I've got a problem on the production server after upgrade from 11.1-p2 to 11.2-p13 Sep 4 18:10:11 master kernel: ix0: link state changed to DOWN Sep 4 18:34:03 master kernel: ix0: link state changed to UP Sep 4 18:34:07 master kernel: ix0: link state changed to DOWN Sep 4 18:34:31 master kernel: ix0: link state changed to UP Sep 4 19:11:49 master kernel: ix0: link state changed to DOWN about if Sep 4 19:32:59 master kernel: ix0: <Intel(R) PRO/10GbE PCI-Express Network Driver, Version - 3.2.12-k> mem 0xfbc00000-0xfbdfffff,0xfbe04000-0xfbe07fff irq 11 at device 0.0 on pci4 Sep 4 19:32:59 master kernel: ix0: Using MSI-X interrupts with 9 vectors Sep 4 19:32:59 master kernel: ix0: Ethernet address: ac:1f:6b:6a:4c:ce Sep 4 19:32:59 master kernel: ix0: netmap queues/slots: TX 8/2048, RX 8/2048
Hello, what NIC are you using? And with what type of connection (RJ-45, SFP, Twinax)? Peter
Type of connection RJ-45 and everything is alright. The server works in ovh. The problem appeared after the upgrade to 11.2-p2 ix0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> metric 0 mtu 1500 options=e407bb<RXCSUM,TXCSUM,VLAN_MTU,VLAN_HWTAGGING,JUMBO_MTU,VLAN_HWCSUM,TSO4,TSO6,LRO,VLAN_HWTSO,RXCSUM_IPV6,TXCSUM_IPV6> ether ac:1f:6b:6a:xx:xx hwaddr ac:1f:6b:6a:xx:xx inet 51.68.XXX.XXX netmask 0xffffff00 broadcast 51.68.XXX.XXX inet 54.38.XXX.XXX netmask 0xffffffff broadcast 54.38.XXX.XXX inet 54.38.XXX.XXX netmask 0xffffffff broadcast 54.38.XXX.XXX inet 54.38.XXX.XXX netmask 0xffffffff broadcast 54.38.XXX.XXX inet 54.38.XXX.XXX netmask 0xffffffff broadcast 54.38.XXX.XXX inet 54.38.XXX.XXX netmask 0xffffffff broadcast 54.38.XXX.XXX nd6 options=29<PERFORMNUD,IFDISABLED,AUTO_LINKLOCAL> media: Ethernet autoselect (1000baseT <full-duplex>) status: active ix0: <Intel(R) PRO/10GbE PCI-Express Network Driver, Version - 3.2.12-k> mem 0xfbc00000-0xfbdfffff,0xfbe04000-0xfbe07fff irq 11 at device 0.0 on pci4 ix0: Using MSI-X interrupts with 9 vectors ix0: Ethernet address: ac:1f:6b:6a:xx:xx ix0: netmap queues/slots: TX 8/2048, RX 8/2048 ix0@pci0:3:0:0: class=0x020000 card=0x15ad15d9 chip=0x15ad8086 rev=0x00 hdr=0x00 vendor = 'Intel Corporation' device = 'Ethernet Connection X552/X557-AT 10GBASE-T' class = network subclass = ethernet THX
I updated NIC driver from 3.2.12-k to 3.2.18 and set mtu to 9000 and hw.intr_storm_threshold=0 I will report what will happen next.
Do you use DHCP client to get IP address for this interface? If so, take a look at https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=229432 for a workaround and solution.
(In reply to Eugene Grosbein from comment #4) I use static addresses only
(In reply to Compri from comment #5) You didn't report on what happened after updating. Did anything change?
(In reply to Eric Joyner from comment #6) Now I use FreeBSD 11.2-RELEASE-p4 and 3.2.18 ix0 driver from official website. I don't see any problems. I think is everything is OK.
(In reply to Compri from comment #7) Worth noting that Intel also provides its upstream ix driver [1] via ports/packages. See https://www.freshports.org/net/intel-ix-kmod/ (currently 3.3.6). [1] And em, ixl drivers via intel-em-kmod / intel-ixl-kmod
This seems to be fixed. Please report back if I misunderstood.