On Wed, Jul 11, 2007 at 10:20:08AM +0400, Pavel D. Kuzin wrote: > node1:/usr/local/apache/bin# clusvcadm -d hosting > Member node1 disabling hosting...success > node1:/usr/local/apache/bin# rg_test test /etc/cluster/cluster.conf start > service hosting > Running in test mode. > Starting hosting... > /usr/share/cluster/clusterfs.sh: line 729: logAndPrint: command not found > mount: /dev/mapper/Hosting-Hosting_data already mounted or /usr/hosting busy > <err> 'mount -t gfs /dev/mapper/Hosting-Hosting_data /usr/hosting' > failed, error=32 > Failed to start hosting Ok, so it actually *failed* to mount. Now, if it was already mounted, that's a bug - clusterfs.sh shouldn't try to mount the device a second time. According to your output below, it *was not* mounted. Is there anything in dmesg or /var/log/messages which you think might help? > +++ Memory table dump +++ > 0xb74b01f8 (16 bytes) allocation trace: > 0xb74adb4c (16 bytes) allocation trace: > 0xb74a5bb8 (16 bytes) allocation trace: > 0xb74ace34 (16 bytes) allocation trace: > 0xb749b270 (16 bytes) allocation trace: > 0xb74ad0b8 (16 bytes) allocation trace: > 0xb74a5d64 (16 bytes) allocation trace: > 0xb74a561c (16 bytes) allocation trace: > 0xb74b382c (776 bytes) allocation trace: > --- End Memory table dump --- Hrm, that's not good, but ultimately not the cause of the problem. -- Lon Hohberger - Software Engineer - Red Hat, Inc. -- Linux-cluster mailing list Linux-cluster@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/linux-cluster
