Sridhar Seshasayee's activity
From 02/28/2021 to 03/29/2021
03/29/2021
- 06:23 AM RADOS Backport #50018 (In Progress): pacific: Test Failed with: "Scrubbing terminated -- not all pgs we...
- 06:23 AM Ceph Revision f6a79be6 (ceph): qa/tasks: Add wait_for_clean() check prior to initiating scrubbing.
- Fixes: https://tracker.ceph.com/issues/49983
Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>
(cherry picked f...
03/25/2021
- 04:44 PM RADOS Bug #49983 (Fix Under Review): Test Failed with: "Scrubbing terminated -- not all pgs were active...
- 02:42 PM RADOS Bug #49983 (Resolved): Test Failed with: "Scrubbing terminated -- not all pgs were active and cle...
- Test Run:
https://pulpito.ceph.com/nojha-2021-03-23_23:04:33-rados-wip-40323-2-distro-basic-gibba/5991116/
Failur... - 04:31 PM Ceph Revision b5570238 (ceph): qa/tasks: Add wait_for_clean() check prior to initiating scrubbing.
- Fixes: https://tracker.ceph.com/issues/49983
Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>
03/24/2021
- 04:19 PM Ceph Revision 4eee3db1 (ceph): PendingReleaseNotes: Document mclock scheduler refinements and profiles
- Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>
03/23/2021
- 12:45 PM Ceph Feature #49698 (Resolved): osd: Add refinements to the mclock built-in profiles implementation.
- 12:45 PM Ceph Backport #49699 (Resolved): pacific: osd: Add refinements to the mclock built-in profiles impleme...
- 12:43 PM Ceph Bug #49844 (Resolved): qa: Add additional wait_for_clean() check in lost_unfound tasks.
- 12:42 PM Ceph Backport #49886 (Resolved): pacific: qa: Add additional wait_for_clean() check in lost_unfound ta...
03/18/2021
- 05:25 PM Ceph Revision 19bc3511 (ceph): osd: Disable sleep times for all best effort clients of mclock
- If mClockScheduler is scheduling IOs then the various sleep options
for the best effort clients of mclock viz. pg_del... - 05:25 PM Ceph Revision 82950511 (ceph): osd: handle config change for cost per io and cost per byte options
- Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>
(cherry picked from commit 33c258a973c9b284194678c5332f918e2e... - 05:24 PM Ceph Revision 54a3b51b (ceph): osd: Add config options for cost per io & byte for the mclock scheduler
- The cost per io and cost per byte options for hdd and ssd are specified
and set to default values determined using ex... - 05:05 PM Ceph Backport #49886 (In Progress): pacific: qa: Add additional wait_for_clean() check in lost_unfound...
- 05:01 PM Ceph Backport #49886 (Resolved): pacific: qa: Add additional wait_for_clean() check in lost_unfound ta...
- https://github.com/ceph/ceph/pull/40221
- 05:05 PM Ceph Revision 35b20c0e (ceph): qa/tasks: Add additional wait_for_clean() check in lost_unfound tasks.
- At the end of the lost_unfound tests add an additional wait_for_clean()
check to ensure that recoveries get enough ti... - 04:58 PM Ceph Bug #49844 (Pending Backport): qa: Add additional wait_for_clean() check in lost_unfound tasks.
- 03:48 PM Infrastructure Bug #49864: Passed jobs classified as dead due to "Could not report results to http://paddles.fro...
- Similar Failures:
https://pulpito.ceph.com/sseshasa-2021-03-18_10:14:02-rados-master-distro-basic-smithi/5976702/
h... - 07:33 AM Ceph Revision 88df4723 (ceph): qa/tasks: Add additional wait_for_clean() check in lost_unfound tasks.
- At the end of the lost_unfound tests add an additional wait_for_clean()
check to ensure that recoveries get enough ti...
03/16/2021
- 08:01 PM Ceph Bug #49844 (Fix Under Review): qa: Add additional wait_for_clean() check in lost_unfound tasks.
- 07:56 PM Ceph Bug #49844 (Resolved): qa: Add additional wait_for_clean() check in lost_unfound tasks.
- Add an additional wait_for_clean() check in lost_unfound tasks so that failures/timeouts down the line can be avoided.
03/10/2021
- 01:30 PM Ceph Revision 2da09122 (ceph): osd: Add config options for cost per io & byte for the mclock scheduler
- The cost per io and cost per byte options for hdd and ssd are specified
and set to default values determined using ex... - 01:30 PM Ceph Revision 33c258a9 (ceph): osd: handle config change for cost per io and cost per byte options
- Signed-off-by: Sridhar Seshasayee <sseshasa@redhat.com>
- 01:30 PM Ceph Revision 18fab905 (ceph): osd: Disable sleep times for all best effort clients of mclock
- If mClockScheduler is scheduling IOs then the various sleep options
for the best effort clients of mclock viz. pg_del... - 10:00 AM Ceph Backport #49699 (In Progress): pacific: osd: Add refinements to the mclock built-in profiles impl...
- 09:59 AM Ceph Revision 71e31bb3 (ceph): osd: Refine implementation that sets built-in profile parameters.
- mclock clients are categorized into client, background_recovery and
background_best_effort. Each client category is r... - 09:59 AM Ceph Revision 52c79e32 (ceph): osd: Fix the OpSchedulerItem cost scaling calculation.
- Calculate the scaled cost for an OpSchedulerItem in msec based on the
overall osd capacity (across all shards). Remov... - 09:42 AM Ceph Feature #49698 (Resolved): osd: Add refinements to the mclock built-in profiles implementation.
- Creating master tracker issue to help backport the associated PR.
Also available in: Atom