Try adding '-kv' to the boot arguments. I'd also (if you can) disable gdm (leaving a text console). If it locks up again, from the console press F1-a. Hopefully that should drop you into the kernel debugger.
If not, you could also try attaching a null modem cable to ttya (should be COM1), and add the following boot arguments (assuming I remember them correctly): "-kv -B console=ttya". Then at a lockup, sending a break over the serial connection should drop you into the kernel debugger from the console. At that point, you can then type '$<systemdump' to force a crash. Upon reboot, it should create some files under /var/crash/<hostname>. (unix.## and vmcore.##). If not, run (as root), 'savecore'. The resulting dump files can help determine what is happening on the system when it becomes unresponsive. On Fri, May 14, 2010 at 6:27 AM, MichaelHoy <michael....@unn.ac.uk> wrote: > Running SunOS 5.11 snv_134 > > I’m testing OpenSolaris as a potential NAS solution for our Windows users to > provide storage via CIFS. > I’ve installed and then patched OpenSolaris on an HP DL380 G5 using the > cpqary3 drivers. I’ve an LSI SAS3801E connected to a JBOD of 24 disks and > created a RAID-Z zpool using 3*7 raid groups. > The only device driver issue I could find with the install was the HP ILO > card which I couldn’t care less about – I’ve since disabled it. > All firmware is the latest including the LSI. The disks are Seagate 1TB SAS - > ST31000640SS. > > No problems adding to AD. > Idmaps wildcarded to local (solaris) users and smbautohome config works fine. > Using the correct chmod to set the acls and no problems there. > > I’m aware of the issue with the base install hence my need to upgrade. > (http://opensolaris.org/jive/thread.jspa?threadID=118670&tstart=105) > > I’ve been copying a fair amount of data (about 20TB) to this server over the > last two weeks and observing the metrics – especially the 4 port network > aggregation using LACP which is new to me. All seemed to be working well. > > I'm doing snaps and creating/deleting zfs. > > On at least four occasions the server has become completely unresponsive to > anything other than pings – no ssh or console access. I’ve needed to switch > off the server to get it back. > > Nothing obvious (to me at least) is in the logs and there are no cores I can > find. > > I’ve already posted in the cifs discussions and it was suggested I post here. > (http://opensolaris.org/jive/thread.jspa?threadID=128634&tstart=0) > > Some advice would be appreciated. > Can anyone help? > > Thanks in advance. Michael > -- > This message posted from opensolaris.org > _______________________________________________ > storage-discuss mailing list > storage-discuss@opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/storage-discuss > _______________________________________________ storage-discuss mailing list storage-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/storage-discuss
