Nick Webb wrote: > I'll be doing all of these on 64-bit systems, good to know. I just got > a new project today that will start out around 5TB and likely grow to 7 > or 8 soon. Not all these systems have enough ram + swap to get to 8GB > of virtual memory, I wonder how it works without that much, I hope it > doesn't just fail?
xfs_repair runs out of ram and dies. I needed to add heaps of swap on this one system. > Also, you can run xfs_check online, right? That's my impression, so at > least the server isn't down during the check, although it will likely be > crawling. its xfs_repair that takes the ram xfs_check is fast and simple i think. > Yes I've heard this as well from some RedHat/CentOS friends. My intent > would be to make sure all the systems with XFS have a good UPS, and are > setup to shutdown on power loss. Of course, we'll have a good backup > plan as well. That should mitigate things to the point it doesn't > really matter... rdiff-backup rocks for this, with nagios... dave -- ubuntu-server mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-server More info: https://wiki.ubuntu.com/ServerTeam
