NetXMS Forum

Thank you very much for your quick reply. Our use case is: based on historical IP addresses (for example, IPs logged when users access critical systems), we want to look up the MAC address associated with that IP at that point in time, so we can identify the specific endpoint. We look forward to this new feature. Many thanks!

Автоматически — нет. У сервера есть проверка собственной связи (Beacon.Hosts), но к потере связи она привязывает только SYS_NODE_DOWN. SYS_NODE_UNREACHABLE от нод через тоннель и за прокси, а также SYS_NODE_UP и ваши icmp ok не коррелируются. На это завёл тикет: https://github.com/netxms/netxms/issues/3753 Настройками шквал "Node up" убирается так: В Server Configuration задать Beacon.Hosts — два-три внешних адреса через запятую (не шлюз локальной сети, он при пропаже интернета продолжит отвечать) — и перезапустить сервер. При потере связи со всеми адресами сервер генерирует SYS_NETWORK_CONN_LOST, при восстановлении — SYS_NETWORK_CONN_RESTORED; стандартные правила держат на это время аларм с ключом NET_CONN_LOST. Добавить правило на SYS_NETWORK_CONN_RESTORED, выше стандартного "Terminate NetXMS server network connection loss alarm when connection restored": действие DoNothing, Delayed by 15m, timer key UPLINK_RECOVERY. Из правил 26, 32 и 34 вынести Normal MailAlarm (вместе с его "Do not run if timer ... is active") в отдельное правило, стоящее непосредственно перед исходным: те же source и event, флаг "Accept correlated events" как у исходного, "Stop event processing" не включать. В исходном остаются Terminate alarms, DoNothing и Cancel timers. В каждом из трёх новых правил — filtering script:return FindAlarmByKey("NET_CONN_LOST") == null && CountScheduledTasksByKey("UPLINK_RECOVERY") == 0; Отдельное правило нужно потому, что скрипт в фильтре исходного правила отключил бы вместе с письмом и закрытие аларма, и отмену таймеров. Note: в течение 15 минут после восстановления связи письма "Node up" и "icmp ok" не отправляются вообще, в том числе от нод, которые в это время действительно поднялись. Алармы при этом закрываются как обычно.

Hi, most likely this is not a hang - installer is waiting for database initialization, which runs in separate console window. If initialization fails, that window shows the error and "PRESS ANY KEY TO CONTINUE", and it can end up behind the installer window. With SQLite it usually fails either because folder in "Path to database file" does not exist (file is created automatically, folder is not), or because database file is already there from previous attempt. Log switch does not change anything here. On second run installer finds existing installation and "Initialize database" is not checked by default, so this step is just skipped - which also means database may still be not initialized. Is NetXMS Core service running and can you log in? If not, please share what you entered as path to database file and last lines of C:\NetXMS\log\netxmsd.log.

Hi! We just published patch release 6 for version 6.2. Hopefully that will be last patch release before major release 7. Notable changes since previous patch release: -  Box selection of elements on network maps -  Configurable size of port boxes in port view dashboard element -  Pie chart dashboard element can show actual values instead of percentages -  Status indicator dashboard element uses object or DCI name as label when custom label is not set -  Object context macros are expanded in dashboard web page element URLs; new macro form %~ URL-encodes expansion result -  New web API endpoints for posting events and for managing mapping tables, persistent storage, and SNMP traps -  New NXSL function FindDCIById; FindAllDCIs accepts null as node argument to search across all data collection targets -  Server console command to recommission decommissioned node; decommissioned nodes can no longer be returned to managed state via NXSL, web API, parent object, or network discovery -  Scheduled task operations and maintenance mode changes made from scheduled tasks are written to audit log -  NXSL scripts run from scheduled tasks and AI assistant tools execute with user security context -  LDAP diagnostic messages from libldap are logged on TLS and bind failures -  File delivery policy files are hidden by default in server file store views; "Show hidden files" option added -  Agent: user sessions on Linux are read from systemd-logind, utmp is kept as fallback -  Agent: processes started in user sessions get environment of session user; external subagent loader no longer deletes master agent policy files at startup -  Agent: new configuration parameter ExternalSubagentWatchdogExcludedUsers; watchdog skips sessions without logged on user -  Agent: authentication and encryption requirements enforced for all commands; AI file operation tools restricted to configured root folder and master server -  Windows server installer ships JSON templates; Windows client installer always runs Visual C++ runtime installer Full list of closed issues: -  #805 / NX-594 (Box selection on network maps) -  #2328 / NX-2120 (Configurable size of boxes in port view on dashboard) -  #2680 / NX-2474 (Add option to show numeric value instead of % on bar chart) -  #3649 (Autobind can re-link an object that is being deleted, leaving a dangling child reference in containers and templates) -  #3651 (Interface Traffic SNMP template auto-applies together with agent template, duplicating traffic DCIs) -  #3652 (Windows client installer runs vcredist only when the nxshell component is selected) -  #3653 (DefaultDCIPollingInterval / DefaultDCIRetentionTime descriptions still say "for newly created DCI") -  #3654 (Changing DataCollection.DefaultDCIPollingInterval does not resync agent cache mode DCIs) -  #3655 (nxdbmgr -C on MS SQL creates user without rights to create tables) -  #3656 (Windows agent: process enumeration skips last entry, Process.Count returns 0 for newest process) -  #3662 (Unsynchronized read of DelegateObject object set (possible use-after-free)) -  #3665 (NXSL scripts run from scheduled tasks and AI assistant tools have no user security context) -  #3666 (NXSL function SQLQuery leaks database connection on every call) -  #3668 (missing label for dci in Status indicator) -  #3669 (Feature Request: Expand Object Context Macros in Dashboard Web Page Elements) -  #3670 (No size check on agent certificate mapping data, node save fails with 'value too long for type character varying(500)') -  #3672 (Agent certificate mapping by public key overflows varchar(500) and breaks node persistence) -  #3674 (Linux agent reports no user sessions on hosts without utmp (Debian 13, systemd 257)) -  #3675 (Agent: CMD_AI_EXECUTE_TOOL and CMD_AI_GET_TOOLS have no server class check and aifileops has no path restriction) -  #3676 (Agent: RequireAuthentication / RequireEncryption not enforced for commands handled in receiver thread) -  #3678 (Server crashes on login when group membership search path exceeds 256 entries) -  #3680 (LDAP: log libldap diagnostic message on TLS/bind failures) -  #3681 (Auto-link nodes on custom maps does nothing since 6.2.0) -  #3682 (Threshold events generated on maintenance leave are correlated to the maintenance event and skipped by EPP) -  #3683 (PostgreSQL driver: failed unbuffered SELECT leaves the connection busy, next unbuffered query fails with "PQsendQuery failed") -  #3685 (v5migrate status: "Active workers" counts queued + running objects, "Pending objects" is a stale batch snapshot) -  #3686 (V5 data migration worker holds a pooled DB connection for the whole object, including while paused or throttled) -  #3687 (Config parameter Objects.Nodes.DefaultDecommissionExpirationTime is missing on fresh installs and is never readable by the client) -  #3688 (NXSL NetObj::manage() bypasses the decommission guard, leaving a node that looks managed but is still scheduled for deletion) -  #3689 (Add server console command to recommission a decommissioned node) -  #3690 (SNMP GET: noSuchObject/noSuchInstance returned as empty value instead of unsupported (regression since 6.0.0)) -  #3691 (Zone deletion removes credential tables by object ID instead of zone UIN, leaves ssh_credentials) -  #3692 (Object deletion leaves rows in dashboard_associations, object_ai_data, port_stop_list, cluster_resources, radios, dci_delete_list) -  #3694 (DCTable deletion skips dct_threshold_instances; DCI deletion never cleans cond_dci_map) -  #3695 (Discovery re-manages unmanaged node via Node::changeIPAddress() (sets status directly, bypassing setMgmtStatus)) -  #3696 (nxdbmgr upgrade 61.35->61.36 fails on PostgreSQL: GetDataCollectionTargets() selects from 7.0-only tables) -  #3700 (Status poll re-manages individually unmanaged interfaces and network services on network path problem (resetStatus bypasses setMgmtStatus)) -  #3704 (MIB DISPLAY-HINT: multi-octet "d" fields are rendered one octet at a time, and separators are emitted before the field instead of after) -  #3705 (Decommission with "Clear IP addresses" leaves node in address index: former IP reported as "already in use" until server restart) -  #3706 (DCI deleted while busy (collection or instance discovery) is never deleted from database) -  #3707 (Web client: downloads fail silently for file names outside Latin-1 (Content-Disposition not encoded)) -  #3708 (netxmsd crash: CMD_GET_DCI_VALUES on table DCI without last value dereferences null table) -  #3711 (Scheduled tasks created from client (incl. scheduled maintenance) are not written to audit log) -  #3712 (SYS_MAC_ADDR_CHANGED: interfaceName parameter missing for pseudo-interface) -  #3714 (Show hidden files toggle for server file store; hide file delivery policy files by default) -  #3715 (ProcessExecutor: no bounds check on argv[256] when building child argument list) -  #3718 (Agent: external subagent loader deletes master agent policy files at startup) -  #3719 (Agent: option to exclude user accounts from external subagent watchdog) -  #3720 (Windows server installer does not ship JSON templates) -  #3721 (MSSQL template has no auto-apply filter) -  #3722 (null node in FindAllDCIs) -  #3724 (Agent: external subagent watchdog stops at first session even if nobody is logged on to it) -  #3725 (NXSL: sometimes 'while' loop compiled incorrectly) -  #3733 (nxreportd ignores ReportingServer.ResultsRetentionTime (misspelled property name in housekeeper)) -  #3734 (Delayed incident creation from EPP drops description and truncates text at semicolons) -  #3735 (Isolation forest anomaly detection looks back seconds instead of days (Timestamp unit mismatch in IsAnomalousValue)) -  #3736 (Formatted DCI value has a space between multiplier prefix and unit ("2.31 Gi B")) -  #3737 (AI anomaly profiles are regenerated every second day: 24 h cooldown equals the daily task period) -  #3738 (Asset management: unique attribute check rejects writing the asset's own current value) -  #3739 (Asset management: date attribute validation accepts invalid dates (isValidDate logic inverted)) -  #3740 (Oxidized module: backup timestamps ignore the UTC offset reported by Oxidized) -  #3741 (LLM provider type openai falls back to Ollama URL and model when URL/Model are omitted) -  #3742 (AI tasks keep running after their owner's account is disabled) -  #3743 (netsvc: TLS.Certificate.* option list starts at the StartTLS protocol argument) -  #3746 (OTLP event forwarder: dropped events are not reflected in forwarder counters and status) -  #3747 (AI: task prompt lacks opening <instructions> tag; show ai providers reports Mistral as Ollama) -  #3748 (nxshell samples call removed method getAllChilds())

Hi, Yes, Module = mcp enables it, and you'll want Module = aitools as well - most of the tools the MCP server exposes come from that module. The module has no settings of its own. It adds the endpoint to the server's built-in HTTP listener, so port and address are set in the [WEBAPI] section of netxmsd.conf: Module = mcp Module = aitools [WEBAPI] Address = any # default is loopback only Port = 8000 # default The endpoint path is fixed: http://<server>:8000/mcp (streamable HTTP transport). No separate web service is needed - the listener is part of netxmsd and enabled by default. Module = webapi only adds the REST API endpoints, MCP doesn't depend on it. Clients authenticate with an "Authorization: Bearer <token>" header. Issue a persistent token in the user's properties, and give that user the "Use AI assistant" system right, otherwise requests get 403. You're right that this isn't documented, ticket: https://github.com/netxms/netxms-doc/issues/87

Hi, This is a bug in the server, not your configuration: https://github.com/netxms/netxms/issues/3682 Here's what's actually happening: when the node leaves maintenance mode, the server re-generates threshold events for every threshold whose state changed during maintenance. That's where your two "task running" events at 04:00 came from. Those events can be processed while the node is still flagged as in maintenance, and then they get correlated to the "maintenance enter" event, which is why they reference 8766568. By default, EPP rules ignore correlated events, so the rule that should terminate the alarm never sees them. It's a timing race, which is why it only happens sometimes. Workaround until it's fixed: enable Accept correlated events on the EPP rule that terminates the alarm on OS_PROCESS_OK. As a side effect, the alarm will also be terminated when the task recovers while the node is still in maintenance, which is probably what you want anyway. If the same rule also creates the alarm, split it into two rules and set the flag only on the terminating one. Otherwise alarms will be created during maintenance too.

Hi, Confirmed, this is a bug in the 6.2 management client, not in the server or your setup. Since 6.2.0 the Auto-link action sends an empty node list to the server, so it reports success and creates nothing. Layer 2 maps are not affected because their links are built by the server from topology data. Fix is tracked in https://github.com/netxms/netxms/issues/3681 and will be in the next 6.2 release. Until then, "Link objects" still works, but it links two objects at a time and doesn't bind the link to interfaces.

Great! Thanks Alex.

Hello Alex, you are right. My mistake was that I downloaded the wrong file, now it works as expected. Thank you

Yes, it's already fixed in the code and will be in 6.2.6, which isn't released yet: https://github.com/netxms/netxms/issues/3656. Only the agent on the Windows machine needs to be updated.

Hi, Still not fixed - nothing in 6.2.4 or 6.2.5 changed this part of the agent. Remaining problem is now tracked in https://github.com/netxms/netxms/issues/3621 (#3137 was closed after the split). To move it forward, please post: exact warning lines from the server log - package name in them shows which part of the agent produces the duplicates nxagentd log with debug level 6 for tag "winnt", covering a full logon and logoff System.InstalledProducts table taken twice - once with a user logged on, once without whether these are RDS/terminal servers, and whether profiles are roaming, FSLogix or UPD

Great job, thanks everyone! Alessio.

Thank you Victor and Alex.  I appreciate the response!

Thanks for the info, after reading the docs about Agent External Metrics I converted the actions to shell form metrics and now it works fine. (Before I read the docs I tried the format you provided, but that had some issues which didn't trigger an alarm so I have no idea what was wrong with that.)

We just published patch release 5 for NetXMS 6.2. Notable changes: -  New network device drivers for Arista and Mellanox switches, and for Dell OS10 and Force10 (FTOS) switches -  New web API endpoints for pushing DCI values, getting the last value of table DCIs, and deleting objects; nodes' primary host name or IP address can be changed via web API -  Web API requests are handled by a thread pool (size is configurable, default 16) instead of a single thread -  Weather subagent: added Bright Sky provider (DWD open data for Germany) -  Unnecessary trailing zeros are removed from formatted fractional values -  Event processing policy editor opens large rule chains faster -  Data collection scripts are no longer restricted by Scripts.RestrictWriteAccess -  Windows installers bundle OpenSSL 3.5 LTS instead of 3.0.x, which has reached end of life And full list of closed issues: -  #2854 / NX-2651 (Driver support for Arista and Mellanox switches - Port View is Blank) -  #3423 (New network device driver: Dell Networking OS10 / Force10 (FTOS) switches) -  #3542 (nxdbmgr upgrade to 62.25 overwrites customized interface event messages) -  #3550 (Editing attributes of LDAP-synced users: silently overwritten on next sync) -  #3575 (Object deletion: removeDCI option is mutable between request accept and worker execution) -  #3609 (WebAPI: collected table DCI data is not reachable) -  #3610 (Value cache placeholder (empty value, timestamp 1 ms) reported as collected value in current-values) -  #3612 (AI assistant function ask-user-choice always fails: tool schema declares every parameter as string) -  #3613 (Web API endpoints for delete objects and modify object primaryName/ipaddress) -  #3615 (Critical BUG: system crash while create json obj in EEP->action->script) -  #3616 (Web API for pushdci) -  #3617 (NXSL JsonArray set()/insert() silently succeed on out-of-range index) -  #3618 (Connection history generates MAC events for interfaces rejected by Hook::CreateInterface) -  #3619 (Scripts.RestrictWriteAccess blocks agent command execution in script-origin DCIs with no per-DCI override) -  #3620 (Denied call under Scripts.RestrictWriteAccess surfaces as an unrelated NXSL error, and the parameter description understates its scope) -  #3622 (nxmc: create rule editor content lazily in the event processing policy editor) -  #3623 (Rack images crash the web client) -  #3630 (File manager subagent ignored read-only restriction for file permission change and move operations) -  #3631 (Remove trailing zeros if more than 2) -  #3634 (Physical container rebinding is not serialized; rapid placement changes can leave the object tree on a stale container) -  #3635 (Weather subagent: Bright Sky provider (DWD open data for Germany)) -  #3636 (Null dereference from unvalidated client input: delegate object in getLastValuesByDciId, element index in DashboardBase) -  #3637 (Server crash: MySQL/MariaDB drivers alloca the whole field length, overflowing the thread stack on large image library entries) -  #3638 (AI assistant: questions asked from a delegated skill never reach the user and time out) -  #3639 (Network discovery discards the SNMP credentials it found; every discovered node is created with community "public") -  #3640 (Built-in SNMP agent reads SNMP.Agent.Community, but the configured parameter is SNMP.Agent.CommunityString) -  #3641 (Management console: scrollable dashboards leak a ControlListener on every rebuild, UI thread ends up in a layout loop after hours) -  #3642 (WebAPI: all REST requests are serialized on a single libmicrohttpd thread) -  #3643 (Move bundled OpenSSL in Windows installers from 3.0.x (EOL) to 3.5 LTS)

Got it, thank you very much for your reply!

Thank you for pointing me in the right direction, I had not created the webhook correctly.

Your fresh install working is the useful data point - this is something about that specific server, not the pinboard code. Here's what's actually happening: all web UI settings live in a single properties file on the machine running the web UI, under the servlet container's work directory, named nxmc.preferences.<uuid>. The uuid comes from a cookie called nxmcStoreId in your browser. If that cookie doesn't come back on the next login, the web UI generates a new uuid and opens an empty file - and the old one stays on disk with all your settings still in it. Pinboard, the "Show server clock" checkbox and column widths all live in that one file, which is why they go together. Three things to check, and they split the problem cleanly: How many contexts are deployed. Look at /var/lib/tomcat10/work/Catalina/localhost/ - if more than one nxmc context is there (nxmc and nxmc-6.2.4, say), each has its own state directory, and the files you find may belong to a context you are no longer using. The web UI logs a line "State directory:" at every startup naming the one actually in use. Whether a new file appears per login. List the state directory, log in again, list it again. A new nxmc.preferences.<uuid> each time means the cookie is not coming back. The same file with a new timestamp means it is, and the problem is elsewhere. The cookie itself. Browser dev tools, Application - Cookies, look for nxmcStoreId. It is set with a 90 day lifetime, so it should survive a logout. If your nginx config caches or strips Set-Cookie on the proxied responses, it won't. If you already have a file there containing pinboard$<number> entries, that settles most of it on its own: saving works, and the session that comes up empty is reading a different file. The web UI log goes to catalina.out on a Tomcat install - grep it for "State directory", "local preferences" and "Invalid store ID". (-Dnxmc.logfile is only set in our Jetty instructions, so on Tomcat everything lands in the console log.) Note: even once this works, that directory is Tomcat's per-context work directory and it is cleared on war redeploy. So web UI settings are lost on every web UI upgrade by design - that is likely part of what you saw across the three version upgrades.

We just published patch release 4 for version 6.2. Notable changes since previous patch release: -  Server provides native MCP (Model Context Protocol) endpoint, replacing standalone Java MCP server -  Agent-side data collection using Nagios-compatible monitoring plugins via ExternalCheck configuration option -  Generic weather subagent with support for multiple providers (MET Norway and Open-Meteo) -  New network device drivers for Aruba CX switches and Cisco Catalyst 9800 wireless controllers -  MQTT subagent supports instance discovery, wildcard topics, and parameterized metrics in extractors -  MQTT topic payloads are no longer truncated at 255 bytes -  Option to synchronize node names with SNMP sysName on configuration poll -  Dashboards can be shared via anonymous public access links -  Network maps show link details on mouse-over -  Connection point search uses LLDP/CDP peer information in addition to FDB data -  Simplified adding multiple DCI-based or object-based checks to business services -  Multiple conditions can be added to log viewer filters at once -  Shift+click range selection of cells in DCI table value viewer -  Option to show total value and change font size in pie charts -  "Export all to CSV" action added to Interfaces view -  Script library scripts can be provided as context to AI assistant in script editor -  OpenTelemetry log record body exposed as event parameter -  ICMP statistics are no longer reported for nodes without valid primary IP address -  Fixed handling of 6-field cron expressions in recurrent scheduled tasks -  Fixed L2 topology building on Cisco switches that provide bridge port mappings in per-VLAN SNMP contexts -  Fixed SNMPv1 trap encoding that caused traps to be dropped by Net-SNMP receivers -  Fixed Linux agent crash when requesting Process.\* metric with "average" aggregation for non-existent process -  Fixed double-counting of shared memory in Process.MemoryUsage with "sum" aggregation causing false memory alarms -  Fixed memory leak in management client -  Fixed housekeeper and v5 data migration stall when throttle watermarks are set to 0 -  Fixed inverted "disable 802.1x status poll" node polling property -  Fixed vendor detection and management interface reporting in generic Cisco driver Full list of closed issues -  #1279 (DCI table value viewer: Shift+click range selection of cells) -  #3082 / NX-2880 (Add multiple conditions to log viewer filter at once) -  #3257 (Do not report ICMP statistics for nodes without valid primary IP address) -  #3262 (Topology.DefaultDiscoveryRadius ignored for network maps saved by pre-5.1 clients (radius stored as 65535)) -  #3413 (Huawei V-STP bridge ID equals master chassis' own base bridge address, contradicting getStpBridgeId() assumption) -  #3422 (New network device driver: Aruba CX (AOS-CX) switches) -  #3479 (Cisco L2 topology: FDB is read per-VLAN but bridge ports are read in the default SNMP context, so ifIndex resolves to 0) -  #3525 (6-field cron expression on recurrent scheduled tasks almost never fires (seconds field always matched)) -  #3529 (Replace standalone Java MCP server with native MCP endpoint in server (mcp module)) -  #3531 (Web client: node stays attached to the cursor after dragging past the viewport edge) -  #3532 (Web client: "Save map image to file" does nothing) -  #3533 (WebAPI returns 415 for bodyless POST requests without Content-Type header) -  #3536 (MAC Address web ui bugs) -  #3538 (Housekeeper throttle watermarks of 0 permanently stall housekeeper and v5 data migration) -  #3539 (ConfigWriteStr treats a failed existence query as "variable does not exist" and silently loses the write) -  #3540 (nxdbmgr upgrade: stale TimescaleDB extension version reported as an unrelated SQL failure) -  #3543 (NXSL: PostEventEx return value lost; origin argument rejects OPENTELEMETRY) -  #3551 (Memory leak - nxmc jar client (tested on windows)) -  #3553 (pie chart - add total % and change font size) -  #3558 (PostgreSQL process template raises false memory alarm — Process.MemoryUsage(...,sum) double-counts shared memory) -  #3559 (MQTT subagent silently truncates topic payload at 255 bytes) -  #3560 (MQTT extractors: wildcard topics with instance discovery) -  #3561 (MQTT extractors: parameterized metrics ($1 placeholders) documented but not implemented) -  #3563 (Network map: show link details on mouse-over) -  #3564 (Agent-side data collection via Nagios-compatible monitoring plugins (ExternalCheck)) -  #3567 (OpenTelemetry log record body not exposed as event parameter) -  #3569 (WebAPI object handlers ignore EDIT\_COMMENTS, EDIT\_RESP\_USERS and the template-parent READ exception) -  #3570 (Generic weather subagent with multiple providers (MET Norway) and Open-Meteo commercial API support) -  #3571 (Unbounded alloca over ACL size in AccessList::getUserRights and SaveGraph) -  #3573 (WebAPI: path placeholder object IDs parsed with base 0 and no end-pointer check) -  #3574 (FormatISO8601Timestamp() dereferences NULL for pre-1970 timestamps on builds without gmtime\_r) -  #3576 (libnxsnmp: SNMPv1 traps encoded with INTEGER time-stamp and wrong enterprise OID, dropped by Net-SNMP receivers) -  #3577 (Float DCI values between 0 and 1 displayed as raw six-decimal string) -  #3578 (Linux agent crashes when requested Process.\* metric for non-existent process and 'average' aggregate) -  #3579 (ReadWebsocketFrame corrupts payload of unmasked non-data frames (uninitialized mask)) -  #3580 (Connection point search ignores LLDP/CDP peer data in node overview and IP address search) -  #3581 (Objects.Nodes.SyncNamesWithDNS has no effect unless Objects.Nodes.ResolveNames is disabled) -  #3585 (snmp traps filter doesnt work) -  #3589 (Object tool translations for region-qualified language codes (zh\_TW, pt\_BR) never resolve at runtime) -  #3593 (nxmc on Windows: "unhandled exception (NoClassDefFoundError: org/netxms/bridge/Platform)" on exit) -  #3595 (Vendor field is not populated for nodes with CISCO-GENERIC driver) -  #3596 (Missing "Export all to CSV" action in Interfaces view) -  #3597 (Simplify adding multiple DCI-based or object-based checks to business service check list) -  #3598 (Management interface reported as front panel port on stackable Cisco switches (CISCO-GENERIC driver)) -  #3599 (Implement driver for Cisco C9800 wireless controllers) -  #3600 (Sync node name with SNMP sysName on configuration poll) -  #3601 (NXSL: FindAlarmById() and FindAlarmByKey() return alarm with empty relatedEvents) -  #3603 (AI assistant sidebar: provide script library script as context in script editor) -  #3605 (Node polling property "disable 802.1x status poll" is inverted) -  #3606 (Add "Enable public access" (anonymous access link) for dashboards)

Thanks Alex, super helpful.