Dnia środa, 6 czerwca 2007 22:27, James C. McPherson napisał:
> Jakub Wartak wrote:
> > Today I've hit the same bug,
> >
> > Host: VMware Server 1.0.2 build-39867, running on top of Linux 2.6.17.8
> > Solaris 10U3 works 100% good with LSIlogic emulation (tunned in VMX
> > configuration files). SNV_65 32bit/64bit with LSILogic sillently blows up
> > whole VMware thread responsible for Solaris VM , BusLogic doesn't detect
> > any disks at all. The same happend on another VMware Server with
> > LSILogic, IDE works fine on both.
> >
> > As far as I can remember (today I've lost one of my disks in my
> > RAID0-based lab server which on top VMware was running so I cannot
> > copy&paste the log from VMware;)) this happens just before some
> > unsupported SCSI commands by VMware.
>
> Hi Jakub,
> Absent any smoking gun implicating mpt, I'm really thinking
> that this is a problem with VMWare rather than Solaris.

Yeah, it seems that it is. Reproduced this bug after adding LSIlogic controler 
and SCSI disks to VMware. Part of VMware server log for snv65 VM just after 
issuing "devfsadm -c disk".

Jun 07 11:57:51: vcpu-0| SCSI-DEV0:0: Unsupported command REPORT LUNS issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:0: Unsupported command REPORT LUNS issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:0: Unsupported command REPORT LUNS issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:0): No transfer information for 
*UNKNOWN (0x1b)* (0x1b)
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:0: Unsupported command *UNKNOWN (0x1b)* 
issued. --ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:0: Unsupported command LOG SENSE issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:0): INQUIRY request with EVDP set
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:0): MODE SENSE(6) for unsupported 
page 0x8
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:0): No transfer information for 
PERSISTENT RESERVE IN (0x5e)
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:0: Unsupported command PERSISTENT RESERVE 
IN issued. --ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:1: Unsupported command REPORT LUNS issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:1: Unsupported command REPORT LUNS issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:1: Unsupported command REPORT LUNS issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:1): No transfer information for 
*UNKNOWN (0x1b)* (0x1b)
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:1: Unsupported command *UNKNOWN (0x1b)* 
issued. --ok
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:1: Unsupported command LOG SENSE issued. 
--ok
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:1): INQUIRY request with EVDP set
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:1): MODE SENSE(6) for unsupported 
page 0x8
Jun 07 11:57:51: vcpu-0| SCSI DEVICE (scsi0:1): No transfer information for 
PERSISTENT RESERVE IN (0x5e)
Jun 07 11:57:51: vcpu-0| SCSI-DEV0:1: Unsupported command PERSISTENT RESERVE 
IN issued. --ok
Jun 07 11:57:51: vcpu-0| Caught signal 11 -- tid 6731
Jun 07 11:57:51: vcpu-0| SIGNAL: eip 0x809dadc esp 0xafaf3290 ebp 0xafaf3298
Jun 07 11:57:51: vcpu-0| SIGNAL: eax 0xa0 ebx 0x8747010 ecx 0x85c5e40 edx 0x0 
esi 0x8747010 edi 0x87498d0
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf3290 : 0x00000000 0x000000a0 
0xafaf32b8 0x0824f8e0
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf32a0 : 0x085befd0 0x00000020 
0x000000a0 0x081ab878
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf32b0 : 0xb7b62808 0x00000000 
0xafaf3328 0x0824fcc2
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf32c0 : 0x085c5e40 0x08747010 
0x000028c8 0x0806787b
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf32d0 : 0xb7b5cd08 0xb7b5cd08 
0x01000710 0x08510450
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf32e0 : 0x00000005 0x084ecb58 
0x00000000 0x0805741a
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf32f0 : 0x00000000 0x00000000 
0x006937c0 0x084ee060
Jun 07 11:57:51: vcpu-0| SIGNAL: stack 0xafaf3300 : 0xf2a69574 0x0004324d 
0x01010dfe 0x08747010
un 07 11:57:51: vcpu-0| Backtrace:
Jun 07 11:57:51: vcpu-0| Backtrace[0] 0xafaf2e68 eip 0x805af90
Jun 07 11:57:51: vcpu-0| Backtrace[1] 0xafaf2f38 eip 0x80f354a
Jun 07 11:57:51: vcpu-0| Backtrace[2] 0xafaf2fa8 eip 0x80f32da
Jun 07 11:57:51: vcpu-0| Backtrace[3] 0xafaf3298 eip 0xffffe420
Jun 07 11:57:51: vcpu-0| Backtrace[4] 0xafaf32b8 eip 0x824f8e0
Jun 07 11:57:51: vcpu-0| Backtrace[5] 0xafaf3328 eip 0x824fcc2
Jun 07 11:57:51: vcpu-0| Backtrace[6] 0xafaf3368 eip 0x819d565
Jun 07 11:57:51: vcpu-0| Backtrace[7] 0xafaf3388 eip 0x82bc5ff
Jun 07 11:57:51: vcpu-0| Backtrace[8] 0xafaf33a8 eip 0x82bcbf9
Jun 07 11:57:51: vcpu-0| Backtrace[9] 0xafaf33d8 eip 0x82c55cb
Jun 07 11:57:51: vcpu-0| Backtrace[10] 0xafaf33e8 eip 0x82bcce1
Jun 07 11:57:51: vcpu-0| Backtrace[11] 0xafaf3458 eip 0x8068449
Jun 07 11:57:51: vcpu-0| Backtrace[12] 0xafaf34c8 eip 0xb7f6a0bd
Jun 07 11:57:51: vcpu-0| Backtrace[13] 00000000 eip 0xb7d9792e
Jun 07 11:57:51: vcpu-0| Unexpected signal: 11.
Jun 07 11:57:51: vcpu-0| Backtrace:
Jun 07 11:57:51: vcpu-0| Backtrace[0] 0xafaf2a48 eip 0x805af90
Jun 07 11:57:51: vcpu-0| Backtrace[1] 0xafaf2e68 eip 0x80c02db
Jun 07 11:57:51: vcpu-0| Backtrace[2] 0xafaf2f38 eip 0x80f35ad
Jun 07 11:57:51: vcpu-0| Backtrace[3] 0xafaf2fa8 eip 0x80f32da
Jun 07 11:57:51: vcpu-0| Backtrace[4] 0xafaf3298 eip 0xffffe420
Jun 07 11:57:51: vcpu-0| Backtrace[5] 0xafaf32b8 eip 0x824f8e0
Jun 07 11:57:51: vcpu-0| Backtrace[6] 0xafaf3328 eip 0x824fcc2
Jun 07 11:57:51: vcpu-0| Backtrace[7] 0xafaf3368 eip 0x819d565
Jun 07 11:57:51: vcpu-0| Backtrace[8] 0xafaf3388 eip 0x82bc5ff
Jun 07 11:57:51: vcpu-0| Backtrace[9] 0xafaf33a8 eip 0x82bcbf9
Jun 07 11:57:51: vcpu-0| Backtrace[10] 0xafaf33d8 eip 0x82c55cb
Jun 07 11:57:51: vcpu-0| Backtrace[11] 0xafaf33e8 eip 0x82bcce1
Jun 07 11:57:51: vcpu-0| Backtrace[12] 0xafaf3458 eip 0x8068449
Jun 07 11:57:51: vcpu-0| Backtrace[13] 0xafaf34c8 eip 0xb7f6a0bd
Jun 07 11:57:51: vcpu-0| Backtrace[14] 00000000 eip 0xb7d9792e
Jun 07 11:57:51: vcpu-0| Core dump limit is 51200 kb.
Jun 07 11:57:51: vcpu-0| Attempting to dump core...
[..]

> If you could repeat your testing with the nightly build
> from 28 May 2007 which includes the changes for
>
> 6550849 WARNING: mpt TEST_UNIT_READY failure
> 6556832 After ctlr brought to online, some paths failed to come back
> 6554029 mpt should get maxdevice from portfacts, not IOCfacts
>
> and let me know success/failure, that would be great.

Sure I'll try it, but probbably next week... 

> Have you (or Erast) logged this issue with VMWare?

I haven't yet.

-- 
Jakub Wartak
-vnull
http://vnull.pcnet.com.pl
_______________________________________________
storage-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/storage-discuss

Reply via email to