* Updated Database->resync_databases() to never run on non-striker machines. On Strikers, before a resync, _age_out_data() is called to clear old data in long-off databases.
* Created System->check_memory() that is loosely based on anvil-check-memory, but checks to see if it's being controlled by a systemctl started daemon and, if so, reads the RAM in use from it's status output.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated the logic of when to boot a node or DR host that was found to be off for unknown reasons to require both poewr and temperature to be OK, and checks against the new 'feature::scancore::disable::boot-unknown-stop' config variable.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created Network->is_our_interface() which returns '1' if an interface is one managed by an Anvil!. Also updated scan-network to use this to determine when an interface alert should be a warning or notice level alert.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created ScanCore->agent_shutdown() that writes out the time the scan agent last ran, and how many databases were available when it last ran.
* Updated ScanCore->agent_startup() to read the the last run data created above.
* Updated Database->connect() to set 'sys::database::last_db_count' to the scan agent's recorded last DB count.
* Updated all agents to call ScanCore->agent_shutdown() at the end of their run.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Fixed a bug where anvil-safe-stop was not recording the stop-reason. Also made '--poweroff' an alias for '--power-off'.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated Jobs->get_job_uuid() to accept the new 'incomplete' parameter that, when set, will look for jobs whose progress is > 1 and < 100.
* Updated ScanCore-agent_startup() to take the new 'no_db_ok' parameter which returns with '0' if no DB is available and that parameter is set to '1'.
* Fixed a logging bug in 'anvil-join-anvil'.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated DRBD->allow_two_primaries() to take the 'set_to' parameter which can be 'yes' to all and 'no' to disallow dual-primary.
* Updated ocf:alteeve:server to call allow_two_primaries() with 'set_to' = 'no' instead of calling 'adjust' after a migration completes.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated scan-cluster to get the CIB from pcs instead of reading the CIB from disk.
* Updated anvil-daemon to always call striker-prep-database at log level 2 while trying to find the cause of rare postgres config failures. Also updated striker-prep-database to use the new method of initializing the DB.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Bumped scancore's scan delay from 30 seconds to 60.
* Shorted the age-out time to 24 hours and again boosted the archive thresholds. As we get a feel for the amount of data collected on multi-Anvil! systems over time, we may continue to tune this.l
* Moved Database->archive_database() to be called daily by anvil-daemon, instead of during '->connect' calls.
* Added locking to Database->_age_out_data to avoid resyncs mid-purge. Also moved the power, temperature and ip_address columns into the same 'to_clean' hash as it was duplicate logic.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created Database->_age_out_data() to delete records from the database that are old enough to no longer be useful. This is designed to significantly reduce the size of the database, allowing a better focus on performance.
* Changed Database->connect() to default to NOT check for resync, reworking the old 'no_resync' to 'check_for_resync', so that resync checks happen on demard, instead of by default.
* Updated get_tables_from_schema() to now allow 'schema_file' to be set to 'all', which then loads the schema files of all scan agents as well as the core anvil schema file. Fixed a bug where commented out tables were being counted.
* Re-enabled triggering resyncs on 'last_updated' differences.
* Fixed a bug in scan-ipmitool where the history_id column in history.scan_ipmitool_value was incorrect.
* Created a new tool called striker-show-db-counts that shows the number of records in all public and history schema tables for all databases.
* Updated anvil-update-states to detect when a libvirtd NAT'ed bridge exists and to delete it when found.
Signed-off-by: Digimer <digimer@alteeve.ca>
* DRBD is now configured to a ping-timeout of 3 seconds.
* Created Log->switches() that returnes the command line switches used by Anvil! tool command line calls based on the active log levels / secure logging. Appended this to all invocations of our tools.
* Updated Database->resync_databases() to now only skip 'jobs' and 'variables' tables with less than 10 record differences. All other differences will trigger a resync.
* Created System->_check_anvil_conf() that, as you might guess, checks in anvil.conf exists and created it (using defaults), if not. It also checks to see if the 'admin' group and user exists and creates them, if not.
* Updated anvil-daemon to check anvil.conf on start up and in each loop. Created the function check_journald() that checks (and sets, if needed) that journald logging is persistent.
* Made striker-manage-peers to check_if_configured on the Database->connect() when updating anvil.conf and the target UUID is the local machine. Also created a loop to make the reconnection a lot more robust.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Changed the alteeve repo RPM to the new cimmunity/enterprise repo
* Fixed a bug where 'fence_data::updated' was causing the fences web page to break.
* Fixed a bug in Database->insert_or_update_network_interfaces() where certain interfaces were being repeatedly added to the database.
* Fixed a bug in Database->_find_behind_databases() was marking DBs as behind even though they had less than 10 columns off.
* Fixed a bug in Get->host_name() where, if the host name was changed on disk but the environment variable was still the old name, it would cause the hostname to waffle back and forth and cause constant updated to /etc/hosts.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated the scan agent timeout to 60 seconds. Also made the scan agent exit code log entries more helpful.
* Updated System->collect_ipmi_data() to now better handle duplicate sensor names. Now, instead of simply appending an integer, we find the hex address and use that in the sensor name when duplicates exist. This solves the problem of the sensor names not being consistently shown in order.
* Fixed message bugs (bad variable insertions) in scan-apc-pdu and scan-apc-ups.
* Fixed schema procedure bugs in the 'temperature' and 'ip_address' tables where the columns were in bad order, causing constanty updates.
Incomplete work;
* Create the shell of 'anvil-manage-storage', but virtually no logic exists in it yet.
* Started work on anvil-safe-start to deal with an issue where DRBD resources don't start when a server is running on a peer.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Fixed a bug in Database->get_tables_from_schema where history.X and X tables were being stored in the table list.
* Updated ocf:alteeve:server to no do resyncs on DB connect.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Added a new parameter to Database->connect() called 'no_resync' that, if set, prevents a resync check being performed. Updated ->resync_databases() to find a uuid_column where the table name ends in 'ies' and the UUID column is 'y_uuid'. Updated ->resync_databases() to not fire on updated table age anymore, and to trigger only if the number of rows differ in a given table by more than 10.
* Updated Log->entry() to prefix a tool's name, when the new 'log::scan_agent' value is set. Also set this value in ScanCore->agent_startup(), to help differentiate log entries.
* Fixed a bug in scancore's main loop where it logged the sleep message at the start of the run.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created Cluster->get_fence_methods() that parses all fence methods out of a recorded CIB and stores the in a hash for a given host_uuid.
* Fixed a bug in ScanCore->post_scan_analysis_striker() where the short_host_name was not being stored correctly.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created DRBD->check_if_syncsource() and ->check_if_synctarget() that return '1' if the target host is currently SyncSource or SyncTarget for any resource, respectively.
* Updated DRBD->update_global_common() to return the unified-format diff if any changes were made to global-common.conf.
* Created ScanCore->check_health() that returns the health score for a host. Created ->count_servers() that returns the number of servers on a host, how much RAM is used by those servers and, if available, the estimated migration time of the servers. Updated ->check_temperature() to set/clear/return the time that a host has been in a warning or critical temperature state.
* Finished ScanCore->post_scan_analysis_node()!!! It certainly has bugs, and much testing is needed, but the logic is all in place! Oh what a slog that was... It should be far more intelligent than M2 though, once flushed out and tested.
* Created Server->active_migrations() that returns '1' if any servers are in a migration on an Anvil! system. Updated ->migrate_virsh() to record how long a migration took in the "server::migration_duration" variable, which is averaged by ScanCore->count_servers() to estimate migration times.
* Updated scan-drbd to check/update the global-common.conf file's config at the end of a scan.
* Updated ScanCore itself to not scan when in maintenance mode. Also updated it to call 'anvil-safe-start' when ScanCore starts, so long as it is within ten minutes of the host booting.
Signed-off-by: Digimer <digimer@alteeve.ca>
* The get_mmeory endpoint was completed.
* The get_replicated_storage endpoint was completed, though it requires testing and likely has issues.
To prepare for the get_status endpoint work, I needed to update ScanCore and modules to track the host_status. This commit contains the work needed for this.
* Updated ScanCore->post_scan_analysis_striker() to use configured fence devices (except PDUs) to check if a target host is off or on, in there is no host_ipmi interface. In all cases, if a machine can be confirmed on or off, the host_status is now updated.
* To support the above fence based power checks, updated scan-cluster to store the on-disk CIB in the new scan_cluster -> scan_cluster_cib colume.
* Updated ScanCore->parse_cib() to map stonith primitive IDs to fence agents. Updated ->parse_crm_mon() to not call if the executable doesn't exist to avoid unhelpful error messages in the logs when called from a Striker.
* Update DRBD->gather_data() to get the size data from /sys/block/drbd<minor>/size' x '/sys/block/drbd<minor>/queue/logical_block_size so it works when a device is Secondary (and can't be promoted).
* Updated Database->get_hosts_info() to record the short host name as well as the stored host name. Created ->update_host_status() as a wrapper to ->insert_or_update_hosts() that only updates the host status.
* Updated anvil-join-anvil to disabled ksm and ksmtuned daemons.
* Updated scancore and anvil-daemon to set the host_status to 'online' on startup.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Fixed a bug in ScanCore->agent_startup() where a (thankfully broken) check to append tables to the 'sys::database::check_tables' would cause an infinite loop as both were pointers to the same anonymous array.
* Fixed a bug in scan-ipmitool where the scan_ipmitool_variables table didn't use a host_uuid reference, causing resyncs of that table to sync for all hosts and cause DB errors when the scan_ipmitool record from another host wasn't sync'ed yet.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated Get->os_type() to use 'cat' instead of Storage->read_file() because 'rsync' may not be available when it is called during striker-initialize-host calls.
* Updated Database methods to skip 'oui' and 'state' during resync.
* Updatedb striker-initialize-host to detect when it's initializing a CentOS Stream Node / DR Host and enable the HA repo.
* Created the tools/striker-auto-initialize-all tool, which is very much incomplete, that will allow for the rapid creation of a full Anvil! from freshly installed machines autonomously.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Cleaned up a log of logging to reduce the amount of log entries when running at log level 1.
* Bumped the scan-ipmitool default 'jump' range to 10c.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Started work on Get->available_resources() that will take an 'anvil_uuid' and figure out what resources are still available for use by new servers or that can be added to existing servers.
* Fixed a bug in ScanCore->agent_startup() where tables weren't being generated properly from the agent's SQL file.
* Made Storage->change_mode() return silently if it's called without a mode being passed. This happens frequently and is harmless so it's not worth filling the logs with errors.
* Renamed the 'start_time' key to 'at_start' when recording files' MD5 sums in Storage->record_md5sums and ->check_md5sums.
* When we moved the directory scan logic out of the 'scancore' daemon and into 'Storage->scan_directory', the logic to record scan agent names in 'scancore::agent::<file>' was removed. This broke a few things and, so, it was restored when it was found that a file starts with 'scan-' and the directory matches the scancore agent directory.
* Moved the 'scancore' daemon's 'load_agent_strings' to 'Words'
* Updated Words->parse_banged_string() to look for variables in the format 'value=X:units=Y' and translate it properly.
* Fixed a bug in scan-ipmitool where discovered sensor INSERT SQL queries were queued, but not committed.
* Fixed a bug in scan-storcli where a while loop was broken, preventing execution.
* Fixed a bug in the 'scancore' daemon where it wouldn't exit if sums changed. Fixed a bug where alerts weren't being sent between loops. Fixed a bug where command-line log level wasn't surviving inside the main loop.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created Database->get_power() that loads data from the special 'power' table.
* Fixed a bug in calls to Network->ping() where some weren't formatted properly for receiving two string variables.
* Updated Database->get_anvils() to record the machine types when recording host information.
* Updated Database->get_hosts_info() to also load the 'host_ipmi' column.
* Updated Database->get_upses() to store the link to the 'power' -> 'power_uuid', when available.
* Created ScanCore->call_scan_agents() that does the work of actually calling scan agents, moving the logic out from the scancore daemon.
* Created ScanCore->check_power() that takes a host and the anvil it is in and returns if it's on batteries or not. If it is, the time on batteries and estimate hold-up time is returned. If not, the highest charge percentage is returned.
* Created ScanCore->post_scan_analysis() that is a wrapper for calling the new ->post_scan_analysis_dr(), ->post_scan_analysis_node() and ->post_scan_analysis_striker(). Of which, _dr and _node are still empty, but _striker is complete.
** ->post_scan_analysis_striker() is complete. It now boots a node after a power loss if the UPSes powering it are OK (at least one has mains power, and the main-powered UPS(es) have reached the minimum charge percentage). If it's thermal, IPMI is called and so long as at least one thermal sensor is found and it/they are all OK, it is booted. For now, M2's thermal reboot delay logic hasn't been replicated, as it added a lot of complexity and didn't prove practically useful.
* Created System->collect_ipmi_data() and moved 'scan_ipmitool's ipmitool call and parse into that method. This was done to allow ScanCore->post_scan_analysis_striker() to also call IPMI on a remote machine during thermal down events without reimplementing the logic.
* Updated scan-ipmitool to only record temperature data for data collected locally. Also renamed 'machine' variables and hash keys to 'host_name' to clarify what is being stored.
* Updated scancore to clear the 'system::stop_reason' variable.
* Added missing packages to striker-manage-install-target.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Fixed a bug in Database->check_agent_data() where the list of tables wasn't passed in, and thus the table list wasn't then passed on to Database->_find_behind_databases().
* Started work on a new method called Storage->parse_lsblk().
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated ScanCore->agent_startup() to use Database->get_tables_from_schema() when the 'tables' parameter isn't passed.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Created (but not finished) scan-apc-pdu
* Added support to tracking maintenance-mode for nodes in Cluster->parse_cib
* Created Remote->read_snmp_oid().
* Created Server->get_definition.
* Updated Server->get_status() to write-out server XML files on-demand.
* Finished scan-cluster.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated the servers table to remove the 'not null' constraints on the server_start_after_server_uuid, server_pre_migration_file_uuid and server_post_migration_file_uuid columns.
* Updated ScanCore->agent_startup() to connect to the database(s) when there isn't a table list.
* Updated Server->migrate_virsh() to test for DB access before making DB calls (to allow ocf:alteeve:server to function even if all ScanCore DBs are offline).
* Updated ocf:alteeve:server to connect to the databases (though work without it), and changed '$FILE_NAME' to be 'ocf:alteeve:server' (to make logging more legible)
* Created the skeleton for 'scan-storage'.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Renamed the 'defitintions' table to 'server_definitions' to clarify the purpose, and made all the 'server' columns have then 'not null' constraint.
* Created Database->insert_or_update_servers(), ->get_servers(), ->insert_or_update_server_definitions() and ->get_server_definitions().
* Updated scancore, anvil-daemon, and scan agents to not run unless they're run with root privs.
* Got scan-server to update the servers / server_definition tables and the on-disk file when needed.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Added support in anvil.conf to disable scan agents with 'scancore::<agent_name>::disable', and added handling this to agents. Also allowed for '--force' to override this setting.
* Updated ScanCore->agent_startup() to allow for empty scan agent table lists.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Started work on the new scan-cluster scan agent that will parse out and store data from the pacemaker CIB.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated Alert->check_alert_sent() and the 'alert_sent' table to remove 'alert_name' as it really isn't helpful given record_locator.
* Created 'Database->purge_data' that takes an array of tables and, in reverse, purges them from the database(s). This also disables archiving and resync functions now.
Signed-off-by: Digimer <digimer@alteeve.ca>
* Updated Alert->register to take a hash reference for message variables to simplify when a caller plans to log and register an alert at the same time.
* Updated Convert->bytes_to_human_readable() to name the 'size' variable used internally for 'bytes' to actually be 'bytes' for better consistency.
* Created multiple new Database methods;
** ->check_condition_age() is meant to be used by scan agents to see how long a given condition has been in play (ie: how long ago power was lost to a UPS or a sensor became unreadable).
** ->insert_or_update_health() handles recording data to the new 'health' table, used for determining ideal hosts for servers between nodes.
** ->insert_or_update_power() handles recording data to the new 'power' table, used for determining how power events are handled.
** ->insert_or_update_temperature() handles recording temperature data to the new 'temperature' table, used to determine how thermal events are handled.
* Got a lot more done on the scan-hardware scan agent. Only part left now is post-scan health processing.
Signed-off-by: Digimer <digimer@alteeve.ca>
Digimer
Fabio M. Di Nitto