-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Vicky,
On 11/5/12 5:12 AM, vicky007aggar...@yahoo.co.in wrote: > Mark actually i have a scenario in which my application receives > variable amount of request depending on time . So at one point i > have an approx around 400 users concurrently accessing my > application as a result i have an approx 400 threads but when the > request count decreases the thread count still remains 400 which > eventually eat up my system resources. If you have 400 threads running, your resources are already "eaten". Does anything get worse over time? Honestly, if your system can't handle its peak load, then you need to make other arrangements (such as load-balancing, or lowering the thread pool size at the expense of average response time). In a default configuration, Tomcat should drop idle threads after 1 minute of idle time per thread (so 60 seconds + some slop after a thread handles its last request, the request-processing thread should die). If that isn't happening, please post your configuration and your observations (and be *very* specific, including your methodology for counting live threads, the timestamps at which you applied the user load and sampled the live threads, etc.). > To cater to this situation i would like to know the option by which > i can close my unused threads as maxSparethreads doesnt exist > anymore in tomcat 6.0 maxIdleTime But honestly, I don't think you'll need it unless the default of 1 minute isn't appropriate for you. - -chris -----BEGIN PGP SIGNATURE----- Version: GnuPG/MacGPG2 v2.0.17 (Darwin) Comment: GPGTools - http://gpgtools.org Comment: Using GnuPG with Mozilla - http://www.enigmail.net/ iEYEARECAAYFAlCYh18ACgkQ9CaO5/Lv0PBOMACeOiDOSJXpcA+0WCg4kABa07X1 6TcAn3OXp/NPUWAbWDRl5O0g0keQ4NWA =WQTx -----END PGP SIGNATURE----- --------------------------------------------------------------------- To unsubscribe, e-mail: users-unsubscr...@tomcat.apache.org For additional commands, e-mail: users-h...@tomcat.apache.org
