This patch series bases on previously sent patch series: array/container freezing by Dan Williams, and my and Maciej Trela's patches for OLCE and check pointing for external metadata.
This patch series implements:
1. Patches 0001 to 0018 : fixes for freezing, OLCE and check pointing
2. Patches 0018 to 0021 : implements new migration features for external metadata:
- Migration raid0 to raid5
- Migration raid5 to raid0
- Chunk size migration
We are planning in the few days add fixes for raid10 <-> raid0 takeover also.
Migration feature reuses code flow introduced for OLCE (Online Capacity Expansion) and uses the same grow/reshape flow in mdadm/mdmon.
Migration is executed in the following way:
1. mdadm: reshape_super() prepares metadata update and sends it to mdmon
2. mdadm: waits for reshape array state
3. monitor: receives metadata update and applies it.
4. monitor: metadata update triggers managemon.
5. managemon: updates array (md) configuration and starts reshape
6. mdadm: finds that reshape is started and continues it using check pointing
7. mdadm: reshape is finished and manage_reshape() finalizes array:
- Sets array size as is given in metadata
- Performs takeover to raid0 if necessary
In current patches placement of manage_reshape() function call was changed.
It is moved to end of array processing to use common code form Grow.c for external metadata reshape case
(we do not need to duplicate existing code) as it would do the same things as code for native metadata.
New manage_reshape() placement causes a few things to do in implementation only and simplifees code.
Migrations command line:
1. Execute migration raid0->raid5:
mdadm - -grow /dev/md/array_name -level 5 -layout=left-asymmetric
This converts n-disks raid0 array to (n+1)-disks raid5 array. Additional raid5 array disk is user from spares pool
2. Execute migration raid5->raid0:
mdadm - -grow /dev/md/array_name -level 0
This converts n-disks raid5 array to n-disks raid0 array.
3. Execute chunk size migration
mdadm - -grow /dev/md/array_name -chunk N
where N is ne chunk size value
In current patch series testing of 'experimental' flag was changes according to Neil's suggestion, as I've found it useable for my testing purposes.
The rest of comments to already reviewed patches (OLCE/check pointing) will be implemented in next step.
---
Adam Kwolek (21):
Migration: Chunk size migration
Migration raid0->raid5
Migration: raid5->raid0
imsm: Use geo structure
Change manage_reshape() placement
FIX: Start mdmon after changing level
FIX: Checkpointing: Enable checkpointing for non-growing migrations
FIX: Level field can be NULL
imsm Fix: Core during rebuild on array details read
WORKAROUND: md reports idle state during reshape start
FIX: Honor !reshape state on wait_reshape() entry
FIX: Unfreeze not only container for external metadata
FIX: Cannot exit monitor after takeover
imsm: FIX: Continue expansion after -As command
imsm: FIX: Reshape Cancel message on 2nd array doesn't work correctly
imsm: FIX: Guard size setting
FIX: Allow for reshape without backup file
FIX: Use MDADM_EXPERIMENTAL environment variable
imsm: FIX: imsm_check_reshape_conditions change ret_val
imsm: Change some debug output information
imsm: Remove unnecessary code
mdadm/mdadm/Detail.c | 1
mdadm/mdadm/Grow.c | 178 ++++++++----
mdadm/mdadm/Makefile | 6
mdadm/mdadm/managemon.c | 113 ++++++-
mdadm/mdadm/mdadm.h | 7
mdadm/mdadm/mdmon.h | 4
mdadm/mdadm/mdstat.c | 2
mdadm/mdadm/monitor.c | 11 +
mdadm/mdadm/msg.c | 5
mdadm/mdadm/restripe.c | 1
mdadm/mdadm/super-intel.c | 693 +++++++++++++++++++++++++++++++++------------
mdadm/mdadm/util.c | 88 +++++-
12 files changed, 830 insertions(+), 279 deletions(-)
--
Signature
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
