Repository navigation
Conversation
mjmac
force-pushed
the
mjmac/DAOS-19803
branch
from
October 29, 2026 17:55
53a4a60 to
1c1a7ac
Compare
|
Ticket title is 'rebuild/widely_striped.py:RbldWidelyStriped.test_rebuild_widely_striped - self-heal-property lookup caused a misleading rebuild timeout' |
mjmac
force-pushed
the
mjmac/DAOS-19803
branch
3 times, most recently
from
October 29, 2026 19:08
45371a3 to
bd45104
Compare
mjmac
changed the title
mjmac
force-pushed
the
mjmac/DAOS-19803
branch
from
October 29, 2026 19:21
39c8e55 to
b8434de
Compare
Simplifies the implementation of dmg pool query by leveraging the fact that the pool service's query RPC already returns a property list. Instead of special-casing self_heal with a separate query mask and PoolGetProp call, extend PoolQuery to allow for an optional set of properties to be returned in a single call. This also future-proofs the code for other properties that may be needed to construct a PoolQuery response down the road. Removes unneeded DPI_SELF_HEAL_POLICY bit from daos_pool.h, as the only consumer was the control plane. Adds a new control plane error to indicate to RPC clients that the targeted replica cannot service a request because none of its engines are running. Control clients treat it as a redirect to the replica's peers. The MS retry loop also now prefers a routing error over whichever error arrived last, and retries against the replicas it has learned rather than its original candidates. Together these fix rebuild test timeouts caused by the prop query stalling for minutes on a replica with no running engines. Features: RbldWidelyStriped Signed-off-by: Michael MacDonald <mjmac@vectorsolutionsgroup.com>
mjmac
force-pushed
the
mjmac/DAOS-19803
branch
from
October 29, 2026 19:24
b8434de to
30de439
Compare
Collaborator
|
Test stage Functional Cluster Box Medium Provider MD on SSD completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-19395/10/execution/node/1700/log |
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Simplifies the implementation of dmg pool query by leveraging the
fact that the pool service's query RPC already returns a property list.
Instead of special-casing self_heal with a separate query mask and
PoolGetProp call, extend PoolQuery to allow for an optional set of
properties to be returned in a single call. This also future-proofs the
code for other properties that may be needed to construct a PoolQuery
response down the road. Removes unneeded DPI_SELF_HEAL_POLICY
bit from daos_pool.h, as the only consumer was the control plane.
Adds a new control plane error to indicate to RPC clients that the
targeted replica cannot service a request because none of its engines
are running. Control clients treat it as a redirect to the replica's
peers. The MS retry loop also now prefers a routing error over whichever
error arrived last, and retries against the replicas it has learned
rather than its original candidates. Together these fix rebuild test
timeouts caused by the prop query stalling for minutes on a replica
with no running engines.
Signed-off-by: Michael MacDonald mjmac@vectorsolutionsgroup.com