Date: Mon, 1 Aug 2016 09:16:34 -0500 From: Rob Belics <robbelics@gmail.com> To: freebsd-net@freebsd.org Subject: Troubleshooting Idrops Message-ID: <CAPu-kW_kafX1xV4SJoOFnFDEdwb7-FZVV%2BUBDwz%2BjO%2BMGs6X_w@mail.gmail.com>
next in thread | raw e-mail | index | archive | help
I have a VPS server running FreeBSD 10.3-RELEASE-p4 and nginx. It contains three very low volume web sites that have been up for about three years. I was tinkering with TLS and SSL ciphers by eliminating TLSv1 and TLSv1.1 with different ciphers when I noticed my daily "Network interface status" report one morning saying I was getting Idrops of 48665. Network interface status: Name Mtu Network Address Ipkts Ierrs Idrop Opkts Oerrs Coll Drop em0 1500 <Link#1> xx:xx:3c:cd:7e:c7 225569570 0 48665 923463 0 0 0 em0 - xxxx::xxx:3cf fe80::216:3cff:fe 0 - - 4 - - - em0 - 107.xxx.xx.xx mysite1.co 94833 - - 0 - - - em0 - 107.xxx.xx.0 mysite2.co 479981 - - 920067 - - - lo0 16384 <Link#2> 783 0 0 783 0 0 0 lo0 - ::1 ::1 0 - - 0 - - - lo0 - xxxx::1%lo0 xxxx::1%lo0 0 - - 0 - - - lo0 - your-net localhost 783 - - 783 - - - I reverted my TLS/SSL changes but, the next day, that exact same number of Idrops happened and continued for a couple of days afterwards. I just don't know what I could have done to cause this and am looking for troubleshooting help since it's been so long since I've had to deal with this and forgotten nearly everything. All the sites seem to function normally and I should note that, besides the nginx server, there are also two nodejs servers listening via proxy. I do nothing with IPv6. Here is part of vmstat -z where I notice FAILs: ITEM SIZE LIMIT USED FREE REQ FAIL SLEEP UMA Hash: 128, 0, 5, 26, 7, 0, 0 4 Bucket: 16, 0, 8, 496, 21344, 0, 0 6 Bucket: 24, 0, 0, 336, 121, 0, 0 8 Bucket: 32, 0, 2, 376, 1600, 0, 0 12 Bucket: 48, 0, 0, 0, 97831, 0, 0 16 Bucket: 64, 0, 12, 303, 9585, 8, 0 32 Bucket: 128, 0, 14, 389, 46423, 0, 0 64 Bucket: 256, 0, 20, 235, 48362, 0, 0 128 Bucket: 512, 0, 19, 101, 23133, 0, 0 mbuf_packet: 256, 30975, 256, 253,455561904,97330, 0 mbuf: 256, 30975, 2, 254, 2124523, 0, 0 mbuf_cluster: 2048, 4840, 509, 3, 17874,194660, 2 mbuf_jumbo_page: 4096, 2419, 0, 2, 10659, 0, 0 mbuf_jumbo_9k: 9216, 716, 0, 0, 0, 0, 0 mbuf_jumbo_16k: 16384, 403, 0, 0, 0, 0, 0 mbuf_ext_refcnt: 4, 0, 0, 0, 0, 0, 0 256 Bucket: 1024, 0, 27, 45, 56839,5856, 0 vmem btag: 28, 0, 6648, 1848, 68924, 58, 0 I ran netstat -s and can post that here if it's OK for something that big or if someone wants something specific from that. In addition, what can I do to see these drops without having to wait till the next day for the report? I know I can do netstat -i but that contains drops for the current day.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAPu-kW_kafX1xV4SJoOFnFDEdwb7-FZVV%2BUBDwz%2BjO%2BMGs6X_w>