@eyedeekay
&eche|on
&zzz
+FreefallHeavens
+R4SAS
+RN
+Romster
+T3s|4
+acetone
+cims
+dr|z3d_
+hagen
+nilbog
+orignal
+postman
+qend-irc2p
+snex
+wodencafe
Arch
Dann
Holmes
Irc2PGuest12976
Irc2PGuest1526
Irc2PGuest28234
Irc2PGuest32287
Irc2PGuest51098
Irc2PGuest57041
Irc2PGuest88488
NiceBoat
OfficialCIA
Onn4l7h
Onn4|7h
Over
SilentWave
Sisyphus
Sleepy
T3s|4_
Wikk__
Zapek
aargh4
ac9f
ahiru
ananas_
aqua
calamares_
dr4wd3
duanin2
fa
i2potus
leopold
mahlay
makoto
n2_
not_bob_afk2
poriori
pory
profetikla
r00tobo_BNC
rapidash
test3847473
thetia
uop23ip
user1
vivid_reader56
x74a6
zelgomer
dr|z3d
got a deadlock for you, zzz. twice today..
dr|z3d
Found one Java-level deadlock:
dr|z3d
=============================
dr|z3d
"SimpleTimer2 1/4":
dr|z3d
waiting for ownable synchronizer 0x00000000fff05310, (a java.util.concurrent.locks.ReentrantReadWriteLock$NonfairSync),
dr|z3d
which is held by "SimpleTimer2 3/4"
dr|z3d
"SimpleTimer2 3/4":
dr|z3d
waiting to lock monitor 0x00007fa2a8139350 (object 0x00000000fbbfbda0, a java.lang.Object),
dr|z3d
which is held by "UDPPktHandler 1/4"
dr|z3d
"UDPPktHandler 1/4":
dr|z3d
waiting for ownable synchronizer 0x00000000fff05310, (a java.util.concurrent.locks.ReentrantReadWriteLock$NonfairSync),
dr|z3d
which is held by "SimpleTimer2 3/4"
dr|z3d
stacktrace: cake.i2p/view/KSDA0hp56l_KwxV7YSbJ87IzVP8wIldrFgVJPIdPu_djPxkU2tvd/KSDA0hp56l.txt
zzz
sigh...
dr|z3d
> Invalid DbStore attempt - RouterInfo [8Igqrr] was published 4y in the future
dr|z3d
clock skew? :)
zzz
dr|z3d, I'm a little stumped on that deadlock because the first two threads aren't holding any locks at all
zzz
2 is waiting for 3. 1 and 3 are waiting for somebody not listed
dr|z3d
didn't we have a similar SimpleTimer2 deadlock before that you fixed? but yeah, beats me.
dr|z3d
there was some tangential information about GC, but I doubt that's relevant.
zzz
ok, must be a logging glitch. thread 2 has the write lock in locked_rebuildRouterInfo
zzz
maybe a quirk of ReadWriteLocks
dr|z3d
ok, this is getting a bit surreal: > Invalid DbStore attempt - RouterInfo [MGpD1~] was published 12y in the futur
zzz
eyedeekay, thanks for fixing my butchery of the blog post
zzz
for some reason the undraft didn't 'take' though
zzz
launchpad and our deb repo updated
zzz
eyedeekay, also, up until last year, month and day directories were always two digits e.g. 06; you started the one digit trend this year
zzz
so the links are a little messy. but I don't think that's the problem
eyedeekay
Yeah I noticed that yesterday when I was doing the reddit post, going to make sure changing it doesn't break anything then git mv them because of that effect
eyedeekay
Also still got Android to finish up, probably another couple hours there
zzz
git move or rename the directories? noooooooo it will break all the links
zzz
I already tested. you can't do /08/ if the directory is /8/
eyedeekay
That is what I was worried about
zzz
unless you add some rewrite rules
zzz
it's messy because there's both month and day single digits
zzz
all this year plus one last year
eyedeekay
Maybe I'll do rewrite rules to fix it up later then, right now I'm pretty focused on making sure Android is ready
zzz
so why wasn't the post undrafted? it is live at the link
zzz
oh wait. it is in there
zzz
it's just not sorted right
zzz
it's the 3rd entry
eyedeekay
Yeah that's why I noticed it and was going to try and move it
zzz
oh, because 4 > 3 > 22
eyedeekay
But until I can rewrite the links I can't do that
zzz
I guess it's not sorting numerically
zzz
ok, all makes sense now
zzz
the first problematic one is 2021/11/2/, then everything this year
eyedeekay
I'll get them fixed, either by changing how it sorts or rewriting the URL's and moving the misplaced ones
zzz
I was doing fine with two digits but then just blindly followed what you were doing :(
zzz
blog readme updated to specify two digits