Date: Sat, 18 Dec 2010 01:14:37 +0600 From: Eugene Grosbein <eugen@grosbein.pp.ru> To: net@freebsd.org Subject: netisr queue overflows Message-ID: <4D0BB69D.4000106@grosbein.pp.ru>
next in thread | raw e-mail | index | archive | help
Hi! I run pretty busy 4-core router that has 4 gigabit ethernet interfaces: em0 and em1 grouped to lagg0 that has IP address and igb0 and igb1 grouped to lagg1 that has no IP address but many vlans that carry PPPoE traffic. sysctl net.inet.ip.intr_queue_drops shows me 15408340 packet drops due to netisr queue overflow. I have in /boot/loader.conf: net.isr.maxthreads=4 # default value #net.isr.maxqlimit=10240 # for rtsock net.route.netisr_maxqlen=4096 # for other protocols (IP & PPPoE?) net.isr.defaultqlimit=4096 How do I obtain current queue lengths for monitoring/tuning? I see over 125Kpss for lagg0 and 132Kpps for lagg1. /etc/sysctl.conf: net.isr.direct=0 net.isr.direct_force=0 net.inet.ip.intr_queue_maxlen=4096 dev.em.0.rx_int_delay=200 dev.em.0.tx_int_delay=200 dev.em.0.rx_abs_int_delay=4000 dev.em.0.tx_abs_int_delay=4000 dev.em.0.rx_processing_limit=4096 dev.em.1.rx_int_delay=200 dev.em.1.tx_int_delay=200 dev.em.1.rx_abs_int_delay=4000 dev.em.1.tx_abs_int_delay=4000 dev.em.1.rx_processing_limit=4096 dev.igb.0.rx_processing_limit=4096 dev.igb.1.rx_processing_limit=4096 kern.ipc.nmbclusters=100000 kern.ipc.nmbjumbop=100000 kern.ipc.maxsockbuf=83886080 net.graph.maxdgram=8388608 net.graph.recvspace=8388608 Eugene Grosbein
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4D0BB69D.4000106>