From e7094c60a054658372981fd983811c8670042424 Mon Sep 17 00:00:00 2001 From: xin liang Date: Fri, 15 Nov 2024 11:22:36 +0800 Subject: [PATCH] Dev: ChangeLog: update ChangeLog for release 4.6.1 --- ChangeLog | 143 ++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 143 insertions(+) diff --git a/ChangeLog b/ChangeLog index dfc5912a0b..9f1ec9d17d 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,146 @@ +* Fri Nov 15 2024 Xin Liang +- Release 4.6.1 +- Dev: bootstrap: more robust implementation for ssh_merge (bsc#1230530) +- Fix: log: The report DEBUG log message is not displayed in the log file +- Dev: report.utils: For a sequence of archived log files, check the modify time +- Fix: report.utils: Fix the performance issue (bsc#1232821) +- Dev: report.utils: Add debug info for the log file types +- Fix: report.collect: Make sure the log is not None before using it (bsc#1232821) +- Dev: bootstrap: Add log info when starting pacemaker.service +- Dev: utils: Introduced `detect_duplicate_device_path` function in utils +- Fix: cibconfig: Disable auto add advise values for operations (bsc#1231386) +- Fix: ui_cluster: Stop renaming cluster name when using qdevice +- Dev: utils: Check node is reachable by using both ping and ssh +- Dev: report: do not capture stderr when unarchiving tarballs +- Dev: report: make error messages easier to parse for hawk2 (bsc#1228899) +- Fix: report: find_shell should accept hacluster user (bsc#1228899) +- Dev: healthcheck: Enhace log clarity +- Dev: ui_cluster: Change the completer for crm cluster health +- Dev: report: add a reminder to use `crm cluster health` (bsc#1228899) +- Dev: ui_cluster: add a reminder to use "crm cluster init/join ssh" to +initialize ssh (bsc#1228899) +- Dev: scripts: add a reminder to use `crm cluster health` to fix hacluster +passwordless ssh authentication (bsc#1228899) +- Dev: scripts: generate readable messages when ssh authentication fails (bsc#1228899) +- Fix: report: should not try interactive authentication when stdin is not a +tty (bsc#1228899) +- Dev: main: remove upgradeutil (bsc#1228899) +- Fix: bootstrap: check is_nologin more robustly (bsc#1228251) +- Dev: completers: Reuse node completer for cluster remove and health +- Dev: bootstrap: Enhance log clarity during crm cluster remove process +- Dev: utils: Catch PermissionError when reading files +- Dev: utils: Load CIB_file env before some readonly commands +- Dev: Rename variables to make them look like boolean type +- Fix: utils: Stop providing the detailed and precise sudoer rules to "hack" +the privilege (bsc#1229093) +- Dev: ui_node: Improve command 'clearstate ' +- Dev: Rename the option 'add_advised_op_values' to 'ra_advised_op_values' +- Dev: Add a new internal flag to control auto add time units on operation +- Fix: Add a new option 'fa_advised_op_values' (bsc#1228858) +- Fix: utils: group check failure when os.getgroups() returns empty (bsc#1229030) +- Fix: report: When 'core.no_ssh' set to 'yes', crm report works in local mode (bsc#1228899) +- Fix: completers: deferring calls to crm_mon until envsetup() is called (bsc#1228271) +- Fix: utils: allow haclient group to use cluster level commands (bsc#1228271) +- Fix: scripts: call commands as current user when running as hacluster (bsc#1228271) +- Fix: prun: allow pcopy_to_remote to use intercept (bsc#1228271) +- Fix: upgradeutil: Immdiately return if config.core.no_ssh is set (bsc#1228899) +- Fix: ui_cluster: Handle 'crm cluster start --all' command (bsc#1228899) +- Fix: ui_cluster: Handle 'crm cluster stop --all' command (bsc#1228899) +- Fix: utils: Define utils.NoSSHError exception and ssh wrapper function (bsc#1228899) +- Fix: config: Add 'core.no_ssh' option (bsc#1228899) +- Dev: sh: Ensure CommandFailure Exception is Picklable Across Processes (bsc#1229686) +- Fix: report: Error output of crm_verify should be recorded in report result (bsc#1229686) +- Fix: Don't add time units to values for existing CIB (bsc#1228817) +- Fix: bootstrap: drop environ SSH_AUTH_SOCK before checking passwordless ssh +when it is not enabled (bsc#1228950) +- Fix: bootstrap: should check if sudo is available when running `cluster join +-c` with a non-root destination user (bsc#1228950) +- Fix: bootstrap: should check if sudo is available when running `cluster init +-N` with a non-root destination user (bsc#1228950) +- Dev: Add option core.add_advised_op_values, default value is 'yes' +- Dev: Rename variable 'add_default_op_values' to 'add_advised_op_values' +- Revert "Dev: ui_configure: Deprecate configure erase sub-command" (bsc#1228713) +- Fix: hahealth.py script (bsc#1228271) +- Dev: qdevice: numbers automatically the steps to generate client certs +- Dev: prun: add more possible libexec PATH for sftp-server +- Fix: qdevice: config "tls" should accept value "required" (bsc#1227649) +- Fix: bootstrap: refine the wording in the question asking TLS config for +qdevice (bsc#1227649) +- Fix: qdevice: TLS certs should always be generated for qnetd (bsc#1227649) +- Dev: Rename variables to distinguish the concept of adding default +- Fix: cli: loading cib should be a soft depenedency +- Dev: ui_context: refactor NON_FUNCTIONAL_COMMAND and NON_FUNCTIONAL_OPTIONS +- Revert changes to `quit` and `up` in ui_context from #1300 (#1466) +- Fix: cibconfig: do not load cib to check if cib is changed (#1466) +- Dev: ra: Refine RAInfo.params interface to exclude some completion results +- Dev: ui_context: Don't append space when completion ends with equal sign +- Dev: ui_configure: Complete required parameters first +- Dev: ra: Guess stonith class type when using fence agents +- Fix: upgradeutil: refine error handling (bsc#1226147, bsc#1223371) +- Fix: cibconfig: Disable complete advised operation values when adding a rsc_template +- Dev: ui_resource: Improve completers for 'crm resource' command +- Fix: cmd_status: call crm_mon without shell (#1429) +- Dev: log an error when fencing node without stonith device configured and running +- Fix: healthcheck: KeyError when local nodename not found in cib (bsc#1223438) +- Dev: utils: Add info when property is newly added +- Dev: corosync: Show corosync cpg status in `crm corosync status cpg` +- Dev: ui_context: ignore requirements when option --help is specified +- Dev: ui_context: Skip querying CIB when in a sublevel or help command +- Dev: bootstrap: Setup the stage dependency on init and join side (bsc#1175865, bsc#1219940) +- Dev: bootstrap: Enhance stage validation +- Fix: healthcheck: Add crmsh.constants.SSH_OPTION when doing ssh in check_local +- Fix: healthcheck: Missing 'id_' prefix while checking the ssh key existence +- Fix: main: Set PATH environment first +- Dev: bash_completion: Enable complete 'id=' on non-interactive mode +- Fix: bootstrap: Detect cluster service on init node before saving the +canonical hostname (bsc#1222714) +- Fix: ui_node: When `utils.list_cluster_nodes` return None, try to get ip +list from corosync.conf +- Dev: report: Collect quorum/qdevice/qnetd status +- Dev: ui_context: make help subcommands to exit with 0 (#1374) +- Dev: bootstrap: Add all nodes' keys to qnetd authroized_keys even init +- Fix: bootstrap: fix NameError +- Dev: qdevice: Refactor init_qdevice function +- Dev: qdevice: Add all nodes' keys to qnetd authorized_keys +- Dev: bootstrap: Give a warning when detecting $SSH_AUTH_SOCK but not use +--use-ssh-agent option +- Fix: sh: Return the value of AuthorizationError.diagnose if it is not None +- Dev: ssh_key: Extract duplicate code about list keys from ssh-agent to a function +- Dev: bootstrap: Refactor qdevice user parsing and finding +- Fix: utils: set env `CIB_shadow` using `os.environ` (bsc#1205925) +- Fix: pass env to child process explicitly (bsc#1205925) +- Fix: term: unset env `COLUMNS` and `ROWS` (bsc#1205925) +- Fix: sh: pass env to child process explicitly (bsc#1205925) +- Fix: report: Show different perspectives of cluster +- Fix: ui_node: prevent traceback on node online +- Fix: ui_node: prevent traceback on node standby +- Fix: add diagnostic messages when crm report cannot find ssh-agent (bsc#1219538) +- Dev: use ClusterShell for the changes in the previous commit +- Fix: crm report hangs in the root passwordless with ssh-agent +- Fix: report: use ClusterShell for ssh (bsc#1220170) +- Fix: ui_node: clearstate command needs adjustment (bsc#1219831) +- Fix: bootstrap: fail to join a cluster with hostname alias +- Fix: bootstrap: clear stall data about ssh users left possiblely from +previous setups (bsc#1219476) +- Fix: report: Escape special characters in pattern (bsc#1220022) +- Dev: ui_configure: verify all cib objects even if there is no primitive +resource configured +- Dev: ui_configure: bundle: support podman container +- Dev: constants: Update the bundle help text +- Dev: parse: Assign meaningful id to the bundle's port-mapping and storage-mapping +- Dev: cibconfig: Hide the id of bundle port-mapping and storage-mapping +- Dev: cibconfig: Change the multiline format for bundle +- Dev: utils: Return empty list if corosync.conf does not exist +- Dev: report: Enable crm report even cib.xml does not exist +- Dev: utils: Query pacemaker_remote node correctly +- Fix: ui_cluster: Can't start cluster with --all option if no cib (bsc#1219052) +- Fix: bootstrap: ssh key of the init node is duplicated in the +authorized_keys files of other node (bsc#1218940) +- Fix: sh: guide users to setup key-based ssh authentication when +non-interactive authentcation fails (bsc#1219045) +- Fix: parse: Automatically append 's' as default time unit for timeout and interval (#1304) +- Dev: ui_configure: Update the operation id after the monitor interval changed + * Tue Jan 9 2024 Xin Liang - Release 4.6.0 - Fix: report: Unable to gather log files that are in the syslog format (bsc#1218491)