Snap_schedule does not always work.
Hi Teams, *Ceph-version*: Quincy, Reef *OS*: Almalinux 8 *Issue*: snap_schedule doesn't create the scheduled snapshots consistently. *Description:* Hi team, We are currently working in a 3-node ceph cluster. We are currently exploring the scheduled snapshot capability of the ceph-mgr module. To enable/configure scheduled snapshots, we followed the following link: https://docs.ceph.com/en/quincy/cephfs/snap-schedule/ Using this we were able to schedule a snapshot for the subvolume which we created. Initially, it was not working but after two days it started working. Later on a fresh cluster we again tried to schedule the snapshot for a subvolume and it worked fine. But next time when we again tried the snapshot creation on a fresh cluster it didn't work this time. We have observed this inconsistency in scheduling the snapshots for a long time now. Everytime we follow the exact same steps to add a snap_schedule but sometimes it works and sometimes it doesn't. We have a chronyd service running and the timezone set to UTC timezone. Could you please help us out with this? Kindly let me know if you require any kind of logs for this Thanks and Regards, Kushagra Gupta
Hello Kushagr, Snap-schedule no longer accepts a --subvol argument, so it's not easily possible to schedule snapshots for subvolumes. Could you tell the commands used to schedule snapshots for subvolumes ? -- Milind On Wed, Sep 27, 2023 at 11:13 PM Kushagr Gupta <kushagrguptasps.mun@gmail.com> wrote:
Hi Teams,
*Ceph-version*: Quincy, Reef *OS*: Almalinux 8
*Issue*: snap_schedule doesn't create the scheduled snapshots consistently.
*Description:* Hi team,
We are currently working in a 3-node ceph cluster. We are currently exploring the scheduled snapshot capability of the ceph-mgr module. To enable/configure scheduled snapshots, we followed the following link:
https://docs.ceph.com/en/quincy/cephfs/snap-schedule/
Using this we were able to schedule a snapshot for the subvolume which we created. Initially, it was not working but after two days it started working.
Later on a fresh cluster we again tried to schedule the snapshot for a subvolume and it worked fine. But next time when we again tried the snapshot creation on a fresh cluster it didn't work this time.
We have observed this inconsistency in scheduling the snapshots for a long time now. Everytime we follow the exact same steps to add a snap_schedule but sometimes it works and sometimes it doesn't.
We have a chronyd service running and the timezone set to UTC timezone. Could you please help us out with this? Kindly let me know if you require any kind of logs for this
Thanks and Regards, Kushagra Gupta _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
-- Milind
Hi Milind,Team Thank you for the response @Milind.
Snap-schedule no longer accepts a --subvol argument, Thank you for the information.
Currently, we are using the following commands to create the snap-schedules: Syntax: *"ceph fs snap-schedule add /<volume-name>/<subvolume_group>/<subvolume> <frequency_of_snapshot> <Timeinformat_yyyymmddTHH:MM:ss>"* *"ceph fs snap-schedule retention add <path> <retention_spec_or_period>"* Example: *"ceph fs snap-schedule add /volumes/hns/test/e4ef5ce8-ebaf-46c9-ae69-6bc70602d619 1h 2023-06-07T11:05:00"* *"ceph fs snap-schedule retention add **/volumes/hns/test/e4ef5ce8-ebaf-46c9-ae69-6bc70602d619 h 4"* We were able to create snap_schedules with the above commands along with the retention policies. This command should start creating the scheduled snapshots (As per the document after 1 hour from the scheduled time) *.* In some instances, the scheduler created the scheduled snapshot. But on a fresh setup when we executed the same commands as per the setup, the scheduler did not create the scheduled snapshots. We have observed this behavior multiple times. Could you please help us out? Kindly let me know if anything else is required Thanks and Regards, Kushagra Gupta On Thu, Sep 28, 2023 at 8:00 AM Milind Changire <mchangir@redhat.com> wrote:
Hello Kushagr, Snap-schedule no longer accepts a --subvol argument, so it's not easily possible to schedule snapshots for subvolumes. Could you tell the commands used to schedule snapshots for subvolumes ?
-- Milind
On Wed, Sep 27, 2023 at 11:13 PM Kushagr Gupta <kushagrguptasps.mun@gmail.com> wrote:
Hi Teams,
*Ceph-version*: Quincy, Reef *OS*: Almalinux 8
*Issue*: snap_schedule doesn't create the scheduled snapshots
consistently.
*Description:* Hi team,
We are currently working in a 3-node ceph cluster. We are currently exploring the scheduled snapshot capability of the ceph-mgr module. To enable/configure scheduled snapshots, we followed the following link:
https://docs.ceph.com/en/quincy/cephfs/snap-schedule/
Using this we were able to schedule a snapshot for the subvolume which we created. Initially, it was not working but after two days it started working.
Later on a fresh cluster we again tried to schedule the snapshot for a subvolume and it worked fine. But next time when we again tried the snapshot creation on a fresh
cluster
it didn't work this time.
We have observed this inconsistency in scheduling the snapshots for a long time now. Everytime we follow the exact same steps to add a snap_schedule but sometimes it works and sometimes it doesn't.
We have a chronyd service running and the timezone set to UTC timezone. Could you please help us out with this? Kindly let me know if you require any kind of logs for this
Thanks and Regards, Kushagra Gupta _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
-- Milind
the filesystem path specified for scheduling snapshots is incorrect see below ... On Thu, Sep 28, 2023 at 9:52 AM Kushagr Gupta <kushagrguptasps.mun@gmail.com> wrote:
Hi Milind,Team
Thank you for the response @Milind.
Snap-schedule no longer accepts a --subvol argument, Thank you for the information.
Currently, we are using the following commands to create the snap-schedules: Syntax: "ceph fs snap-schedule add /<volume-name>/<subvolume_group>/<subvolume> <frequency_of_snapshot> <Timeinformat_yyyymmddTHH:MM:ss>" "ceph fs snap-schedule retention add <path> <retention_spec_or_period>"
Example: "ceph fs snap-schedule add /volumes/hns/test/e4ef5ce8-ebaf-46c9-ae69-6bc70602d619 1h 2023-06-07T11:05:00" "ceph fs snap-schedule retention add /volumes/hns/test/e4ef5ce8-ebaf-46c9-ae69-6bc70602d619 h 4"
the path needs to be shortened as follows assuming that the path is a system generated path ... "ceph fs snap-schedule add /volumes/hns/test 1h 2023-06-07T11:05:00" "ceph fs snap-schedule retention add /volumes/hns/test h 4" the .snap dir is created one level up than the actual path returned py the 'getpath' command where: 'hns' is the subvolume group name and 'test' is the subvolume name
We were able to create snap_schedules with the above commands along with the retention policies. This command should start creating the scheduled snapshots (As per the document after 1 hour from the scheduled time) . In some instances, the scheduler created the scheduled snapshot. But on a fresh setup when we executed the same commands as per the setup, the scheduler did not create the scheduled snapshots. We have observed this behavior multiple times.
Could you please help us out? Kindly let me know if anything else is required
Thanks and Regards, Kushagra Gupta
On Thu, Sep 28, 2023 at 8:00 AM Milind Changire <mchangir@redhat.com> wrote:
Hello Kushagr, Snap-schedule no longer accepts a --subvol argument, so it's not easily possible to schedule snapshots for subvolumes. Could you tell the commands used to schedule snapshots for subvolumes ?
-- Milind
On Wed, Sep 27, 2023 at 11:13 PM Kushagr Gupta <kushagrguptasps.mun@gmail.com> wrote:
Hi Teams,
*Ceph-version*: Quincy, Reef *OS*: Almalinux 8
*Issue*: snap_schedule doesn't create the scheduled snapshots consistently.
*Description:* Hi team,
We are currently working in a 3-node ceph cluster. We are currently exploring the scheduled snapshot capability of the ceph-mgr module. To enable/configure scheduled snapshots, we followed the following link:
https://docs.ceph.com/en/quincy/cephfs/snap-schedule/
Using this we were able to schedule a snapshot for the subvolume which we created. Initially, it was not working but after two days it started working.
Later on a fresh cluster we again tried to schedule the snapshot for a subvolume and it worked fine. But next time when we again tried the snapshot creation on a fresh cluster it didn't work this time.
We have observed this inconsistency in scheduling the snapshots for a long time now. Everytime we follow the exact same steps to add a snap_schedule but sometimes it works and sometimes it doesn't.
We have a chronyd service running and the timezone set to UTC timezone. Could you please help us out with this? Kindly let me know if you require any kind of logs for this
Thanks and Regards, Kushagra Gupta _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io
-- Milind
-- Milind
participants (2)
-
Kushagr Gupta
-
Milind Changire