crsctl stop crs -f 停止不成功 (SYSTEMDG in QUIESCING state)

There exist two workarounds to solve this problem:

1.  kill these processes: ohasd, gipcd, mdnsd, gpnpd, evmd, and crsd

ohasd, gipcd, mdnsd, gpnpd, evmd, & crsd

OR

2.  reboot the node

This leaves the SYSTEMDG diskgroup in QUIESCING state (HTML report from Document 470211.1 ):

 

GROUP_NUMBER NAME STATE TYPE TOTAL_MB FREE_MB OFFLINE_DISKS
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
3 SYSTEMDG QUIESCING NORMAL 894240 892024 0
0 RECO DISMOUNTED 0 0 0
0 DATA DISMOUNTED 0 0 0

 

Note: "The QUIESCING state occurs when the CRSCTL utility attempts to dismount a disk group that contains the Oracle Cluster Repository (OCR). The disk group cannot be dismounted until Cluster Ready Services (CRS) exits, because the disk group contains the OCR."

As workaround, please kill the next background processes (as root user) on each node:

ohasd, gipcd, mdnsd, gpnpd, evmd, & crsd

Example:

ps -fea | grep ohasd.bin | grep -v grep
ps -fea | grep gipcd.bin | grep -v grep
ps -fea | grep mdnsd.bin | grep -v grep
ps -fea | grep gpnpd.bin | grep -v grep
ps -fea | grep evmd.bin | grep -v grep
ps -fea | grep crsd.bin | grep -v grep

 
ohasd, gipcd, mdnsd, gpnpd, evmd, & crsd

Example:
 

[root@node1 ~]# ps -fea | grep ohasd.bin | grep -v grep
root 13783 1 0 Oct25 ? 00:01:41 /bin/ohasd.bin reboot
[root@node1 ~]#
[root@node1 ~]# ps -fea | grep gipcd.bin | grep -v grep
oracle 13938 1 0 Oct25 ? 00:00:01 /bin/gipcd.bin
[root@node1 ~]#
[root@node1 ~]# ps -fea | grep mdnsd.bin | grep -v grep
oracle 13943 1 0 Oct25 ? 00:00:05 /bin/mdnsd.bin
[root@node1 ~]#
[root@node1 ~]# ps -fea | grep gpnpd.bin | grep -v grep
oracle 14178 1 0 Oct25 ? 00:00:03 /bin/gpnpd.bin
[root@node1 ~]#
[root@node1 ~]# ps -fea | grep evmd.bin | grep -v grep
oracle 14481 1 0 Oct25 ? 00:00:02 /bin/evmd.bin
[root@node1 ~]#
[root@node1 ~]# ps -fea | grep crsd.bin | grep -v grep
root 14684 1 0 Oct25 ? 00:05:31 /bin/crsd.bin reboot

[root@dmorldb03 ~]# kill -9 13783 13938 13943 14178 14481 14684