CA-393194: Fix pvremove failure #694
Conversation
|
@rdn32 we probably want to backport this to 8.2 as I suspect it will resolve your partner issue. |
MarkSymsCtx
force-pushed
the
private/stephenche/CA-393194
branch
from
331ee84 to
9132e8f
Compare
MarkSymsCtx
changed the title
|
Whilst this change is in itself fine we will need to add unit tests for the code modified in order to be able to release it. Either in this PR or in a follow-up. test_lvutil.py has separate classes for the various operations and currently does not have anything for VG/PV management |
Sure. I will add unit tests in this PR. |
In lvmohba test, the test failed at `SR.destroy(sr_ref) -> removeVG -> pvremove ` with the lvm error message "No devices to process". The root cause is that the PV which was to be removed was not really used as PV when it comes to multipathed devices. The LVM official document says: "Since each LUN has multiple device nodes in /dev that point to the same underlying data, they all contain the same LVM metadata and thus LVM commands will find the same metadata multiple times and report them as duplicates. These duplicate messages are only warnings and do not mean the LVM operation has failed. Rather, they are alerting the user that only one of the devices has been used as a physical volume and the others are being ignored." Please refer to https://docs.redhat.com/en/documentation/ red_hat_enterprise_linux/6/html/logical_volume_manager_administration /duplicate_pv_multipath#duplicate_pv_multipath This fix is to find the real PV in a VG before removing the VG. Signed-off-by: Stephen Cheng <stephen.cheng@cloud.com>
stephenchengCloud
dismissed stale reviews from TimSmithCtx and MarkSymsCtx
via
e0ecb42
stephenchengCloud
force-pushed
the
private/stephenche/CA-393194
branch
from
9132e8f to
e0ecb42
Compare
|
Added unit tests. |
In lvmohba test, the test failed at
SR.destroy(sr_ref) -> removeVG -> pvremovewith the lvm error message "No devices to process".The root cause is that the PV which was to be removed
was not really used as PV when it comes to multipathed devices.
The LVM official document says:
"Since each LUN has multiple device nodes in /dev that point to the same underlying data,
they all contain the same LVM metadata
and thus LVM commands will find the same metadata
multiple times and report them as duplicates.
These duplicate messages are only warnings
and do not mean the LVM operation has failed.
Rather, they are alerting the user that only one of the devices has been used as a physical volume and the others are being ignored."
Please refer to https://docs.redhat.com/en/documentation/red_hat_enterprise_linux/7/html/logical_volume_manager_administration/duplicate_pv_multipath#duplicate_pv_multipath
This fix is to find the real PV in a VG before removing the VG.
Test passed: 4038532