Carsten,
Looks like:
http://bugs.opensolaris.org/view_bug.do?bug_id=6807184
Bug ID 6807184
Synopsis rge driver drops off network
State 3-Accepted (Yes, that is a problem)
Category:Subcategory driver:rge
Keywords opensolaris
Reported Against snv_106
Duplicate Of
Introduced In
Commit to Fix
Fixed In
Release Fixed
Related Bugs
Submit Date 18-February-2009
Last Update Date 24-February-2009
Any other pci network card that you can try to verify the issue?
More things I noticed glancing over your data below (may or may not be
unrelated)
..Remco
I am really no expert on this but I'll give you my 2 cents on what I saw
glancing over the data you sent. It appears to me there is something
funny going on the power management on your system:
http://opensolaris.org/os/project/tesla/Work/CPUPM/
I think it should work...however:
[snip]
Aug 11 16:46:19 sun rootnex: [ID 263444 kern.warning] WARNING: dmar0,1
generated a fault event when translating DMA write
Aug 11 16:46:19 sun on address 0xbe000000 for PCI(0, 2, 0), the reason is:
Aug 11 16:46:19 sun The Write field in a page-table entry is Clear
when DMA write
Aug 11 16:46:19 sun rootnex: [ID 263444 kern.warning] WARNING: dmar0,2
generated a fault event when translating DMA read
Aug 11 16:46:19 sun on address 0xf000e000 for PCI(0, 29, 7), the
reason is:
Aug 11 16:46:19 sun The Read field in a page-table entry is Clear when
DMA read
Aug 11 16:46:19 sun rootnex: [ID 630785 kern.info] Start to enable the
dmar units
Aug 11 16:46:19 sun rootnex: [ID 306289 kern.info] dmar0,0 enabled
Aug 11 16:46:19 sun rootnex: [ID 306289 kern.info] dmar0,1 enabled
Aug 11 16:46:19 sun rootnex: [ID 306289 kern.info] dmar0,2 enabled
Aug 11 16:46:19 sun pcplusmp: [ID 419660 kern.info] pcplusmp: irq 0x9
vector 0x84 ioapic 0x4 intin 0x9 is bound to cpu 1
Aug 11 16:46:19 sun amd_iommu: [ID 251261 kern.info] NOTICE: amd_iommu:
No AMD IOMMU ACPI IVRS table
Aug 11 16:46:19 sun rootnex: [ID 263444 kern.warning] WARNING: dmar0,1
generated a fault event when translating DMA write
Aug 11 16:46:19 sun on address 0xbe000000 for PCI(0, 2, 0), the reason is:
Aug 11 16:46:19 sun The Write field in a page-table entry is Clear
when DMA write
[snip]
[snip]
Aug 11 16:46:21 sun genunix: [ID 308332 kern.info] root on
rpool/ROOT/opensolaris-3 fstype zfs
Aug 11 16:46:21 sun unix: [ID 190185 kern.info] SMBIOS v2.5 loaded (1101
bytes)
Aug 11 16:46:21 sun unix: [ID 517869 kern.info] NOTICE: cpu_acpi: _PSS
package not found.
Aug 11 16:46:21 sun unix: [ID 388705 kern.warning] WARNING: cpu_acpi:
error parsing _PSS for CPU 0
Aug 11 16:46:21 sun unix: [ID 196753 kern.warning] WARNING: cpupm_init:
processor 0: unable to initialize P-state support
Aug 11 16:46:21 sun unix: [ID 374799 kern.info] NOTICE: cpu_acpi: _CST
evaluate failure
Aug 11 16:46:21 sun unix: [ID 518192 kern.warning] WARNING: cpu_acpi:
error parsing _CST for CPU 0
Aug 11 16:46:21 sun unix: [ID 726263 kern.info] NOTICE: cpu_idle_init:
Failed to cache ACPI C-state data
Aug 11 16:46:21 sun unix: [ID 763091 kern.warning] WARNING: cpupm_init:
processor 0: unable to initialize C-state support
[snip]
Is it possible to disable power management on the CPU, just to see if it
makes any difference?
> What I've noticed is, that the onboard graphic isn't fully supported
by OpenSolaris' X11, but I don't care, because I "scvadm disable
system/gdm" and do everything through the text consoleor via SSH.
Hopefully a "loose" hanging device isn't going to cause issues....
Carsten Wegner wrote:
"fmdump -ev" gives:
TIME CLASS ENA
Jul 28 00:09:11.1307 ereport.io.scsi.cmd.disk.tran 0x0cf585deb5e00001
Jul 28 17:21:35.6707 ereport.io.scsi.cmd.disk.tran 0x050e646c32000001
Alll other hardware related messages should be in the log.
My mainboard is a Jetway JNF 93 LF with INTEL GM45 (Cantiga) + ICH9M chipset
and two Gb LAN Ports. An Intel Penryn Mobile Core 2 Duo CPU is powering
OpenSolaris. The board holds 4 GB of RAM and there are 4 x 1TB WD HDDs attached
to the SATA ports on the board. The bootdisk is a 8 GB SSD. The I/O expierience
is naturally quiet good with such a powerfull setup - copying 700 MB from
RAMDisk to the ZFS raidz1 on the SATA HDDs is done in two seconds.
What I've noticed is, that the onboard graphic isn't fully supported by OpenSolaris' X11,
but I don't care, because I "scvadm disable system/gdm" and do everything
through the text consoleor via SSH.
If it is not CIFS, which would be the best group to post to?
And what about all these lines in the log:
Aug 11 20:07:47 sun smbd[454]: [ID 775558 daemon.debug] smb_door_srv_func:
execute server routine(opcode=2)
Aug 11 20:07:47 sun smbd[454]: [ID 775558 daemon.debug] smb_door_srv_func:
execute server routine(opcode=0)
Aug 11 20:07:47 sun smbsrv: [ID 138215 kern.notice] NOTICE: smbd[NT
Authority\Anonymous]: daten access denied: IPC only
repeating for thousand times?
And:
Aug 11 21:13:31 sun sshd[539]: [ID 685508 auth.info] libgss
dlopen(/usr/lib/gss/mech_spnego.so.1): ld.so.1: sshd: fatal:
/usr/lib/gss/mech_spnego.so.1: open failed: No such file or directory
I am glad that out there are some people who understand their business and are
offering help!
Kind regards
Carsten
_______________________________________________
cifs-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/cifs-discuss
