Description of problem: When using gdeploy to setup ctdb it picks up the list from hosts as ctdb_nodes and stores it into /etc/ctdb/nodes, regardless of finding hostnames or ip addresses. Since ctdb expects ip address format for entries in /etc/ctdb/nodes it fails to start with error message. Version-Release number of selected component (if applicable): gdeploy-2.0.2-14.el7rhgs.noarch How reproducible: Always. Steps to Reproduce: 1. use hostnames instead of ip addresses in [hosts] section of gdeploy config file 2. don't specify ctdb_nodes list 3. run gdeploy Actual results: gdeploy fails with error messages: PLAY [gluster_servers] ************************************************************************************************************************ TASK [start/stop/restart/reload services] ***************************************************************************************************** failed: [gl-t3] (item=ctdb) => {"failed": true, "item": "ctdb", "msg": "Unable to start service ctdb: Job for ctdb.service failed because the control process exited with error code. See \"systemctl status ctdb.service\" and \"journalctl -xe\" for details.\n"} failed: [gl-t2] (item=ctdb) => {"failed": true, "item": "ctdb", "msg": "Unable to start service ctdb: Job for ctdb.service failed because the control process exited with error code. See \"systemctl status ctdb.service\" and \"journalctl -xe\" for details.\n"} failed: [gl-t1] (item=ctdb) => {"failed": true, "item": "ctdb", "msg": "Unable to start service ctdb: Job for ctdb.service failed because the control process exited with error code. See \"systemctl status ctdb.service\" and \"journalctl -xe\" for details.\n"} to retry, use: --limit @/tmp/tmpyUmGR5/service_management.retry and from /var/log/log.ctdb: 2017/11/21 00:44:12.235065 ctdbd[7040]: CTDB starting on node 2017/11/21 00:44:12.236956 ctdbd[7040]: ../ctdb/common/system_util.c:123 Failed to translate gl-t1 into sin_addr 2017/11/21 00:44:12.236973 ctdbd[7040]: Failed to load nodes file "/etc/ctdb/nodes" Expected results: gdeploy should not put hostnames into /etc/ctdb/nodes Additional info: full trace see https://github.com/mattmuench/bug-traces/blob/master/bug-ctdb_nodes--gdeploy-2.0.2-14
Hi, PR[1] fixes the issue. When ctdb_nodes are not set, the default behaviour of gdeploy is to pick up hosts from the host list. gdeploy should throw error when hostnames are given instead of IPs in this case. [1] https://github.com/gluster/gdeploy/pull/471 Thanks, Devyani Kota
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2018:1958