-
Notifications
You must be signed in to change notification settings - Fork 310
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
DAOS-17271 pool: Fix handle_event -DER_NONEXISTs #16081
base: master
Are you sure you want to change the base?
Conversation
When handling the exclusion of multiple ranks, pool_svc_update_map_internal aborts the whole request and returns -DER_NONEXIST if any of the ranks is absent in the pool map. This is correct for the dmg case, but problematic for the handle_event case, where ranks not in the pool map should simply be ignored. Signed-off-by: Li Wei <liwei@hpe.com>
Ticket title is 'Aurora test cluster can't exclude a faulty engine from a pool: handle_event(): failed to exclude ranks: DER_NONEXIST(-1005): 'The specified entity does not exist'' |
@@ -7033,7 +7033,7 @@ pool_svc_update_map_internal(struct pool_svc *svc, unsigned int opc, bool exclud | |||
inval_tgt_addrs); | |||
if (rc != 0) | |||
goto out_map; | |||
if (inval_tgt_addrs->pta_number > 0) { | |||
if (src == MUS_DMG && inval_tgt_addrs->pta_number > 0) { |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If need to revise the patch otherwise: src and skip_rf_check arguments are not documented in the comment preceding the function code. Incidentally, this may be the first and only use of the src argument in this function.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK.
Test stage Functional Hardware Medium Verbs Provider completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16081/1/execution/node/1557/log |
When handling the exclusion of multiple ranks,
pool_svc_update_map_internal aborts the whole request and returns -DER_NONEXIST if any of the ranks is absent in the pool map. This is correct for the dmg case, but problematic for the handle_event case, where ranks not in the pool map should simply be ignored. (See the Jira ticket for more.)
Steps for the author:
After all prior steps are complete: