github/repmgr
1
0
Fork 0
mirror of https://github.com/EnterpriseDB/repmgr.git synced 2026-03-22 22:56:29 +00:00
Code Issues Packages Projects Releases 1 Wiki Activity

Improve replication connection check

Browse Source 
Previously the check verifying that a node has connected to its upstream
merely assumed the presence of a record in pg_stat_replication indicates
a successful replication connection. However the record may contain a
state other than "streaming", typically "startup" (which will occur when
a node has diverged from its upstream and will therefore never
transition to "streaming"), which needs to be taken into account when
considering the state of the replication connection to avoid false
positives.
This commit is contained in:
Ian Barwick
2020-08-27 16:05:12 +09:00
parent c50a2d049c
commit 0630d9644e
5 changed files with 90 additions and 35 deletions
Download Patch File Download Diff File Expand all files Collapse all files

56
dbutils.c
View File

@@ -5758,16 +5758,16 @@ get_node_replication_stats(PGconn *conn, t_node_info *node_info)
NodeAttached
is_downstream_node_attached(PGconn *conn, char *node_name)
is_downstream_node_attached(PGconn *conn, char *node_name, char **node_state)
{
PQExpBufferData query;
PGresult *res = NULL;
int c = 0;
const char *state = NULL;
initPQExpBuffer(&query);
appendPQExpBuffer(&query,
" SELECT pg_catalog.count(*) "
" SELECT pid, state "
" FROM pg_catalog.pg_stat_replication "
" WHERE application_name = '%s'",
node_name);
@@ -5786,31 +5786,53 @@ is_downstream_node_attached(PGconn *conn, char *node_name)
return NODE_ATTACHED_UNKNOWN;
}
if (PQntuples(res) != 1)
{
log_verbose(LOG_WARNING, _("unexpected number of tuples (%i) returned"), PQntuples(res));
termPQExpBuffer(&query);
/*
* If there's more than one entry in pg_stat_application, there's no
* way we can reliably determine which one belongs to the node we're
* checking, so there's nothing more we can do.
*/
if (PQntuples(res) > 1)
{
log_error(_("multiple entries with \"application_name\" set to \"%s\" found in \"pg_stat_replication\""),
node_name);
log_hint(_("verify that a unique node name is configured for each node"));
termPQExpBuffer(&query);
PQclear(res);
return NODE_ATTACHED_UNKNOWN;
}
c = atoi(PQgetvalue(res, 0, 0));
termPQExpBuffer(&query);
PQclear(res);
if (c == 0)
if (PQntuples(res) == 0)
{
log_verbose(LOG_WARNING, _("node \"%s\" not found in \"pg_stat_replication\""), node_name);
log_warning(_("node \"%s\" not found in \"pg_stat_replication\""), node_name);
PQclear(res);
return NODE_DETACHED;
}
if (c > 1)
log_verbose(LOG_WARNING, _("multiple entries with \"application_name\" set to \"%s\" found in \"pg_stat_replication\""),
node_name);
state = PQgetvalue(res, 0, 1);
if (node_state != NULL)
{
*node_state = palloc0(strlen(state) + 1);
strncpy(*node_state, state, strlen(state));
}
if (strcmp(state, "streaming") != 0)
{
log_warning(_("node \"%s\" attached in state \"%s\""),
node_name,
state);
PQclear(res);
return NODE_NOT_ATTACHED;
}
PQclear(res);
return NODE_ATTACHED;
}