We've noticed on one of our test machines, with 5 NICs (nge0 and 4xhme) that when we disable nwam, it cannot be started again.
[We did experience issues with the nge driver, which didn't like being plumbed/unplumbed, so I uninstalled the driver so it never got used again - it reported a DMA error in the cases where it couldn't be plumbed again.] When we do a svcs -xv, we are told that nwamd is offline, but there is also the message that the startup script is being run - which seems contradictory. This also seemed to happen once when switching between ncps, in this case I believe that nwam crashed. Is there any changes in the NWAM gate that might effect how SMF is working? The only solution we found to the issue was to reboot. No matter what we did we couldn't get the NWAM service to start again. Looking in the service log, the last message was that nwamd was shutting down. After that, it didn't seem to even acknowledge the svcadm enable command - which is why I believe it's an issue in SMF itself. Attached is a log of nwamd where it went down switching ncps and also caused this issue. Thanks, Darren. -------------- next part -------------- An embedded message was scrubbed... From: [email protected] Subject: nwamd.log-SMF Date: Thu, 06 Aug 2009 13:36:41 +0100 (IST) Size: 203191 URL: <http://mail.opensolaris.org/pipermail/nwam-dev/attachments/20090806/1891720a/attachment.nws>
