Clients should not wait infinitely for etcd to start if it's unavailable

2024-08-28 02:03:35 +03:00
parent 4121c66281
commit f848c450a4
4 changed files with 17 additions and 9 deletions
@@ -485,8 +485,6 @@ static int run(cli_tool_t *p, json11::Json::object cfg)
        p->ringloop = new ring_loop_t(RINGLOOP_DEFAULT_SIZE);
        p->epmgr = new epoll_manager_t(p->ringloop);
        p->cli = new cluster_client_t(p->ringloop, p->epmgr->tfd, cfg_j);
-        // Smaller timeout by default for more interactiveness
-        p->cli->st_cli.etcd_slow_timeout = p->cli->st_cli.etcd_quick_timeout;
        p->loop_and_wait(action_cb, [&](const cli_result_t & r)
        {
            result = r;