Skip to content
This repository has been archived by the owner on May 3, 2024. It is now read-only.

Disable alternative pool versions by default #1958

Closed
andriytk opened this issue Jul 5, 2022 — with Board Genius Sync · 4 comments · Fixed by #1959
Closed

Disable alternative pool versions by default #1958

andriytk opened this issue Jul 5, 2022 — with Board Genius Sync · 4 comments · Fixed by #1959
Assignees
@andriytk

Comments

@andriytk Board Genius Sync
Copy link
Contributor

andriytk commented Jul 5, 2022 •
edited
Loading

In case of some failure, we want the I/O to be done in the base pool version in the degraded mode always without switching to the alternative pool version. Later, when we introduce the side pool versions, we will do the degraded I/O in them and use them for faster SNS-recovery, which will try to recover only the objects which have the side pool versions instead of traversing all objects in the cluster.

Relates to Seagate/cortx-hare#2123.
@cortx-admin Board Genius Sync
Copy link

For the convenience of the Seagate development team, this issue has been mirrored in a private Seagate Jira Server: https://jts.seagate.com/browse/CORTX-33537. Note that community members will not be able to access that Jira server but that is not a problem since all activity in that Jira mirror will be copied into this GitHub issue.

@andriytk andriytk self-assigned this Jul 5, 2022
andriytk added a commit to andriytk/cortx-motr that referenced this issue Jul 5, 2022
@andriytk
CORTX-33537: disable alternative pvers by default
42de7ea 
We want the I/O to be done in the base pool version in the
degraded mode. Later, when we introduce the side pool versions,
we will do the degraded I/O in them and use them for fasters
SNS-recovery, which will try to recover only the objects which
have the side pool versions instead of traversing all objects
in the cluster.

Closes Seagate#1958.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
@andriytk andriytk mentioned this issue Jul 5, 2022
Merged
12 tasks
andriytk added a commit to andriytk/cortx-motr that referenced this issue Jul 5, 2022
@andriytk
CORTX-33537: disable alternative pvers by default
d304d64 
We want the I/O to be done in the base pool version in the
degraded mode. Later, when we introduce the side pool versions,
we will do the degraded I/O in them and use them for fasters
SNS-recovery, which will try to recover only the objects which
have the side pool versions instead of traversing all objects
in the cluster.

Closes Seagate#1958.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
andriytk added a commit to andriytk/cortx-motr that referenced this issue Jul 6, 2022
@andriytk
CORTX-33537: disable alternative pvers by default
d97232f 
In case of some failure, we want the I/O to be done in the
base pool version in the degraded mode always without switching
to the alternative pool version. Later, when we introduce the
side pool versions, we will do the degraded I/O in them and use
them for faster SNS-recovery, which will try to recover only the
objects which have the side pool versions instead of traversing
all objects in the cluster.

Closes Seagate#1958.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
@cortx-admin cortx-admin removed the dps label Jul 6, 2022 — with Board Genius Sync
andriytk added a commit to andriytk/cortx-motr that referenced this issue Jul 6, 2022
@andriytk
CORTX-33537: disable alternative pvers by default
12f0856 
In case of some failure, we want the I/O to be done in the
base pool version in the degraded mode always without switching
to the alternative pool version. Later, when we introduce the
side pool versions, we will do the degraded I/O in them and use
them for faster SNS-recovery, which will try to recover only the
objects which have the side pool versions instead of traversing
all objects in the cluster.

Closes Seagate#1958.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
andriytk added a commit to andriytk/cortx-motr that referenced this issue Jul 6, 2022
@andriytk
CORTX-33537: allow degraded i/o for the new objects
7881087 
Currently, if no clean pool version (actual or formulaic)
can be found for the new object on its creation, -ENOENT is
returned, which is not good. We want the user to be able to
create new objects even if this implies the degraded i/o on
them.

Solution: return the actual pver at conf_pver_find_locked()
in case when nothing better (cleaner) can be found.

Closes Seagate#1958.
Relates Seagate/cortx-hare#2123.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
andriytk added a commit to andriytk/cortx-motr that referenced this issue Jul 7, 2022
@andriytk
CORTX-33537: allow degraded i/o for the new objects
7c2b423 
Currently, if no clean pool version (actual or formulaic)
can be found for the new object on its creation, -ENOENT is
returned, which is not good. We want the user to be able to
create new objects even if this implies the degraded i/o on
them.

Solution: return the actual pver at conf_pver_find_locked()
in case when nothing better (cleaner) can be found.

Closes Seagate#1958.
Relates Seagate/cortx-hare#2123.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
@cortx-admin Board Genius Sync
Copy link

Andriy Tkachuk commented in Jira Server:

Ready for review at #1959

rkothiya pushed a commit that referenced this issue Jul 11, 2022
@andriytk @rkothiya
CORTX-33537: allow degraded i/o for the new objects
2f43b29 
Currently, if no clean pool version (actual or formulaic)
can be found for the new object on its creation, -ENOENT is
returned, which is not good. We want the user to be able to
create new objects even if this implies the degraded i/o on
them.

Solution: return the actual pver at conf_pver_find_locked()
in case when nothing better (cleaner) can be found.

Closes #1958.
Relates Seagate/cortx-hare#2123.

Signed-off-by: Andriy Tkachuk <andriy.tkachuk@seagate.com>
@cortx-admin Board Genius Sync
Copy link

Gaurav Chaudhari commented in Jira Server:

{panel:bgColor=#c1c7d0}h2. motr - main branch build pipeline SUCCESS
h3. Build Info:

h3. Image Location :

  • cortx-docker.colo.seagate.com/seagate/cortx-all:2.0.0-5740
    cortx-docker.colo.seagate.com/seagate/cortx-rgw:2.0.0-5740
    cortx-docker.colo.seagate.com/seagate/cortx-data:2.0.0-5740
    cortx-docker.colo.seagate.com/seagate/cortx-control:2.0.0-5740
    {panel}

@cortx-admin Board Genius Sync
Copy link

Chandradhar Raval commented in Jira Server:

#1958 is merged

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees

@andriytk andriytk

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

CORTX-33537: allow degraded i/o for the new objects andriytk/cortx-motr
2 participants
@andriytk @cortx-admin