DAOS-16908 object: add client-side target compound rpc pinging on update retry #16093
Conversation
This change adds logic which pings all targets that are involved in the object retry. When the retry function gets an error signifying that the server could not reach clients, the update will ping the relevant targets to establish a connection so the update can retry. Signed-off-by: Yokesh Jayakumar <karthj@google.com>
Signed-off-by: Yokesh Jayakumar <karthj@google.com>
Signed-off-by: Yokesh Jayakumar <karthj@google.com>
Signed-off-by: Yokesh Jayakumar <karthj@google.com>
Previously, I was getting an error in the unit test saying that HG_Finalize could not work since the bulk handle was not being freed. This is because we were incorrectly returning early. Signed-off-by: Yokesh Jayakumar <karthj@google.com>
Signed-off-by: Yokesh Jayakumar <karthj@google.com>
Signed-off-by: Yokesh Jayakumar <karthj@google.com>
|
Ticket title is 'Modify DAOS to use new mercury changes to implement improved firewall handling' |
|
Test stage Build on Leap 15.5 with Intel-C and TARGET_PREFIX completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/1/execution/node/307/log |
|
Test stage Build on EL 8 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/1/execution/node/384/log |
|
Test stage Build RPM on EL 8 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/1/execution/node/292/log |
|
Test stage Build RPM on EL 9 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/1/execution/node/369/log |
|
Test stage Build RPM on Leap 15.5 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/1/execution/node/261/log |
karthjyojay
force-pushed
the
dev/karthj/firewall-simplification-compound-rpc
branch
from
f0f8003 to
25bb7eb
Compare
|
Test stage Build on EL 8 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/2/execution/node/360/log |
|
Test stage Build RPM on EL 8 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/2/execution/node/323/log |
|
Test stage Build RPM on EL 9 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/2/execution/node/320/log |
|
Test stage Build RPM on Leap 15.5 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/2/execution/node/319/log |
|
Test stage Build DEB on Ubuntu 20.04 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/2/execution/node/359/log |
When the transaction's return code is equal to DER_RECONNECT, that means one of the packed operations resulted in the server not being able to establish a connection to a client. When this happens, we should have the client ping the server. In this case, we will retry all targets across all update operations in the compound RPC because we don't know which subrequest resulted in the DER_RECONNECT. Signed-off-by: Yokesh Jayakumar <karthj@google.com>
karthjyojay
force-pushed
the
dev/karthj/firewall-simplification-compound-rpc
branch
from
25bb7eb to
31d0dc9
Compare
karthjyojay
force-pushed
the
dev/karthj/firewall-simplification
branch
3 times, most recently
from
60cee9a to
3493494
Compare
|
Test stage Build RPM on EL 8 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/4/execution/node/306/log |
|
Test stage Build RPM on Leap 15.5 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/4/execution/node/403/log |
|
Test stage Build RPM on EL 9 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/4/execution/node/390/log |
|
Test stage Build DEB on Ubuntu 20.04 completed with status FAILURE. https://build.hpdd.intel.com//job/daos-stack/job/daos/view/change-requests/job/PR-16093/4/execution/node/443/log |
When the transaction's return code is equal to DER_RECONNECT, that means one of the packed operations resulted in the server not being able to establish a connection to a client. When this happens, we should have the client ping the server. In this case, we will retry all targets across all update operations in the compound RPC because we don't know which subrequest resulted in the DER_RECONNECT.
Steps for the author:
After all prior steps are complete: