Repository navigation
Conversation
Before this change, ds_pool_update_handler() for reintegration pool_discard() swallowed a nonzero rc returned by dss_rpc_send() for the POOL_TGT_DISCARD corpc. The code execution continued forward to perform pool map update (transition reintegrating engine targets to UP state). With this change, the code does *not* proceed to pool map update when the dss_rpc_send() call returns an error (e.g., -DER_TIMEDOUT). This gives retries either by the MS host daos_engine dsc_pool_client or by the test program itself (re-issuing the dmg pool reintegrate command) a chance to wait out and possibly recover from whatever communication failures are occurring among some engines and the (recently-restarted) reintegrating engine. Features: pool Signed-off-by: Kenneth Cain <kenneth.cain@hpe.com>
|
Ticket title is 'daos_test/suite.py:DaosCoreTest.test_daos_rebuild_ec - pool reintegrate timeout' |
|
Test stage NLT on EL 8.8 completed with status UNSTABLE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos//view/change-requests/job/PR-19254/1/testReport/ |
|
Test stage Functional Hardware Medium Verbs Provider completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/1/display/redirect |
1 similar comment
|
Test stage Functional Hardware Medium Verbs Provider completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/1/display/redirect |
|
Test stage Functional Hardware Medium completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/1/display/redirect |
1 similar comment
|
Test stage Functional Hardware Medium completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/1/display/redirect |
|
Test stage Functional Hardware Medium completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/2/display/redirect |
1 similar comment
|
Test stage Functional Hardware Medium completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/2/display/redirect |
|
Test stage Functional Hardware Medium Verbs Provider completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/2/display/redirect |
1 similar comment
|
Test stage Functional Hardware Medium Verbs Provider completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net/job/daos-stack/job/daos/job/PR-19254/2/display/redirect |
|
Test stage Functional Hardware Medium completed with status FAILURE. https://jenkins-3.daos.hpc.amslabs.hpecorp.net//job/daos-stack/job/daos/view/change-requests/job/PR-19254/3/execution/node/502/log container/boundary.py test failures are unrelated, and an instance of issue DAOS-18610 (fixed on master and release/2.8 and decided not to be fixed on release/2.6 branch, affecting this PR/test run). |
|
Build #1’s NLT instability appears unrelated to this patch: both affected commands succeeded, but Valgrind reported Go-runtime Syscall6 / read(buf) warnings. These are consistent with the intermittent syscall-suppression issue discussed in DAOS-19035 (comment 179329). Subsequent builds restarted at Test Hardware, so they did not provide an NLT retest. |
|
@daos-stack/daos-gatekeeper the associated ticket has request_for_2.6.6 label, and so that needs to be granted first. |
Before this change, ds_pool_update_handler() for reintegration pool_discard() swallowed a nonzero rc returned by dss_rpc_send() for the POOL_TGT_DISCARD corpc. The code execution continued forward to perform pool map update (transition reintegrating engine targets to UP state).
With this change, the code does not proceed to pool map update when the dss_rpc_send() call returns an error (e.g., -DER_TIMEDOUT).
This gives retries either by the MS host daos_engine dsc_pool_client or by the test program itself (re-issuing the dmg pool reintegrate command) a chance to wait out and possibly recover from whatever communication failures are occurring among some engines and the (recently-restarted) reintegrating engine.
Features: pool
Steps for the author:
After all prior steps are complete: