vBulletin Search Engine Optimization
| |||||||
| Register | FAQ | Members List | Calendar | Search | Today's Posts | Mark Forums Read |
| ||||
| - My understanding is that many volume managers will mark that block/sector as bad and continue on. What happens under an ASM raw disk if a disk block/sector goes bad? - We currently have a Sun Fire V880 w/ (6) disks and only one hard disk controller. We will be procuring another controller for HA (high availability) but that will probably take at least a few weeks (and maybe months). If I set up ASM with one controller and normal redundancy (i.e., two failure groups), will I be able to later add another controller (with each failure group controlled by one controller)? |
| ||||
| GeoPappas wrote: > - My understanding is that many volume managers will mark that > block/sector as bad and continue on. What happens under an ASM raw > disk if a disk block/sector goes bad? > > - We currently have a Sun Fire V880 w/ (6) disks and only one hard disk > controller. We will be procuring another controller for HA (high > availability) but that will probably take at least a few weeks (and > maybe months). If I set up ASM with one controller and normal > redundancy (i.e., two failure groups), will I be able to later add > another controller (with each failure group controlled by one > controller)? > Question 1: I don't know, that's a good question for Metalink (or, figure out a way to test it yourself, but I don't know offhand how to force a "bad block" at the hardware level). Question 2: ASM doesn't know about controllers per se, in fact ASM can actually work with a cooked file. ASM looks for "paths" that identify whole disks, partitions, logical drives, NAS files, and so on. Adding a controller would probably add one or more ASM "paths" on most platforms. If you start with a normal redundant (or better) ASM configuration, you should be able to add new ASM "disks" as your paths change, and drop old ones if necessary. Ans: Test it. -Mark Bole |