New answers tagged cluster
Solved. On the first node: /dev/sdb is the quorum device /dev/sdc is the shared volume (the fence device) But on the second node the devices are inverted: /dev/sdc is the quorum device /dev/sdb is the shared volume (the fence device) The first node has placed (rightly) a scsi reservation on its fence device, but for the second node this device name ...
Consider trying multicast or broadcast the files on the lan and receiving it on all clients from the same stream. It will then take the time for ONE copy... See for instance: http://linux.die.net/man/1/udp-sender Good luck, gerhard
I'd install pdsh and configure it to allow me to run commands in parallel on all servers (or selected individual hosts or groups of servers - as defined in pdsh's /etc/genders file). e.g. after installing pdsh, configuring it to use ssh (rather than the odd default of rsh) and defining an all group that contains all servers: pdsh -g all mv /old /new |& ...
The best rgmanager cluster tutorial I've seen is this: AN!Cluster Tutorial 2 Fencing is the act of one cluster node making sure a failed node can't access the shared resources. The STONITH (Shoot The Other Node In The Head) also called power fencing, basically reboots the failed node. It does that either using a remote console (like HP iLO, Dell DRAC, ...
You might want to look at drbd, which provides RAID-1 like disk redundancy and sharing over the network. Drbd is a block device which is designed to build high availability clusters by providing a virtual shared device which keeps disks in nodes synchronised using TCP/IP. This simulates RAID 1 but avoiding the use of uncommon hardware (shared SCSI buses ...
It works fine using the direct download, you can also rename file with -O parameter wget 'http://downloads.sourceforge.net/project/skewer/Binaries/skewer-0.1.127-linux-x86_64?r=&ts=1445976044&use_mirror=netix' -O skewer-0.1.127-linux-x86_64
Top 50 recent answers are included