sipa>
maybe one day i'll pick up my background flushing design and implement it :)
andrew_mo_ has quit [Ping timeout: 255 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_m_ has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 248 seconds]
andrew_mo_ has joined #bitcoin-core-dev
andrew_m_ has quit [Read error: Connection reset by peer]
<jamesob>
sipa: I think sdaftuar and I went back and forth about this before you got to ccl... very tricky IIRC
<jamesob>
I don't doubt you have something workable for it though :)
<sipa>
i'm pretty sure i have a design that works, but convincing others that it does (and testing it...) may be hard :p
<sipa>
happy to whiteboard it out for you some day, but not over IRC
<jamesob>
does it involve leveldb snapshots?
<sipa>
no
<jamesob>
👍
pablomartin4btc has joined #bitcoin-core-dev
upekkha has quit []
upekkha has joined #bitcoin-core-dev
ghost43_ has joined #bitcoin-core-dev
ghost43 has quit [Ping timeout: 246 seconds]
qxs has quit [Ping timeout: 246 seconds]
qxs has joined #bitcoin-core-dev
<PaperSword>
"maybe one day i'll pick up my background flushing design and implement it 😁" do you have a patch I could use? genuinely excited about this!
Nekorand has quit [Ping timeout: 246 seconds]
pablomartin4btc has quit [Remote host closed the connection]
bugs_ has joined #bitcoin-core-dev
<sipa>
haha, no, it has never been implemented
<sipa>
or experimented with
<sipa>
just an idea
Victorsueca has quit [Ping timeout: 246 seconds]
pablomartin has joined #bitcoin-core-dev
pablomartin is now known as pablomartin4btc
dviola has quit [Quit: WeeChat 4.0.4]
andrew_mo_ has quit [Remote host closed the connection]
pablomartin4btc_ has joined #bitcoin-core-dev
pablomartin4btc has quit [Ping timeout: 246 seconds]
Nekorand has joined #bitcoin-core-dev
andrew_mo_ has joined #bitcoin-core-dev
dviola has joined #bitcoin-core-dev
<dviola>
is there a way to migrate from bdb -> sqlite (descriptor wallet) ?
andrew_mo_ has quit [Ping timeout: 255 seconds]
<dviola>
bdb is a headache and I would like to drop it if possible
<sipa>
migratewallet ?
andrew_mo_ has joined #bitcoin-core-dev
<dviola>
sipa: I haven't tried that yet but I will take a look, thanks
<dviola>
last time I checked it was still experimental, I think
<instagibbs>
i think in this case experimental means don't count on the API to be stable, one off manual should be fine(or something is horribly wrong)
andrew_mo_ has quit [Ping timeout: 255 seconds]
<furszy>
dviola: while you haven't done something out of the ordinary stuff in your legacy wallet. The process should work fine.
<furszy>
sounds good. Try it on latest master. IIRC, there are other bug fixes that haven't been released yet.
<furszy>
(28067)
<dviola>
ok
andrew_mo_ has joined #bitcoin-core-dev
achow101 has quit [Remote host closed the connection]
andrew_mo_ has quit [Ping timeout: 245 seconds]
csknk has joined #bitcoin-core-dev
<real_or_random>
Hm, does someone have an overview over the status of Cirrus CI on the bitcoin-core github org?
<real_or_random>
All repos in that org share the same free limit of 50 credits now
<real_or_random>
Some of them have recent builds, some don't ... But I haven't them individually, and I haven't checked if some of these have migrated to self-hosted runners. I guess we'll run into issues this month, given that we used 350 credits in Aug
andrew_mo_ has joined #bitcoin-core-dev
achow101 has joined #bitcoin-core-dev
andrew_mo_ has quit [Ping timeout: 246 seconds]
<dviola>
should it matter if the node is running pruned or not (for the migration process) ?