ldap/servers/plugins/replication/repl5_replica_config.c | 43 +++++++++++----- ldap/servers/slapd/plugin.c | 2 2 files changed, 32 insertions(+), 13 deletions(-)
New commits: commit ec3f8da524bf406a24a261b48dd5643ccb48ec7a Author: Mark Reynolds mreynolds@redhat.com Date: Wed Jan 20 10:53:55 2016 -0500
Ticket 48218 - cleanAllRUV - modify the existing "force" option to bypass the "replica online" checks
Bug Description: If a replica is down/offline the cleanallruv task even when using the force option, will wait indefinitely until the replica comes back online.
Fix description: The "force" option should also ignore offline replicas.
https://fedorahosted.org/389/ticket/48218
Reviewed by: nhosoi(Thanks!)
diff --git a/ldap/servers/plugins/replication/repl5_replica_config.c b/ldap/servers/plugins/replication/repl5_replica_config.c index 4d7135c..14020b7 100644 --- a/ldap/servers/plugins/replication/repl5_replica_config.c +++ b/ldap/servers/plugins/replication/repl5_replica_config.c @@ -67,7 +67,7 @@ static int replica_cleanallruv_send_extop(Repl_Agmt *ra, cleanruv_data *data, in static int replica_cleanallruv_send_abort_extop(Repl_Agmt *ra, Slapi_Task *task, struct berval *payload); static int replica_cleanallruv_check_maxcsn(Repl_Agmt *agmt, char *basedn, char *rid_text, char *maxcsn, Slapi_Task *task); static int replica_cleanallruv_replica_alive(Repl_Agmt *agmt); -static int replica_cleanallruv_check_ruv(char *repl_root, Repl_Agmt *ra, char *rid_text, Slapi_Task *task); +static int replica_cleanallruv_check_ruv(char *repl_root, Repl_Agmt *ra, char *rid_text, Slapi_Task *task, char *force); static int get_cleanruv_task_count(); static int get_abort_cleanruv_task_count(); static int replica_cleanup_task (Object *r, const char *task_name, char *returntext, int apply_mods); @@ -1801,7 +1801,10 @@ replica_cleanallruv_thread(void *arg) ruv_obj = replica_get_ruv(data->replica); ruv = object_get_data (ruv_obj); while(data->maxcsn && !is_task_aborted(data->rid) && !is_cleaned_rid(data->rid) && !slapi_is_shutting_down()){ - if(csn_get_replicaid(data->maxcsn) == 0 || ruv_covers_csn_cleanallruv(ruv,data->maxcsn) || strcasecmp(data->force,"yes") == 0){ + if(csn_get_replicaid(data->maxcsn) == 0 || + ruv_covers_csn_cleanallruv(ruv,data->maxcsn) || + strcasecmp(data->force,"yes") == 0) + { /* We are caught up, now we can clean the ruv's */ break; } @@ -1816,7 +1819,7 @@ replica_cleanallruv_thread(void *arg) * Even if we are forcing the cleaning, the replicas still need to be up */ cleanruv_log(data->task, data->rid, CLEANALLRUV_ID,"Waiting for all the replicas to be online..."); - if(check_agmts_are_alive(data->replica, data->rid, data->task)){ + if(strcasecmp(data->force, "no") == 0 && check_agmts_are_alive(data->replica, data->rid, data->task)){ /* error, aborted or shutdown */ aborted = 1; goto done; @@ -1857,7 +1860,9 @@ replica_cleanallruv_thread(void *arg) } else { agmt_not_notified = 1; cleanruv_log(data->task, data->rid, CLEANALLRUV_ID, "Failed to send task to replica (%s)",agmt_get_long_name(agmt)); - break; + if(strcasecmp(data->force,"no") == 0){ + break; + } } agmt_obj = agmtlist_get_next_agreement_for_replica (data->replica, agmt_obj); } @@ -1866,7 +1871,7 @@ replica_cleanallruv_thread(void *arg) aborted = 1; goto done; } - if(agmt_not_notified == 0){ + if(agmt_not_notified == 0 || strcasecmp(data->force, "yes") == 0){ break; } /* @@ -1909,7 +1914,7 @@ replica_cleanallruv_thread(void *arg) found_dirty_rid = 0; continue; } - if(replica_cleanallruv_check_ruv(data->repl_root, agmt, rid_text, data->task) == 0){ + if(replica_cleanallruv_check_ruv(data->repl_root, agmt, rid_text, data->task, data->force) == 0){ found_dirty_rid = 0; } else { found_dirty_rid = 1; @@ -1924,7 +1929,7 @@ replica_cleanallruv_thread(void *arg) aborted = 1; goto done; } - if(found_dirty_rid == 0){ + if(found_dirty_rid == 0 || strcasecmp(data->force, "yes") == 0){ break; } /* @@ -2033,7 +2038,8 @@ check_replicas_are_done_cleaning(cleanruv_data *data ) int not_all_cleaned = 1; int interval = 10;
- cleanruv_log(data->task, data->rid, CLEANALLRUV_ID, "Waiting for all the replicas to finish cleaning..."); + cleanruv_log(data->task, data->rid, CLEANALLRUV_ID, + "Waiting for all the replicas to finish cleaning...");
csn_as_string(data->maxcsn, PR_FALSE, csnstr); filter = PR_smprintf("(%s=%d:%s:%s)", type_replicaCleanRUV,(int)data->rid, csnstr, data->force); @@ -2058,15 +2064,22 @@ check_replicas_are_done_cleaning(cleanruv_data *data ) } agmt_obj = agmtlist_get_next_agreement_for_replica (data->replica, agmt_obj); } - if(not_all_cleaned == 0 || is_task_aborted(data->rid) ){ + if(not_all_cleaned == 0 || + is_task_aborted(data->rid) || + strcasecmp(data->force, "yes") == 0) + { break; } - cleanruv_log(data->task, data->rid, CLEANALLRUV_ID, "Not all replicas finished cleaning, retrying in %d seconds",interval); + + cleanruv_log(data->task, data->rid, CLEANALLRUV_ID, + "Not all replicas finished cleaning, retrying in %d seconds", + interval); if(!slapi_is_shutting_down()){ PR_Lock( notify_lock ); PR_WaitCondVar( notify_cvar, PR_SecondsToInterval(interval) ); PR_Unlock( notify_lock ); } + if(interval < 14400){ /* 4 hour max */ interval = interval * 2; } else { @@ -3400,7 +3413,7 @@ replica_cleanallruv_replica_alive(Repl_Agmt *agmt) }
static int -replica_cleanallruv_check_ruv(char *repl_root, Repl_Agmt *agmt, char *rid_text, Slapi_Task *task) +replica_cleanallruv_check_ruv(char *repl_root, Repl_Agmt *agmt, char *rid_text, Slapi_Task *task, char *force) { Repl_Connection *conn = NULL; ConnResult crc = 0; @@ -3410,6 +3423,9 @@ replica_cleanallruv_check_ruv(char *repl_root, Repl_Agmt *agmt, char *rid_text, int rc = -1;
if((conn = conn_new(agmt)) == NULL){ + if(strcasecmp(force, "yes") == 0){ + return 0; + } return rc; }
@@ -3440,6 +3456,11 @@ replica_cleanallruv_check_ruv(char *repl_root, Repl_Agmt *agmt, char *rid_text, ber_bvfree(retsdata); } } + } else { + if (strcasecmp(force, "yes") == 0){ + /* We are forcing, we don't care that the replica is not online */ + rc = 0; + } } conn_delete_internal_ext(conn); slapi_ch_free_string(&data); diff --git a/ldap/servers/slapd/plugin.c b/ldap/servers/slapd/plugin.c index 2d3a5ce..ddf2631 100644 --- a/ldap/servers/slapd/plugin.c +++ b/ldap/servers/slapd/plugin.c @@ -3625,8 +3625,6 @@ plugin_invoke_plugin_pb (struct slapdplugin *plugin, int operation, Slapi_PBlock PR_ASSERT (pb->pb_op);
target_spec = operation_get_target_spec (pb->pb_op); - - PR_ASSERT (target_spec);
rc = plugin_invoke_plugin_sdn (plugin, operation, pb, target_spec);
389-commits@lists.fedoraproject.org
-
Mark Reynolds