CHANGELOG.md

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Unreleased

Added

• disk.py: get_block_devices() lists all local block devices, including ones without a mounted filesystem, so callers can also work with raw or unmounted devices such as multipath SAN volumes
• lftest.py: network() plus network / network_alias arguments on run_container() let a test wire an application container to a backing service (e.g. a database) on a shared network, for multi-container integration tests
• url.py: fetch() / fetch_json() accept a method argument to force the HTTP method, enabling a bodyless POST

Fixed

• redfish.py: Sensors that report an empty min/max range (identical min and max, used by some firmware as a "no limit" placeholder) no longer raise false warnings (#1211)
• url.py: a caller-supplied Content-Length is ignored and recomputed from the request body
• veeam.py: authentication against the Veeam Enterprise Manager API works again
• Installing the library from source (for example pip install --editable .) no longer hangs, which also unblocks the API documentation build
• Resolve the remaining ruff lint violations across the library, including a few robustness fixes: a bare except in disk.py now catches only OSError, mutable default arguments in url.py and rocket.py are no longer shared between calls, and uptimerobot.py uses isinstance() instead of a type() comparison (#118)

v4.2.0 - 2026-06-02

Added

• db_sqlite.py: get_db_path() resolves and returns the absolute path of a database without opening it, so callers that need to seed, migrate or remove a database file rely on a single source of truth instead of rebuilding the path themselves

Security

• db_sqlite.py: SQLite databases are now created in a private, per-user 0700 directory under the system temporary directory instead of directly in the shared, world-writable /tmp. This closes a local symlink attack on the predictable database paths where an unprivileged user could redirect writes from a process running as root to arbitrary files (GHSA-r35r-fpx2-jgr4, thanks to OoYo0uto)

v4.1.0 - 2026-05-29

Added

• db_mysql.py: get_server_info() returns {flavor, version} for the installed MySQL/MariaDB server, probing mysqld, mariadbd, mariadb, mysql and parsing the --version banner with regexes that cover all known output formats (legacy Distrib banner, modern from banner, server-binary banner, MariaDB-suffixed and plain). Accepts a pre-collected banner string to support unit-test fixtures without shelling out. Works without a database connection
• db_mysql.py: get_flavor() returns whether the installed binary is MariaDB or MySQL (thin wrapper around get_server_info()). Useful where systemd-based detection is unreliable (e.g. Fedora aliases mysql.service to mariadb.service)

Fixed

• db_sqlite.py: plugins that cache trend data in /tmp no longer fail with "attempt to write a readonly database" when first run as one user (e.g. root) and later scheduled under another. Each user now gets its own cache file (#181)

Security

• Bump idna to 3.16 in the Python 3.9 lockfile, closing a moderate vulnerability where crafted input to idna.encode() could bypass the CVE-2024-3651 fix

v4.0.2 - 2026-05-18

Fixed

• url.py: import lib.url no longer aborts with AttributeError: module 'ssl' has no attribute 'TLSVersion' on Python interpreters below 3.7 (e.g. RHEL 8's default python3 = 3.6). The TLS version mapping is now built only when ssl.TLSVersion is available; any plugin that doesn't use TLS version pinning keeps importing. Callers that pass tls_min / tls_max get a clear RuntimeError naming the missing requirement instead of the cryptic AttributeError. Note: the lib's supported minimum is still Python 3.9; this only makes one specific import path resilient

v4.0.1 - 2026-05-18

Fixed

• db_mysql.py: connect() aligns the session's character set and collation with the mysql system schema right after the connection is up. Stops queries against mysql.user and mysql.global_priv from aborting with ER 1267 ("Illegal mix of collations") when the server's connection-collation default differs from the system tables' column collations. MariaDB 10.4+ exposes mysql.user as a view over mysql.global_priv with JSON-derived columns whose results carry COERCIBLE coercibility, which makes plain col = 'literal' compares trip whenever collation_connection and the column collation don't match. Lookup uses information_schema.schemata so the right collation is picked across MySQL/MariaDB versions and locale-specific installs. Best-effort: on any error the connection stays usable. Fixes all mysql-* plugins (linuxfabrik/monitoring-plugins#1139)

v4.0.0 - 2026-05-15

Breaking Changes

• base.py: rename module-level constant X86_64 to IS_64BIT. The old name was misleading because it suggested Intel/AMD 64-bit only, while the underlying check (sys.maxsize > 2**32) is True on any 64-bit Python build (aarch64, ppc64le, s390x, riscv64, …). Logic unchanged. Downstream consumers must update imports

Added

• db_mysql.py: check_privileges(conn, *required) replaces the old check_select_privileges(). Without arguments it keeps the previous functional smoke test (SELECT VERSION(), works with GRANT USAGE alone). With arguments it parses SHOW GRANTS FOR CURRENT_USER() and reports any privilege that is missing for the current user, with ALL PRIVILEGES and SUPER short-circuiting to success. Each positional argument can be a string (single privilege required) or a list/tuple of strings (any-of group, useful for cross-version aliases like REPLICATION CLIENT / SLAVE MONITOR / REPLICA MONITOR introduced by MariaDB 10.5+). Enables plugins to declare their actual privilege requirements precisely
• db_mysql.py: four new helpers consolidate patterns that several check plugins implement by hand. get_all_status(conn) and get_all_variables(conn) return the complete SHOW GLOBAL STATUS / SHOW GLOBAL VARIABLES as a dict in one round trip (cheaper than many LIKE '...' queries when a plugin needs more than a handful of values). get_replica_status(conn) issues SHOW REPLICA STATUS and silently falls back to the legacy SHOW SLAVE STATUS, returning the first row or None when the server is not configured as a replica. has_is_role_column(conn) reports whether mysql.user.IS_ROLE exists (MariaDB 10.0.5+ roles) so plugins can gate IS_ROLE = 'N' clauses without each implementing the same probe
• db_sqlite.py: per_second_deltas(filename, name, counters) consolidates the cross-run "delta of cumulative counters between runs" pattern that several check plugins implement by hand. Persists the counters in a local SQLite cache (schema derived from the counters dict so callers no longer have to spell out the table definition per plugin) and returns the per-second rates vs. the previous run. Useful for any cumulative counter that needs to be reported as a per-second rate: /proc and /sys byte counters (disk I/O, network traffic, file descriptors), database status counters, application metrics. Returns None on the first run, on counter resets and on schema changes (auto-rebuilds the cache table in that case). Lets plugins emit per-second rates as perfdata instead of uom='c' continuous counters, so Grafana panels do not need their own non_negative_difference() workaround
• lftest.py: run_mariadb_from_containerfile(containerfile_path, ...) builds a per-plugin Containerfile via testcontainers' DockerImage and yields (container, defaults_file) like run_mariadb(). Plugins move their test-image matrix into unit-test/containerfiles/<name> so each plugin owns its supported MariaDB / MySQL LTS coverage instead of relying on a hardcoded image list in the lib. The MARIADB_LTS_IMAGES constant has been removed (no in-tree caller after the migration)
• lftest.py: run_mariadb / run_mariadb_from_containerfile renamed to run_mysql_compatible / run_mysql_compatible_from_containerfile to reflect that both upstream MySQL and MariaDB images are supported (docker.io/library/mysql:*, docker.io/library/mariadb:*, quay.io/sclorg/mariadb-*, quay.io/sclorg/mysql-*). Family detection picks mariadbd vs mysqld when extra_args is set. Old names are kept as aliases for one release
• net.py: fetch() and fetch_socket() gain an optional dialog parameter for multi-step request/response conversations (regex-driven, no half-close). Enables clean implementations of plugins for protocols like NUT, SMTP, POP3, IMAP and FTP without re-implementing socket handling per plugin
• net.py: fetch() gains a tls=True switch that wraps the socket in a TLS 1.2+ context with SNI, equivalent to calling fetch_ssl(). The legacy fetch_ssl() helper stays for backward compatibility but is now marked deprecated in its docstring
• time.py: now() gains as_type='utc' returning the current UTC time as a naive datetime.datetime. Useful for fields defined as UTC by spec (x509 notBefore / notAfter, HTTP Date, RFC 3339 timestamps), so consumers can stay on the lib helper instead of falling back to datetime.datetime.now(datetime.timezone.utc)
• url.py: fetch() and fetch_json() now speak HTTP/1.0, HTTP/1.1 and HTTP/2 via httpx, with new http_version, tls_min and tls_max parameters for protocol pinning. extended=True now also returns the negotiated TLS version, ALPN protocol, the server certificate in DER form, and a per-phase timings dict with dns, connect, tls, ttfb, transfer and total (all seconds), ready for downstream certificate inspection and HTTPS-availability monitoring. http_version='3' is reserved and currently returns a clear error until QUIC support lands. Existing callers and parameters are unchanged

Changed

• pyproject.toml: declare pypsrp and pywinrm as direct dependencies. lib.winrm imports both at module load time (wrapped in try/except ImportError only so a non-WinRM consumer still loads); previously they had to be pinned in every downstream project that consumed lib.winrm (e.g. monitoring-plugins for dhcp-scope-usage on Windows). Same convention as psutil, smbprotocol etc., which lib.psutil / lib.smb import conditionally but which are declared as hard deps because they are part of lib's published surface
• requirements: one hash-pinned lockfile per supported Python LTS, each in its own lockfiles/pyXX/ subdirectory (py39 to py314). Replaces the single requirements.txt. Dependabot watches each subdirectory separately, except lockfiles/py39/ which is excluded from both version bumps and security PRs: most upstream packages dropped Python 3.9 over 2025/2026, so automated bumps would break pip install --require-hashes on RHEL 8 / Debian 11. The py39 lockfile is regenerated manually as needed
• url.py: fetch() switched its underlying engine from stdlib urllib to httpx. Behaviour for existing callers is preserved (parameters, return-tuple shape, redirect-following, default Connection: close and User-Agent headers, automatic application/x-www-form-urlencoded for POST without explicit Content-Type). response_header in the extended dict is now a plain dict instead of http.client.HTTPMessage; existing consumers only relied on .get() access

Deprecated

• db_mysql.py: check_select_privileges() is deprecated and now a thin backwards-compatible shim that delegates to check_privileges(conn). New code should call check_privileges() directly. The shim exists to keep already-deployed plugins working when the lib is upgraded ahead of the plugin set; it will be removed in a future major release

Fixed

• base.py: oao() now properly HTML-escapes &, < and > into &, < and > instead of replacing < and > with apostrophes. The previous implementation destroyed legitimate threshold descriptions like <= 10 and shell snippets like echo 1 > /proc/sys/..., turning them into '= 10 and echo 1 ' /proc/sys/... in plugin output. HTML-based web UIs (Icinga Web, Naemon-Adagios) render the entities back to the original characters; terminal viewers see the literal entities, which preserves the information. The XSS-protection goal of the original change is still met
• db_sqlite.py: per_second_deltas() no longer crashes with sqlite3.ProgrammingError: Cannot operate on a closed database when the cache table schema mismatches an earlier plugin version. The internal insert() call now uses delete_db_on_operational_error=False so the connection survives the schema-mismatch error; the explicit drop/recreate path below operates on a live connection as intended
• url.py: fetch() with HTTP digest authentication and insecure=True now actually disables certificate verification. Previously the digest auth path silently lost the SSL context
• url.py: fetch() with no_proxy=True now applies the timeout parameter. Previously the no-proxy path called opener.open(request) without a timeout, so hangs were only caught by the outer plugin wrapper
• url.py: import lib.url no longer fails on hosts where httpx is not installed; the import is now lazy and fetch() returns a clear "httpx not installed" error message instead of crashing on import. Plugins that pull lib.url only transitively (for example via lib.net) keep working without httpx. Hosts that actually use fetch() still need httpx[http2] installed via pip or dnf install python3-httpx python3-h2

v3.4.1 - 2026-05-07

Fixed

• librenms.py: get_state() now also maps the LibreNMS alert states WORSE (3), BETTER (4) and CHANGED (5) to WARN/CRIT. Previously only ACTIVE (1) was treated as an alerting state, so open alerts in any of those three states were silently reported as OK. WORSE and BETTER exist in LibreNMS since 1.54 (July 2019); CHANGED was added in LibreNMS 25.2.0 (February 2025) and is now triggered whenever the alert diff detects a change

Security

• ci: Scope GITHUB_TOKEN permissions in the dependabot-auto-merge workflow to the job level, with top-level now read-all. Matches the pattern used by the other Linuxfabrik workflows and addresses the OpenSSF Scorecard Token-Permissions finding.

v3.4.0 - 2026-04-22

Added

• time.py: add macro2timestr(s, format='') to expand time macros in a string. Supports {today}, {yesterday} (rendered with the given format, default ISO 8601) and single strftime components ({%Y}, {%y}, {%m}, {%d}, {%H}, {%M}, {%S}). Unknown {...} tokens pass through unchanged

v3.3.0 - 2026-04-19

Added

• args.py: add a generic --check-security help text so version-style plugins can offer an upstream security-update check with a uniform parameter description

v3.2.0 - 2026-04-14

Added

• url.py: add split_basic_auth(url) helper that extracts userinfo from a URL like https://user:secret@host/path, returns the URL with the userinfo stripped from the netloc, plus a headers dict carrying the matching Authorization: Basic ... entry. Pass both into lib.url.fetch() / lib.url.fetch_json(). This lets apps accept HTTP basic auth via the URL itself instead of exposing separate --username / --password arguments, which keeps the credentials out of ps listings, out of the request line, and out of any proxy access log

v3.1.1 - 2026-04-14

Changed

• human.py: human2seconds() and humanduration2seconds() now accept the Unix-style lowercase day/week markers d and w in addition to the canonical Linuxfabrik uppercase D and W. This lets callers parse duration strings from third-party tools that follow the Unix convention (exim mailq age literals, sleep 3d, systemd timers, etc.) without having to normalize the input first. Uppercase D/W continue to work exactly as before, so no existing caller breaks
• nextcloud.py: run_occ() no longer relies on the Nextcloud occ script being marked executable. It now locates php via shutil.which('php') and invokes sudo -u \#<uid> php <occ> <cmd>, which also works on installations where occ lacks the execute bit or its shebang does not resolve to a working PHP interpreter. If no php is found in PATH, the call returns a descriptive error instead of silently failing

Security

• Harden the CI supply chain: the pre-commit install in the pre-commit-autoupdate workflow is now hash-pinned via .github/pre-commit/requirements.txt (generated with pip-compile --generate-hashes --strip-extras), and dependabot/fetch-metadata is pinned to a commit SHA so all GitHub Actions used in .github/workflows/ are now pinned by hash. The policy is documented in CONTRIBUTING.md under "CI Supply Chain"

v3.1.0 - 2026-04-13

Added

• disk.py: add dir_exists() as the directory-only counterpart to file_exists(). The existing file_exists() wraps os.path.isfile() and therefore returns False for directories, which is easy to miss; callers that want to check for a directory should now use dir_exists()
• lftest.py: add attach_each() helper for iterating over arbitrary lists (e.g. container image matrices, file-based fixtures) with a caller-supplied action, complementing attach_tests() which only works on the TESTS dict shape
• lftest.py: add attach_tests() helper that attaches one test_* method per entry in a plugin's TESTS list, so that test discovery and reporting show the actual number of fixtures instead of a single aggregate test
• lftest.py: add run_mariadb() context manager and MARIADB_LTS_IMAGES constant for container-based MariaDB integration tests. Starts a sclorg or upstream MariaDB container, waits for the TCP listener, yields a temporary client option file, and cleans up on exit. MARIADB_LTS_IMAGES lists the currently supported MariaDB LTS releases (10.6, 10.11, 11.4, 11.8) so the mysql-* monitoring plugins can iterate over a single canonical matrix

v3.0.0 - 2026-04-13

Added

• Add GitHub Actions workflow to automatically build and deploy API documentation to GitHub Pages
• Add bandit and vulture to .pre-commit-config.yaml for security and dead-code checks on every commit
• Add pre-commit hooks
• Add ruff linter and formatter to pre-commit hooks (#117)
• args.py: expand HELP_TEXTS with standard help texts for all common parameters (always-ok, critical, warning, insecure, no-proxy, timeout, url, ignore-regex, match, lengthy, count, test, etc.)
• disk.py: add get_owner()
• lftest.py: add run() function for declarative, data-driven unit tests using subTest()
• nextcloud.py: new library
• txt.py: add exception2text()
• winrm.py: add WINRM_CONFIGURATION_NAME option to run_ps()
• winrm.py: execute ps in run_ps() directly without Invoke-Expression wrapping

Changed

• base.py: get_perfdata() now sanitizes labels by stripping single quotes and replacing = with _
• base.py: get_perfdata() output no longer has trailing semicolons
• base.py: get_table(): document why pure ASCII delimiters are used instead of Unicode box-drawing characters
• base.py: get_worst() now accepts any number of state arguments (*states). Existing two-argument callers keep working unchanged, but plugins that need to combine three or more states in one call no longer have to nest the call - e.g. get_worst(state, used_state, committed_state) instead of get_worst(state, get_worst(used_state, committed_state))
• base.py: deduplicate get_state() operator logic using operator module
• base.py: deduplicate sum_dict() by delegating to sum_lod()
• base.py: improve get_table() performance for large tables
• base.py: move parse_range() and state name mapping to module level
• base.py: remove unused collections import
• db_sqlite.py: reduce unnecessary dictionary object creation
• human.py: deduplicate bits2human()/bps2human()/bytes2human() via shared _to_human() helper
• human.py: deduplicate humanrange2bytes()/humanrange2seconds() via shared _convert_range() helper
• human.py: pre-compute mappings as module constants
• lftest.py: test() now accepts args with fewer than three elements. Plugins can be invoked as --test=path/to/fixture without the trailing ,,0; stderr defaults to the empty string and the return code to 0
• powershell.py: run_ps() now always returns a dict
• Remove pre-built documentation from the repository (now auto-deployed via GitHub Actions)
• txt.py: improve filter_mltext() performance (avoid O(n²) string concatenation)
• txt.py: improve readability of extract_str() fallback logic
• txt.py: remove unused Python 2 type aliases and outdated comments
• txt.py: simplify to_text() and to_bytes() for Python 3 only (remove dead Python 2 codepaths)
• winrm.py: make run_cmd() and run_ps() JEA-aware
• winrm.py: make run_cmd() and run_ps() Kerberos-aware

Removed

• Drop support for Python older than 3.9. The lib now requires Python 3.9 or newer; this matches the oldest still-supported enterprise Linux (RHEL 8) and lets the codebase use modern syntax and standard-library features

Fixed

• base.py: cu() now also escapes HTML characters in the error message, not just in the traceback
• base.py: cu() now detects active exceptions via sys.exc_info() instead of string-matching the traceback
• base.py: get_state() no longer calls sys.exit() on malformed range specs, returns UNKNOWN instead
• base.py: get_table() no longer uses the wrong separator for the second data row when called without a header
• base.py: oao() now escapes HTML characters in the output message to prevent injection in web UIs
• base.py: fix invalid -10 range example in _parse_range() docstring (correct syntax is -10:0)
• cache.py: treat a cache entry as valid up to and including its expire timestamp instead of expiring it one second early (< instead of <=). A key set with expire=now+5 is now still served at now+5 and first becomes unavailable at now+6, matching HTTP Cache-Control max-age and Redis EXPIRE semantics. Callers that relied on the old one-second-early expiry see their cached value live one second longer (#120)
• db_sqlite.py: pass usedforsecurity=False to hashlib.sha1() so bandit no longer flags a non-security SHA1 use as a weak hash (the hash is only used to derive sanitized SQL identifiers)
• db_sqlite.py: rename unused loop variable in rm_db() to silence ruff B007
• Fix --require-hashes pip installs in CI workflows by using pinned versions instead
• grassfish.py: remove unused match() helper that referenced undefined names (re and compiled_custom_id_regex); the function was never called and would have raised NameError at runtime
• human.py: bits2human(), bytes2human() and bps2human() now scale negative values to a unit that matches their magnitude. Before, bytes2human(-1048576) returned -1048576.0B, now it returns -1.0MiB. This matters for counter deltas that can legitimately be negative (counter resets, reclaimed memory, bandwidth drops) (#120)
• human.py: fix incorrect seconds2human() docstring example for sub-second values
• net.py: fix get_netinfo() which called a non-existent get_ip_public() and swallowed the resulting NameError by returning []; the function now leaves public_address as None and callers that need the public IP must use get_public_ip() directly
• powershell.py: fix outdated shebang line
• rocket.py: get_groups_history() no longer mutates a shared default params={} dict (B006) and properly defaults params to None
• shell.py: shell_exec() now applies timeout to the shell=True path (was previously ignored)
• shell.py: close the upstream process's stdout after connecting it to the next pipeline stage. Without this, the upstream process never received EOF/SIGPIPE when the downstream stage exited early, and each pipeline stage leaked a file descriptor until garbage collection caught up (#120)
• txt.py: sanitize_sensitive_data() now also redacts JSON-style fields and HTTP Authorization headers
• txt.py: fix exception2text() missing traceback import (fallback path was dead code)
• txt.py: fix pluralize() not stripping whitespace from comma-separated suffix parts
• txt.py: fix sanitize_sensitive_data() replacing the key name instead of the secret value
• url.py: fetch() and fetch_json() no longer use mutable default arguments for header and data (B006); defaults are now None with initialization inside the function
• url.py: drop the dead timeout=timeout if digest_auth_user else timeout ternary in fetch(); both branches evaluated to timeout, so behavior is unchanged (#120)
• winrm.py: pass parameters correctly in run_cmd() when using pypsrp

Security

• Annotate all remaining bandit low/medium findings with # nosec BXXX comments and a short justification (subprocess helpers with shell=True by design, admin-controlled URLs passed to urlopen, SQL built from sanitized identifiers in db_sqlite). Bandit now runs clean at --severity-level=low --confidence-level=low over the whole lib

v2.4.0 - 2025-09-17

Added

• rocket.py: add get_groups_history(), get_rooms_info(), send2webhook()
• time.py: add get_weekday()

Changed

• args.py: add --stratum parameter
• args.py: add --verbose parameter
• dmidecode.py: collapse near-duplicate data in an admin-friendly way

Fixed

• base.py: get_table() no longer modifies the input data
• redfish.py: check sensor state against user thresholds first

v2.3.0 - 2025-06-20

Added

• endoflifedate.py: add Icinga

Changed

• shell.py: add optional parameter lc_all='C' to shell_exec()

Fixed

• distro.py: incorrect os_family for Devuan in get_distribution_facts() (#87)

v2.2.1 - 2025-05-30

Fixed

• net.py: use ssl.PROTOCOL_TLS_CLIENT as best practice in fetch_ssl()

v2.2.0 - 2025-05-30

Added

• time.py: add get_timezone()
• tools/update-endoflifedate: add Valkey

Changed

• net.py: force fetch_ssl() to use TLS 1.2+
• txt.py: enhance sanitize regex

v2.1.1.15 - 2025-05-07

Changed

• net.py: add fetch_socket() and fetch_ssl(), improve fetch()

v2.1.1.7 - 2025-04-21

Changed

• distro.py: move get_os_info() from version.py to distro.py
• human.py: simplify bits2human() parameters (no more %-syntax)
• Improve code style across all libraries

v2.1.1.5 - 2025-04-19

Added

• txt.py: add sanitize_sensitive_data()

Changed

• base.py, url.py: use txt.sanitize_sensitive_data()
• disk.py: get_real_disks() now ignores loop devices
• docs: improve and convert docstrings to Markdown, create docs folder using pdoc
• shell.py: remove optional output from Windows chcp command

Fixed

• shell.py: fix special character decoding in Windows output by switching to codepage 65001

v2.1.0.7 - 2025-04-08

Fixed

• disk.py: fix static path to udevadm (#85)

v2.1.0.4 - 2025-03-29

Changed

• uptimerobot.py: add handling of PSPs, restructure code

v2.1.0.0 - 2025-03-23

Added

• uptimerobot.py: new library for the UptimeRobot API

v2.0.0.7 - 2025-03-10

Added

• tools/update-endoflifedate: add OpenVPN

Fixed

• txt.py: fix extract_str()

v2.0.0.0 - 2025-02-15

Breaking Changes

• Rename test.py to lftest.py due to nuitka compilation error on Windows.
• Switch from calendar versioning to semantic versioning due to conventional commits and Windows MSI requirements.

Added

• keycloak.py: new library for Keycloak API interactions

Changed

• db_sqlite.py: create_index(), cut(), delete(), insert(), replace(), and select() now delete the underlying SQLite db file by default on OperationalError (e.g., when new SQL references columns missing in an older db file). Disable with delete_db_on_operational_error=False.
• librenms.py: get_state() returns STATE_OK instead of STATE_UNKNOWN
• url.py: improve error messages and comments

Fixed

• disk.py: fix ValueError if value contains =

2024060401 - 2024-06-04

Minor improvements, barely any changes.

2024052901 - 2024-05-29

Breaking Changes

• librenms.py: get_state() now expects numeric status codes

Added

• args.py: add help()
• base.py: add str2bool()
• disk.py: add get_real_disks(), udevadm()
• human.py: add human2seconds()
• txt.py: add get_dm_name(), match_regex()

Changed

• base.py: oao() replaces | character in output with ! (reserved for performance data separator)
• db_sqlite.py: increase connect timeout to 5 seconds in connect(), return False on failure in close()
• grassfish.py: add insecure=False, no_proxy=False, timeout=8 to fetch_json()
• huawei.py: remove hardcoded insecure parameter in get_creds() and get_data()
• icinga.py: remove hardcoded insecure parameter; add insecure=False, no_proxy=False, timeout=3 to all functions
• infomaniak.py: add insecure=False, no_proxy=False, timeout=8 to all functions
• jitsi.py: remove hardcoded insecure and evaluate no_proxy in get_data()
• librenms.py: improve get_data() parameter handling
• net.py: add insecure=False, no_proxy=False, timeout=3 to get_public_ip()
• nodebb.py: evaluate no_proxy in get_data()
• rocket.py: add insecure=False, no_proxy=False, timeout=3 to all functions
• veeam.py: remove hardcoded insecure and evaluate no_proxy in get_token()
• version.py: add insecure=False, no_proxy=False, timeout=8 to check_eol()
• wildfly.py: evaluate insecure and no_proxy in get_data()

Fixed

• base.py: lookup_lod() never uses the default parameter (#82)
• db_mysql.py: fix select() and providing data
• feed.py: incompatible with Azure status RSS (#756)

2023112901 - 2023-11-29

Added

• endoflifedate.py: new auto-built library for end-of-life date tracking
• Published on PyPI, installable via pip install linuxfabrik-lib
• qts.py: new library for the QNAP QTS API
• tools/update-endoflifedate: tool to update endoflifedate.py

Changed

• base.py: cu() appends an optional message, making it a true error message function
• base.py: oao() prints/suffixes ' (always ok)' if always_ok=True
• shell.py: shell_exec() merges OS environment variables with those set by the env parameter
• version.py: check_eol() also fetches and caches info from https://endoflife.date/api

2023051201 - 2023-05-12

Breaking Changes

• db_mysql.py: change from username/password to option file authentication in connect()
• Remove all Python 2 based plugins and libraries, and remove the "3" suffix from all Python 3 based libraries (#589)

Added

• args.py: add number_unit_method type (used in disk-usage monitoring plugin)
• disk.py: add read_env()
• version.py: new library

Changed

• base.py: improve str2state()

Fixed

• smb.py: TypeError: object SMBDirEntry has no attribute from_filename

2023030801 - 2023-03-08

Breaking Changes

• db_mysql3: change from username/password to option file authentication in connect()
• net3: rename get_ip_public() to get_public_ip()
• net3: rename ip_to_cdir() to netmask_to_cdir()

Added

• dmidecode3.py: new library
• grassfish3.py: new library

Changed

• base3.py: make get_worst() more robust
• human3.py: human2bytes() now handles values like "3.0M"
• infomaniak3.py: apply new API version
• shell3.py: shell_exec() also handles timeouts
• wildfly3.py: update library

2022072001 - 2022-07-20

Added

• distro3.py: new library

Changed

• cache3.py: use more unique default names for SQLite databases
• db_mysql3.py: enhance for new mysql checks
• db_mysql3.py: switch from mysql.connector to PyMySQL (#570)
• db_sqlite3.py: use more unique default names for SQLite databases
• disk3.py: add file_exists() function
• Revert Python 3.6+ f-strings to .format() for broader compatibility

2022022801 - 2022-02-28

Added

• human3.py: new library for converting raw numbers and times to human-readable representations
• shell3.py: new library for shell communication
• time3.py: new library for date/time functions
• txt3.py: new library for text handling, encoding, and decoding
• redfish.py: add get_systems*() functions for Systems collection
• winrm.py: add function to run shell commands (#41)
• powershell.py: add PowerShell support (#40)

Changed

• base3: move filter_str() to db_sqlite3.py (#52)
• base3: move get_owner() to nextcloud-version3 (#53)
• base3: move sha1sum() to db_sqlite3.py (#50)
• base3: move x2human and human2x functions to new human.py library (#49)
• base3: move date/time functions to new time3.py library (#55)
• base3: move shell functions to new shell3.py library (#56)
• base3: move text functions to new txt3.py library (#51)
• Lint all libraries (#57)
• Standardize try-except import statements (#60)
• txt3: handle all encoding and decoding (#59)
• url3.py: extend fetch_json() to make fetch_json_ext() obsolete
• veeam: use new fetch_json() instead of fetch_json_ext() (#42)

Removed

• base3: remove yesterday() function (#54)

Fixed

• base: fix hashlib.md5() on FIPS-compliant systems (#30)
• base: fix tuple item assignment error (#43)
• librenms-alerts2: --lengthy causes error (#61)
• nginx-status: TypeError: a bytes-like object is required, not 'str' (#47)
• url3.py: AttributeError: 'str' object has no attribute 'to_bytes' (#62)
• url3: TypeError: a bytes-like object is required, not str (#44)
• Various fixes after linting
• veeam.py: ValueError: need more than 2 values to unpack (#45)
• veeam3.py, huawei3.py: getheader() should be get() in Python 3 (#46)

2021101401 - 2021-10-14

Added

• base: add utc_offset() function (#35)
• db_sqlite: add REGEXP function (#36)
• jitsi.py: new library
• nodebb.py: new library
• test.py: new library for unit testing
• veeam.py: new library

Changed

• base2: improve Unicode, UTF-8, and ASCII handling
• base: get_state() can now evaluate against a range (#34)
• base: version2float() now strips everything except numbers and decimal points (#26)
• base: improve line drawing in get_table() (#7)
• base: make version() more robust (#28)
• cache: make filename for cache configurable (#21)
• db_sqlite: support LIKE statements using a regexp
• get_table(): use ASCII characters only for broadest terminal compatibility (#33)
• url: fetch() and fetch_json() can now also return HTTP status code and response headers (#32)
• url: identify as Linuxfabrik Monitoring Plugin (#24)

Fixed

• base2: AttributeError: 'exceptions.ValueError' object has no attribute 'encode' (#37)
• base2: UnicodeDecodeError: 'ascii' codec can't decode byte 0xc2 (#38)
• base: get_table() does not handle length of UTF-8 correctly (#8)
• cache3: NameError: name 'base' is not defined (#29)
• db_sqlite: 8-bit bytestrings error with text_factory (#20)
• disk: problems with read_csv() (#25)
• librenms3.py: fix error (#27)
• net: fetch() port and timeout must be integers (#22)
• net: socket recv() timeout (#23)

2020052801 - 2020-05-28

Added

• db_mysql.py: new library
• feedparser.py: new library
• icinga.py: new library

2020042001 - 2020-04-20

Changed

• base.py: add shell_exec()
• net.py: improvements
• url.py: improvements

2020041501 - 2020-04-15

Added

• args.py: new library
• base.py: new library
• cache.py: new library
• db_sqlite.py: new library
• disk.py: new library
• net.py: new library
• rocket.py: new library
• url.py: new library

2020022801 - 2020-02-28

Initial release.