Skip to content

DAOS-18949 container: Fix sched_seq assert failures #18269

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

Merged
gnailzenh merged 1 commit into from
May 26, 2026
+14 −14
Merged

DAOS-18949 container: Fix sched_seq assert failures #18269

Show file tree
Hide file tree
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
8 changes: 2 additions & 6 deletions src/container/srv_target.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1045,7 +1045,7 @@ cont_child_start(struct ds_pool_child *pool_child, const uuid_t co_uuid,
D_DEBUG(DB_MD, DF_CONT "[%d]: Container is being destroying (s=%d, d=%d)\n",
DP_CONT(pool_child->spc_uuid, co_uuid), tgt_id, cont_child->sc_stopping,
cont_child->sc_destroying);
rc = -DER_CONT_NONEXIST;
rc = -DER_CONT_DESTROYING;
} else if (cont_child->sc_stopping) {
D_DEBUG(DB_MD, DF_CONT "[%d]: Container is being stopped (s=%d, d=%d)\n",
DP_CONT(pool_child->spc_uuid, co_uuid), tgt_id, cont_child->sc_stopping,
Expand Down Expand Up @@ -1503,14 +1503,10 @@ ds_cont_child_lookup(uuid_t pool_uuid, uuid_t cont_uuid,
if (rc != 0)
return rc;

/**
* Return -DER_CONT_NONEXIST to simplify caller-side handling.
* This may return -DER_CONT_DESTROYING in the future if needed.
**/
if ((*ds_cont)->sc_destroying) {
cont_child_put(tls->dt_cont_cache, *ds_cont);
*ds_cont = NULL;
return -DER_CONT_NONEXIST;
return -DER_CONT_DESTROYING;
}

if ((*ds_cont)->sc_stopping) {
Expand Down
18 changes: 11 additions & 7 deletions src/object/srv_obj_migrate.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1850,21 +1850,23 @@ migrate_get_cont_child(struct migrate_pool_tls *tls, uuid_t cont_uuid,
*/
rc = ds_cont_child_open_create(tls->mpt_pool_uuid, cont_uuid, false, &cont_child);
if (rc != 0) {
if (rc == -DER_CONT_NONEXIST)
if (rc == -DER_CONT_DESTROYING)
D_DEBUG(DB_REBUILD,
DF_RB ": container " DF_UUID
"already destroyed or destroying\n",
DP_RB_MPT(tls), DP_UUID(cont_uuid));
D_ASSERT(cont_child == NULL);
return rc;
}
} else {
rc = ds_cont_child_lookup(tls->mpt_pool_uuid, cont_uuid, &cont_child);
if (rc != 0) {
if (rc == -DER_CONT_NONEXIST)
if (rc == -DER_CONT_NONEXIST || rc == -DER_CONT_DESTROYING)
D_DEBUG(DB_REBUILD,
DF_RB ": container " DF_UUID
"already destroyed or destroying\n",
DP_RB_MPT(tls), DP_UUID(cont_uuid));
D_ASSERT(cont_child == NULL);
return rc;
}
}
Expand Down Expand Up @@ -2262,7 +2264,8 @@ migrate_one_ult(void *arg)
* (nonexistent)
* This is just a workaround...
*/
if (rc != 0 && rc != -DER_CONT_NONEXIST && rc != -DER_DATA_LOSS && tls->mpt_status == 0) {
if (rc != 0 && rc != -DER_CONT_NONEXIST && rc != -DER_CONT_DESTROYING &&
rc != -DER_DATA_LOSS && tls->mpt_status == 0) {
DL_ERROR(rc, DF_RB ": " DF_UOID " rebuild failed, set mpt_fini for tgt %d.",
DP_RB_MPT(tls), DP_UOID(mrone->mo_oid), dss_get_module_info()->dmi_tgt_id);
tls->mpt_status = rc;
Expand Down Expand Up @@ -3274,9 +3277,10 @@ migrate_obj_epoch(struct migrate_pool_tls *tls, struct iter_obj_arg *arg, daos_e
* to rebuild the data, see obj_list_common.
*/
/* If the container is being destroyed, it may return
* -DER_CONT_NONEXIST, see obj_ioc_init().
* -DER_CONT_NONEXIST or -DER_CONT_DESTROYING, see obj_ioc_init().
*/
if (rc == -DER_DATA_LOSS || rc == -DER_CONT_NONEXIST) {
if (rc == -DER_DATA_LOSS || rc == -DER_CONT_NONEXIST ||
rc == -DER_CONT_DESTROYING) {
D_WARN(DF_RB ": mo replicas for " DF_UOID " %d\n", DP_RB_MPT(tls),
DP_UOID(arg->oid), rc);
num = 0;
Expand Down Expand Up @@ -3531,7 +3535,7 @@ migrate_obj_ult(void *data)
if (arg->epoch == DAOS_EPOCH_MAX)
tls->mpt_obj_count++;

if (rc == -DER_CONT_NONEXIST) {
if (rc == -DER_CONT_NONEXIST || rc == -DER_CONT_DESTROYING) {
struct ds_cont_child *cont_child = NULL;

/* check again to see if the container is being destroyed. */
Expand Down Expand Up @@ -4377,7 +4381,7 @@ reint_post_cont_iter_cb(daos_handle_t ih, vos_iter_entry_t *entry,
D_ASSERT(daos_handle_is_valid(cont_toh));

rc = ds_cont_child_lookup(tls->mpt_pool_uuid, entry->ie_couuid, &cont_child);
if (rc == -DER_CONT_NONEXIST) {
if (rc == -DER_CONT_NONEXIST || rc == -DER_CONT_DESTROYING) {
D_DEBUG(DB_REBUILD, DF_RB" co_uuid "DF_UUID" already destroyed or destroying, "
DF_RC"\n", DP_RB_MPT(tls), DP_UUID(entry->ie_couuid), DP_RC(rc));
rc = 0;
Expand Down
2 changes: 1 addition & 1 deletion src/rebuild/scan.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -942,7 +942,7 @@ rebuild_container_scan_cb(daos_handle_t ih, vos_iter_entry_t *entry,
}

rc = ds_cont_child_lookup(rpt->rt_pool_uuid, entry->ie_couuid, &cont_child);
if (rc == -DER_CONT_NONEXIST) {
if (rc == -DER_CONT_NONEXIST || rc == -DER_CONT_DESTROYING) {
D_DEBUG(DB_REBUILD, DF_RB " co_uuid " DF_UUID " already destroyed or destroying\n",
DP_RB_RPT(rpt), DP_UUID(arg->co_uuid));
rc = 0;
Expand Down
Loading