Revision 90382ed9afeafd42ef193f0eadc6b2a252d6c24d authored by Goldwyn Rodrigues on 24 June 2015, 14:30:32 UTC, committed by NeilBrown on 24 July 2015, 03:37:59 UTC
During a node failure, We need to suspend read balancing so that the reads are directed to the first device and stale data is not read. Suspending writes is not required because these would be recorded and synced eventually. A new flag MD_CLUSTER_SUSPEND_READ_BALANCING is set in recover_prep(). area_resyncing() will respond true for the entire devices if this flag is set and the request type is READ. The flag is cleared in recover_done(). Signed-off-by: Goldwyn Rodrigues <rgoldwyn@suse.com> Reported-By: David Teigland <teigland@redhat.com> Signed-off-by: NeilBrown <neilb@suse.com>
1 parent 33e38ac
| File | Mode | Size |
|---|---|---|
| ext | ||
| Kconfig | -rw-r--r-- | 697 bytes |
| Makefile | -rw-r--r-- | 372 bytes |
| array.c | -rw-r--r-- | 1.1 KB |
| hda_bus_type.c | -rw-r--r-- | 1.8 KB |
| hdac_bus.c | -rw-r--r-- | 5.0 KB |
| hdac_controller.c | -rw-r--r-- | 12.8 KB |
| hdac_device.c | -rw-r--r-- | 22.3 KB |
| hdac_i915.c | -rw-r--r-- | 4.7 KB |
| hdac_regmap.c | -rw-r--r-- | 12.0 KB |
| hdac_stream.c | -rw-r--r-- | 18.8 KB |
| hdac_sysfs.c | -rw-r--r-- | 9.6 KB |
| local.h | -rw-r--r-- | 618 bytes |
| trace.c | -rw-r--r-- | 107 bytes |
| trace.h | -rw-r--r-- | 2.1 KB |
Computing file changes ...
