light: safeguard skipping of logfiles in disconnected state.

Found by code inspection, neither in practice nor by testing.

Should not occur in practice, because it could only occur after
marsadm pause-fetch, which is an exceptional state only to be entered
for maintenance or for emergency failover.

Skipping over an incorrect logfile at a secondary may produce an
unnecessary split brain.

Fix the potential problem by doing it only after "primary --force",
and by never creating a new logfile, always by re-using existing
logfiles.
This commit is contained in:
Thomas Schoebel-Theuer 2016-02-07 23:36:46 +01:00
parent f235b76900
commit 8bc1e80488
1 changed files with 0 additions and 7 deletions

View File

@ -3285,13 +3285,6 @@ int _make_logging_status(struct mars_rotate *rot)
(skip_new && !_check_allow(global, parent, "connect"))) {
MARS_INF_TO(rot->log_say, "start switchover from transaction log '%s' to '%s'\n", dent->d_path, rot->next_relevant_log->d_path);
_make_new_replaylink(rot, rot->next_relevant_log->d_rest, rot->next_relevant_log->d_serial, rot->next_relevant_log->new_stat.size);
} else if (!_check_allow(global, parent, "connect")) {
char *new_path = path_make("%s/log-%09d-%s", parent->d_path, log_nr + 1, my_id());
if (strcmp(new_path, rot->next_relevant_log->d_path)) {
MARS_WRN("FORCING PRIMARY LOGFILE '%s'\n", new_path);
_create_new_logfile(new_path);
}
brick_string_free(new_path);
}
} else if (rot->todo_primary) {
if (dent->d_serial > log_nr)