Date: Wed, 30 Aug 2023 09:39:45 +0000 From: bugzilla-noreply@freebsd.org To: virtualization@FreeBSD.org Subject: [Bug 263062] tcp_inpcb leaking in VM environment Message-ID: <bug-263062-27103-h80fx8GM7A@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-263062-27103@https.bugs.freebsd.org/bugzilla/> References: <bug-263062-27103@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D263062 paul vixie <paul@redbarn.org> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |paul@redbarn.org --- Comment #14 from paul vixie <paul@redbarn.org> --- > ... > For now the server already has stopped working three times with: > Aug 10 07:25:38 mailrelay kernel: [zone: udp_inpcb] kern.ipc.maxsockets l= imit > reached >=20 > I guess the DNS queries are the reason for this. > ... this is normal for all recursive dns servers not just unbound, because kami= nsky resistance requires randomized source udp port numbers for cache miss transactions, which in turn requires a moderately large working set of bound udp sockets, with heavy churn. if this turns out to be chipset dependent, we should be looking at atomic t= est and set operations. i long ago turned off hyperthreading because microarchtectural details turned out to be motherboard dependent. there may= be similar problems still lurking, but on a lark, maybe try turning off hyperthreading? --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-263062-27103-h80fx8GM7A>