求助！fmd服务重复coredump无法启动

Question

是台T2000，就是fmd不停的core，挂掉然后自动重启，再core，直到写满/var目录
fmdump看fltlog和errlog都没有错误
查看依赖服务
# svcs -d svc:/system/fmd:default
STATE          STIME    FMRI
online         16:39:08 svc:/system/filesystem/minimal:default
online         16:39:09 svc:/system/sysevent:default
online         16:39:18 svc:/network/rpc/bind:default
online         16:39:20 svc:/system/dumpadm:default
查看服务属性
# svcprop  svc:/system/fmd:default
general/enabled boolean true
general/entity_stability astring Evolving
general/single_instance boolean true
SUNWfmd/entities fmri file://localhost/usr/lib/fm/fmd/fmd
SUNWfmd/grouping astring require_all
SUNWfmd/restart_on astring none
SUNWfmd/type astring path
startup_req/entities fmri svc:/system/sysevent svc:/system/filesystem/minimal svc:/system/dumpadm
startup_req/grouping astring require_all
startup_req/restart_on astring none
startup_req/type astring service
startup_opt/entities fmri svc:/network/rpc/bind
startup_opt/grouping astring optional_all
startup_opt/restart_on astring none
startup_opt/type astring service
start/exec astring /usr/lib/fm/fmd/fmd
start/timeout_seconds count 0
start/type astring method
stop/exec astring :kill
stop/timeout_seconds count 60
stop/type astring method
tm_common_name/C ustring Solaris Fault Manager
tm_man_fmd/manpath astring /usr/share/man
tm_man_fmd/section astring 1M
tm_man_fmd/title astring fmd
restarter/logfile astring /var/svc/log/system-fmd:default.log
restarter/contract count 129
restarter/start_pid count 1224
restarter/start_method_timestamp time 1272274874.040004000
restarter/start_method_waitstatus integer 0
restarter/auxiliary_state astring none
restarter/next_state astring none
restarter/state astring online
restarter/state_timestamp time 1272274874.056201000
查看服务诊断
# svcs -xv
svc:/system/fmd:default (Solaris Fault Manager)
State: offline since Mon Apr 26 17:44:28 2010
Reason: Start method is running.
   See: http://sun.com/msg/SMF-8000-C4
   See: man -M /usr/share/man -s 1M fmd
   See: /var/svc/log/system-fmd:default.log
Impact: This service is not running.
查看/var/svc/log/system-fmd:default.log
。。。
[ Apr 26 17:44:05 Stopping because process dumped core. ]
[ Apr 26 17:44:05 Executing stop method (:kill) ]
[ Apr 26 17:44:05 Executing start method ("/usr/lib/fm/fmd/fmd") ]
[ Apr 26 17:44:18 Method "start" exited with status 0 ]
[ Apr 26 17:44:28 Stopping because process dumped core. ]
[ Apr 26 17:44:28 Executing stop method (:kill) ]
[ Apr 26 17:44:28 Executing start method ("/usr/lib/fm/fmd/fmd") ]
[ Apr 26 17:44:41 Method "start" exited with status 0 ]
[ Apr 26 17:44:50 Stopping because process dumped core. ]
[ Apr 26 17:44:51 Executing stop method (:kill) ]
[ Apr 26 17:44:51 Executing start method ("/usr/lib/fm/fmd/fmd") ]
[ Apr 26 17:45:04 Method "start" exited with status 0 ]
[ Apr 26 17:45:13 Stopping because process dumped core. ]
[ Apr 26 17:45:13 Executing stop method (:kill) ]
[ Apr 26 17:45:13 Executing start method ("/usr/lib/fm/fmd/fmd") ]
[ Apr 26 17:45:27 Method "start" exited with status 0 ]
[ Apr 26 17:45:36 Stopping because process dumped core. ]
[ Apr 26 17:45:37 Executing stop method (:kill) ]
[ Apr 26 17:45:37 Executing start method ("/usr/lib/fm/fmd/fmd") ]
[ Apr 26 17:45:50 Method "start" exited with status 0 ]

Answer 1

关注一下

Answer 2

主要是一个级别高点的合同号，如白金或者金级合同的合同号，就可以去查询了。
另外你的kernel补丁到了多少？如果不是最新的，可以的话，建议可以打下补丁到最新去试试看看。

http://unix.derkeiler.com/Mailin ... 09-01/msg00092.html
这个案例和你的比较相似，他的内核补丁就很低，好像是打补丁好了，问的是可否删除core文件。
或者有维保的话，找下SUN吧。

Answer 3

回复 4# zhmzhouming

    谢谢你的回复，我没有做类似LDOMS的操作，因为我们的工作不涉及这个，我也去sunsolve找了一下，完全只能看个title。是不是要注册帐号并且填一个合同号才有权限啊。
另外我看了下/var/svc/log/system-fmd:default.log前面的纪录，发现从大概20号开始这个服务每4分钟左右重启一次但是没有core，直到22号才开始core了,偶尔还报这个：
fmd: ABORT: failed to populate nvlist: Invalid argument
我知识有限在网上google了相关的东西看不懂。。。

Answer 4

有没有划LDOMS?  sunsolve上有几个fmd dump的列子，可惜我账号权限不够高，无法进去看。我能看到的几个都是因为划了LDOMS导致的fma core dump

进sunsolve 搜索fmd dump，看能不能麻烦有权限的兄弟贴下。如下这个我就无法进去看了：
cpumem-diagnosis module can cause fmd to dump core bug 6477004.

Answer 5

自检

Answer 6

我很郁闷啊，什么错都没报，就是一个劲的core，咋办？
试过停掉服务删除了/var/fm/fmd/下面的全部再重启，还是不行。
要贴core文件看看吗