lp://qastaging/~csurbhi/junk/mdadm.fixes

Branch merges

Propose for merging

No branches dependent on this one.

Related source package recipes

No recipes using this branch. (?)

Create packaging recipe

Related snap packages

Related bugs

Related blueprints

18. By Surbhi Palande on 2010-10-05

Fixed locking

When lockf is used for locking, the lock is released when you close the file.
So, when the file is closed, the process waiting for a lock gets it, but
thereafter the file is unlinked. So when another process asks for a lock on
the same file, it too gets the lock on a new file that it creates. Hence we
have two processes having a lock, one process with a lock on the stale file
and the other on a fresh new one. This breaks our locking scheme. So we
changed the locking to flock() and check if the file is stale using the stat()
and by unlinking the file before closing.

Signed-off-by: Surbhi Palande <email address hidden>

17. By Surbhi Palande on 2010-10-04

changelog changes
Signed-off-by: Surbhi Palande <email address hidden>

16. By Surbhi Palande on 2010-10-01

Partition creation logic fixing

Used the logic of creating partitions from create_mddev().

Signed-off-by: Surbhi Palande <email address hidden>

15. By Surbhi Palande on 2010-09-30

Bug fixing for mdadm map file reading and dev name choosing

Signed-off-by: Surbhi Palande <email address hidden>

14. By Surbhi Palande on 2010-09-29

Add locks for Manage_runstop()

Before deleting an entry from the maps file, lock the file and use the most
updated copy.
Signed-off-by: Surbhi Palande <email address hidden>

13. By NeilBrown <email address hidden> on 2010-09-29

Incremental: lock against multiple concurrent additions to an array.

In two devices are added via -I to one array at the same time, mdadm
can get badly confused.

Signed-off-by: NeilBrown <email address hidden>
Signed-off-by: Surbhi Palande <email address hidden>

12. By NeilBrown <email address hidden> on 2010-09-29

Rename open_mddev to create_mddev

This reflect that fact that more often than not it is creating things
in /dev, and allows for a new open_mddev which does just that.

NeilBrown: 2399204dddcabcf629ccb57be1957010444fc0ea

Signed-off-by: NeilBrown <email address hidden>
Signed-off-by: Surbhi Palande <email address hidden>

11. By NeilBrown <email address hidden> on 2010-09-24

mapfile - when rebuilding, choose an appropriate name is none is found.

When rebuilding the mapfile (mdadm -Ir), if not appropriate name is
found in /dev/md/, try to find an appropriate name, either by looking
in mdadm.conf or by using the name in the metadata.

Used part of the commit from Neil Browns tree:
360b463696e89593645c0f094f0c39ba0fa65944

Signed-off-by: NeilBrown <email address hidden>
Signed-off-by: Surbhi Palande <email address hidden>

10. By NeilBrown <email address hidden> on 2010-09-24

Adjust major number testing to allow for extended minor number in 2.6.28

From 2.6.28, normal md device will be able to have partitions. These
partitions will have a different major number. Sometimes mdadm tests
the major number and so can get confused.
Change these tests to test against get_mdp_major(). mdp does not use
extended minor number and so this test will always be accurate.

Also use /sys/dev links to map major/minor to devnum in sysfs.

Neil Brown: 4ebd3237119b1c1d701ea0c94795631883e449ed

Signed-off-by: NeilBrown <email address hidden>
Signed-off-by: Surbhi Palande <email address hidden>

9. By Neil Brown <email address hidden> on 2010-09-24

Fix an error when assembling arrays that are in the middle of a reshape.

It is important that dup_super always returns an 'st' with the same
->ss and ->minor_version as the st that was passed.
This wasn't happening for 0.91 metadata (i.e. in the middle of a reshape).

Neil Brown: 56f8add211a840faaed325bd16483b55da544e93

Signed-off-by: Surbhi Palande <email address hidden>