PROBLEM: (95291) (PATCH ID: TCR540-030) ******** After applying a recent dsfmgr patch, cfsd may begin core dumping shortly after startup (if it is enabled at startup) or shortly after enabling it. The core dumps appear in the cluster root as "/core.cfsd..#". Syslog daemon.log (/var/adm/syslog.dated/*/daemon.log) messages similar to the following are typically seen: cfsd[525532]: startup cfsd[525532]: reconfiguring cfsd[525532]: cfsd is active cfsd[525532]: AdvFS file domain cluster_root# has ... cfsd[525532]: AdvFS file domain root1_domain# has ... cfsd[525532]: AdvFS file domain root2_domain# has ... cfsd[525532]: AdvFS file domain root3_domain# has ... cfsd[525532]: AdvFS file domain cluster_usr# has ... cfsd[525532]: AdvFS file domain cluster_var# has ... cfsd[525532]: received signal: Segmentation fault (11) cfsd[525532]: aborting PROBLEM: (95297) (PATCH ID: TCR540-044) ******** This patch corrects a problem in which cfsd will terminate prematurely and core dump when a node leaves the cluster very shortly after joining the cluster. In very rare cases, a node that just joined the cluster may leave the cluster before cfsd has had a chance to finish processing the join event. If this occurs, cfsd may terminate prematurely and dump core. Log messages similar to the following in /var/adm/syslog.dated/*/daemon.log are symptomatic of this problem: ... cfsd[698059]: fatal error in ../../../../../src/usr/sbin/cfsd/cfsd_device.c at line 328: cfsd_device_node_up(): cfsd_cludrd_checkpath(): Invalid argument (22)