OpenSolaris

Printable Version Enter a New Search
Bug ID 6740850
Synopsis dlmgmt_post_sysevent failed when try to plumb 4000 vlan interfaces
State 10-Fix Delivered (Fix available in build)
Category:Subcategory kernel:gld
Keywords solx86-nic-nicdrv-e1000g
Responsible Engineer Cathy Zhou
Reported Against snv_90 , snv_98
Duplicate Of
Introduced In solaris_nevada
Commit to Fix snv_100
Fixed In snv_100
Release Fixed solaris_nevada(snv_100)
Related Bugs 6713612 , 6729044
Submit Date 25-August-2008
Last Update Date 8-October-2008
Description
The test case test08 always failed with TIMED_OUT. The end of the journal is:

stdout|
stdout| xplumb_subnet_remote: plumb remote vlan begins!
stdout| xplumb_subnet_remote: plumb remote interface done!
stdout| lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232 index 1
stdout|         inet 127.0.0.1 netmask ff000000
stdout| e1000g0: flags=201004843<UP,BROADCAST,RUNNING,MULTICAST,DHCP,IPv4,CoS> mtu 1500 index 2
stdout|         inet 10.13.28.162 netmask ffffff00 broadcast 10.13.28.255
stdout|         ether 0:14:4f:d1:c5:b8
stdout| e1000g1: flags=201000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4,CoS> mtu 1500 index 3
stdout|         inet 11.0.1.162 netmask ffffff00 broadcast 11.0.1.255
stdout|         ether 0:14:4f:d1:c5:b9
stdout| lo0: flags=2002000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv6,VIRTUAL> mtu 8252 index 1
stdout|         inet6 ::1/128
stdout| clean up finished.

This problem was observed on T2000, X4150 and X2200 M2 when testing the e1000g driver and the bnx driver. The systems were bfu'ed to the latest bits (e.g. daily.0820-nd) from
snv_96.
1) I've reproduced it on X2200M2 with nge and bnx of the bfu from /ws/onv.eng/export/home/nightly/snv_97

2) During case test08, we got two error messages in system log:
dlmgmtd[100017]: [ID 498274 daemon.warning] dlmgmt_post_sysevent(2625) failed: I/O error
Sep  3 19:27:52 airvoice devfsadmd[100164]: [ID 639486 daemon.error] failed to log event: I/O error

3) And after test08 timeout, system hang if we invoke "ifconfig -a" command.

ifconfig -a
lo0: flags=2001000849<UP,LOOPBACK,RUNNING,MULTICAST,IPv4,VIRTUAL> mtu 8232 index 1
        inet 127.0.0.1 netmask ff000000 
nge0: flags=201000843<UP,BROADCAST,RUNNING,MULTICAST,IPv4,CoS> mtu 1500 index 2
        inet 10.13.30.128 netmask ffffff00 broadcast 10.13.30.255
(...no response from here)
The test08 is a vlan test. After some investigation, we found the timeout is caused by ifconfig cmd hung when trying to plumb vlan interface. Please see the comments.

Hope more suggestions from the kernel team.
Thanks
Work Around
N/A
Comments
N/A