Skip to content

DAOS-19212 object: client retry modification if TX_RESTART repeatedly #18594

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.

Sign up for GitHub

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

Open
Nasf-Fan wants to merge 1 commit into
base: master
Choose a base branch
from
+24 −0
Open

DAOS-19212 object: client retry modification if TX_RESTART repeatedly #18594

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 24 additions & 0 deletions src/object/srv_obj.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3031,6 +3031,7 @@ ds_obj_rw_handler(crt_rpc_t *rpc)
uint32_t max_ver = 0;
struct dtx_epoch epoch = {0};
int rc;
int retry = 0;
bool need_abort = false;

D_ASSERT(orw != NULL);
Expand Down Expand Up @@ -3248,6 +3249,17 @@ ds_obj_rw_handler(crt_rpc_t *rpc)
break;
}

/* If we have already retried once, but still failed for -DER_TX_RESTART, then
* it is quite possible that the -DER_TX_RESTART failure is related with server
* overload or some congestion caused RPC delay. Let's ask client to retry with
* some backoff delay. That will avoid increasing server workload/congestion and
* avoid client RPC timeout during server retry repeatedly.
*/
if (++retry > 1) {
rc = -DER_INPROGRESS;
break;
}

/* Only standalone updates use this RPC. Retry with newer epoch. */
orw->orw_epoch = d_hlc_get();
exec_arg.flags |= ORF_RESEND;
Expand Down Expand Up @@ -4019,6 +4031,7 @@ ds_obj_punch_handler(crt_rpc_t *rpc)
uint32_t max_ver = 0;
struct dtx_epoch epoch;
int rc;
int retry = 0;
bool need_abort = false;

opi = crt_req_get(rpc);
Expand Down Expand Up @@ -4159,6 +4172,17 @@ ds_obj_punch_handler(crt_rpc_t *rpc)
break;
}

/* If we have already retried once, but still failed for -DER_TX_RESTART, then
* it is quite possible that the -DER_TX_RESTART failure is related with server
* overload or some congestion caused RPC delay. Let's ask client to retry with
* some backoff delay. That will avoid increasing server workload/congestion and
* avoid client RPC timeout during server retry repeatedly.
*/
if (++retry > 1) {
rc = -DER_INPROGRESS;
break;
}

/* Only standalone punches use this RPC. Retry with newer epoch. */
opi->opi_epoch = d_hlc_get();
exec_arg.flags |= ORF_RESEND;
Expand Down
Loading