1

I've installed latest yukon driver from marvell for Solaris 10 8/11. Got the driver, installed, updated /etc/driver_aliases and worked fine.

After the system is running for a couple of days, it hungs with the following error:

Mar 24 12:13:33 somehost Msg: Unexpected IRQ Status error
Mar 24 12:13:33 somehost Class: Hardware failure
Mar 24 12:13:34 somehost Msg: Unexpected IRQ Status error
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost Msg: Unexpected IRQ Status error
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost yukonx: [ID 610292 kern.notice] yukonx0: -- ERROR --
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost Class: Hardware failure
Mar 24 12:13:34 somehost yukonx: [ID 285869 kern.notice] ERROR: yukonx0: SkDrvEvent: Adapter failed!
Mar 24 12:13:34 somehost yukonx: [ID 610292 kern.notice] yukonx0: -- ERROR --
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost yukonx: [ID 610292 kern.notice] yukonx0: -- ERROR --
Mar 24 12:13:34 somehost last message repeated 1 time
Mar 24 12:13:34 somehost Msg: Unexpected IRQ Status error
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost Msg: Unexpected IRQ Status error
Mar 24 12:13:34 somehost yukonx: [ID 285869 kern.notice] ERROR: yukonx0: SkDrvEvent: Adapter failed!
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost Msg: Unexpected IRQ Status error
Mar 24 12:13:34 somehost yukonx: [ID 285869 kern.notice] ERROR: yukonx0: SkDrvEvent: Adapter failed!
Mar 24 12:13:34 somehost Class: Hardware failure
Mar 24 12:13:34 somehost Nr: 0x265
Mar 24 12:13:34 somehost yukonx: [ID 610292 kern.notice] yukonx0: -- ERROR --
Mar 24 12:13:34 somehost Class: Hardware failure
Mar 24 12:13:34 somehost Msg: Unexpected IRQ Status error

I've disabled apci, disabled devices which was sharing irqs, with no luck. Seems like under some load the system hungs and the only way to recover it is just rebooting the server.

System is:

Gigabyte: GA-MA770T-UD3. Kingston: 2x4g 1333 ddr. changed memories with no luck.

NIC: dlink DGE-530T PCI.

disks: 4 WD RE3 in raid10, zfs, no problems here.

I have the same NIC running on another Solaris server with no issues the other system has only 2gigs of ram.

Any help will be appreciated. Thanks,

SamB
  • 269
  • 2
  • 13
gustavo
  • 11
  • 1
  • It's a bad excuse for hangs, but the Marvell chipsets are just about complete trash. If you want something reliable get a good NIC chip (eg Intel, Broadcom, QLogic, Emulex, Mellanox, Netxen, etc) – Chris S Mar 27 '12 at 19:28
  • you're totally right, I'm waiting the PCIe intel server NIC but just I was asking if I can beat the beast with a workarround until the NIC hits my country because sadly our customs are pretty picky. But until that I would search for a fix. – gustavo Mar 28 '12 at 23:48

0 Answers0