Hi Guys.
I started to play a bit with the HA... I'm in stage of some testing and
projecting the setup and have several questions. Any notes to below
mentioned are more than welcome.
Current setup:
- HW: 5 servers + 2 new ones used now for testing
- OS: Debian woody, sarge and etch
- Location: data center with redundant 100Mbit connection
- Services: apache, ftp, qmail, ldap (, ssh)
Target:
- have no single point of failure after providers ethernet connectons
- 2 LVS directors with VIPs using heartbeat
- redundant data storage (2 linux servers with identical disk arrays)
- worker nodes (real servers) mounting the arrays using clustering
filesystem
On the storage servers I want to use DRBD primary/primary and export the
devices for real nodes... to make it HA, I want to export the devices
thru virtual IP with takeover between this two storage servers... The
real servers will the import the devices thru this VIP and run GFS2 on it...
The question is, if using the VIP with takeover for exporting devices
for GFS is safe... I consider using GRBD or iSCSI... is any of thouse
better for the setup? I tend to use iSCSI but have no good reasons for
it - just the feelings...
Can anyone advice?
Thank you,
--
Kamil Srot
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
