Daemon stall due to threading

[Gingeropolous]

Apparently this is some threading issue. I'm running Lubuntu 64 bit, last update was ~4 weeks ago. bitmonero v0.8.8.7-1720aff.

System behavior: bitmonero, when working normally (and mining on one thread), varies between 100 - 120 %CPU. When this bug manifests, it stays at a steady ~200%. %mem is 6.5% on a 4 gig box.

The print_cn is where the double threading can be observed, with the two headers.

print_cn
2015-Jun-21 16:02:58.945555 Read command: print_cn
2015-Jun-21 16:02:58.971285 Remote Host Peer id Recv/Sent (inactive,sec) State Livetime(sec) Down (kB/s) Dow
n(now) Up (kB/s) Up(now)

Remote Host Peer id Recv/Sent (inactive,sec) State Livetime(sec) Down (kB/s) Down(now) Up (kB/s) Up(now)

[Gingeropolous]

Additional notes: doesn't seem to manifest on testnet. Ran for 24 hours without stall while mining on testnet.

Might be due to mining: ran for 24 hours on main-net without mining and it didn't stall.

[Gingeropolous]

sent a bunch of transactions on testnet, still no problem.

[Gingeropolous]

Just happened on v0.8.8.7-1720aff (2 print_cn headers), but simplewallet is able to synchronize and the daemon is still functioning.

edited to add: nevermind - simeplwallet stalled as well. Was crawling (still refreshing) but ultimately became impatient.

###FASCINATING!!!!

If you try to exit the daemon (using the exit command) you can kill the rogue new thread! I went from 2 print_cn headers to 1. Oh lemme see if I can't grep this. Bah. set_log 2 is just too much.

[Gingeropolous]

new daemon seems to have fixed this