Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
I have serial, eth0 and eth1 connections between the two and simulate test on various situations below. Anyone got the same results on testcase#6 like mine? Need help on resolution! Testcase#1: Reboot primary Server normally Tasks: Reboot primary while copying through NFS server Results: Switch over from PriSrv to SecSrv nicely and can unzip the tar.gz file transferred Testcase#2: Reboot Secondary Server (but currently has St:Primary/Secondary) normally Tasks: from testcast#1, secondary server is at Primary state, so now, we reboot Secondary server during nfs file transfer. Results: Switch over from SecSrv to PriSrv nicely and can unzip the tar.gz file transferred Testcase#3: Reboot Secondary Server (but currently has St:Secondary/Primary) normally Tasks: Reboot/shutdown secondary server while primary server is active Results: Primary server still function ok Testcase#4: Unplug Secondary Server (but currently has St:Secondary/Primary) Tasks: Unplug secondary server to simulate secondary server crashes Results: ok, primary server survived Testcase#5: Remove ethernet cable from Primary Server Tasks: Remove eth0 from primary server while nfs file transferring Results: failover completely and can untar the gz file Testcase#6: CRITICAL: unplug primary server Tasks: Unplug primary server power cable to simulate power outage/crash. Results: FAILED COMPLETELY, FILESYSTEM CORRUPTED AND CANNOT BE RECOVERED. Have to mkfs on drbd :-( -- View this message in context: http://www.nabble.com/filesystem-corrupt-after-unplug-AC-cable-from-primary-server-tf4785079.html#a13689246 Sent from the DRBD - User mailing list archive at Nabble.com.