Note: "permalinks" may not be as permanent as we would like,
direct links of old sources may well be a few messages off.
Am 31.05.2012 um 14:23 schrieb agag: > My first idea is to have in example WWW server and when one crashes, > second > is working without any additional admin tasks. You can do this with a drbd-device as primary/secondary, mounted e.g. as /ha, and having your webspace, logs, configs etc. somewhere under / ha. Then configure pacemaker with IP-failover etc. and it works as long as one of the cluster-nodes is alive. I had such a configuration since ~2004 with heartbeat and drbd-0.7. Disadvantage is, taht you need to take care of all dependencies like cron-jobs, webstats, users etc. and maybe need to modify some scripts. > > Second idea is to have almost identical operating systems. For > example to > use drbd replicated data for /etc and /bin (wich will be modified > only in > primary/first serverd when changes or updates are done). Is it > possible > somehow? You can use it for the whole / (root-filesystem), if you start an virtual machine on a drbd-device. That's most convenient as everything is replicated. It works in primary/secondary, with a downtime of shutdown+bootup. Helmut Wollmersdorfer