~dr|z3d
@RN
@RN_
@StormyCloud
@not_bob_afk
@orignal
@postman
@zzz
%Liorar
%acetone
+BeepyBee
+FreefallHeavens
+H20
+Onn4l7h
+Over
+Sh0ck
+bak83_
+leopold_
+r00tobo
+uop23ip
+xHarr
Arch
BubbRubb
Danny
Debian_GNU_Hurd_2025_released
DeltaOreo
Irc2PGuest16752
Irc2PGuest33667
Maylay
Meow
T3s|4_
ac9f_
anontor
b4dab00m
duck
john231
makoto
mareki2p_
nZDoYBkF__
nilbog-
ntty`
poriori_
profetik1
r00tobo[2]
shiver_
simprelay
solidx66
thetia
tmg86
u5657
vivid_reader56
zer0bitz
dr|z3d
shiver: zzz's LookupDestJob replacement patch is now available in the latest /dev/ build if you want to try it. *if* the issue with spikes was as a result of obscuratus' patch, this may fix it.
shiver
dr|z3d, thank you, will update!
dr|z3d
sure thing, let me know how you get on!
shiver
after disabling floodfill i still got 1 sec. spikes over a short time and 500ms over 5min.
dr|z3d
ok, well, one way or another we'll find a solution, if not zzz's patch.
shiver
restart brb
shiver
i enable floodfill again
dr|z3d
ok
dr|z3d
shiver: there's another config you can try if things are still looking spikey. netdb.maxConcurrent=1
dr|z3d
I'm also tweaking the IterativeSearchJob params here to reduce the number of concurrent searches.
dr|z3d
new /dev/ build up.
shiver
80min uptime and no spikes but cant say for sure
shiver
4h was last time between spikes
dr|z3d
ok, could be positive. the latest build and/or the config above should help more.
shiver
afk, maybe 1h