Hi again,
I would really appreciate some feedback from the smartfox team on my issues.
At the moment (after 3 days uptime) I got 1050 CCU wich is the same as the number of sessions displayed on the admin panel, but I got 2600 "users logged in". I really don't understand what can cause this issue.
When I start the server, both numbers are really close.
Thanks,
Sebastien
Live deployment of SFX2x (2000+ CCU) - Report and Questions.
Hi Lapo,
Thanks for the quick response, but can you also checkout my other questions ?
I have tweaked the various threads values but the server still "lock" at some point, not processing any new incoming connexion, with a bunch of:
Those exceptions never happen when the server is running smoothly.
Can you point me in the right direction or give me some clues about where to look?
Thanks,
Sebastien
Thanks for the quick response, but can you also checkout my other questions ?
I have tweaked the various threads values but the server still "lock" at some point, not processing any new incoming connexion, with a bunch of:
Code: Select all
18 Jul 2011 14:40:17,474 WARN [Scheduler1-thread-1] bitswarm.sessions.DefaultSessionManager - Failed closing connection while removing idle Session: { Id: 164, Type: DEFAULT, Logged: No, IP: 192.168.2.82:11149 }
Can you point me in the right direction or give me some clues about where to look?
Thanks,
Sebastien
Hi,
Our server is running on a remote linux. I cannot run any graphical tool.
Can you explain to me exactly how I can provide you with the necessary information ?
I dont know if it's of any interrest, but we noticed using strace during an issue that one of the java thread was cycling very very quickly with E_TIMEOUT on the same memory address, nothing like other threads.
I'm not a specialist of thoses low level stuff, so I may be saying something stupid there.
Thanks,
Sebastien
Our server is running on a remote linux. I cannot run any graphical tool.
Can you explain to me exactly how I can provide you with the necessary information ?
I dont know if it's of any interrest, but we noticed using strace during an issue that one of the java thread was cycling very very quickly with E_TIMEOUT on the same memory address, nothing like other threads.
I'm not a specialist of thoses low level stuff, so I may be saying something stupid there.
Thanks,
Sebastien