* Throw some shit together, rough outline.
* Reorganize tabs, use tab controllers
* Implement performance (into the latency tab)
* Update resources/views/device/header.blade.php
Co-Authored-By: Jellyfrog <Jellyfrog@users.noreply.github.com>
* Add more tabs
* All controllers created
* Implement routes
* Implement smokeping
* routing and auth
* fix smokeping check
* Implement device dropdown menu
* Update deviceUrl to new style
* Use Gates
* Fix style
* use more appropriate gates
* add show-config gate
remove Laravel helper
* Only show vlan tab if VLANs exist for the device :D
* Fix rancid file check will return false
* revert over-zealous file name changes
* don't need to request the location parameter, just cast to string to avoid bugs when not found
* Move latency tab (ping/performance) to the position of performance instead of ping.
Co-authored-by: Jellyfrog <Jellyfrog@users.noreply.github.com>
* Support FreeNAS/TrueNAS 11.3
Seems the latest release changes the SysObjectID from FreeBSD to
their own enterprise number.
I've left the old FreeBSD enterprise number in place, so both
older and newer releases will be detected.
* Test data
It's also possible to do zpool IO from the MIB, but diskIO on
LibreNMS makes a lot of assumptions about it coming from the UCD
MIB.
This changeset adds detection for FreeNAS and TrueNAS - despite
having their own enterprise number, the snmpd uses the FreeBSD
one.
The naming is a bit all over the place, because this reflects
reality - IXSystems produce FreeNAS and TrueNAS, the product lines
have merged to eliminate FreeNAS, but TrueNAS still calls itself
FreeNAS everywhere except the GUI.
* opengear MemPools
Discovery and Polling of opengear mempools - opengear Detail Memory Usage is already collected however that can not be used to trigger rules.
* add test
opengear mempools tests
* remove newline
* Added sensors for outlets on Schleifenbauer devices. Also use the outlet name when configured.
* Updated tests for Schleifenbauer
* Fixed on test for Schleifenbauer
* Optimized polling by retriving only outlets from the device with the network connection, not all devices on the data-bus.
* Fixed code styling
* Use shorthand comparisons for names
* Fix for conflicting $entPhysicalIndex
* Add OS detection for Greenbone appliances
They don't expose anything fancy over SNMP, but this makes them
display interface names correctly. Under the hood they're unix-y.
Greenbone logo is from the AGPLv3 gsa, which is GPL compatible:
https://github.com/greenbone/gsa/blob/master/gsa/public/img/greenbone.svg
The mib is included for completeness, but it doesn't expose any
graphable values.
* Fix the vendor's logo
* BGP Polling: Add error code polling
* Rework describe_bgp_error_code and fix bgp error fields migration
* Add real test data for IOS-XR and Arista EOS
* OS detection and sensors for Dell Compellent
... or Dell storage as it is now known. Under the hood it's
referred to as Compellent, so I've named it as that to avoid
confusion.
It is not part of the Dell group or using the Dell MIB folder -
Dell have doubled up on some OID definitions, which means the MIB
here and a couple of the existing ones conflict.
The change adds OS detection and a frankly ludicrous amount of
sensors. I have access to a relatively low end device, so at the
moment everything under *ctlr* is my best guess based on the other
table layouts.
* Set sensor state_name to avoid collisions
* Regenerate test data
* Sev and Rel terminology varies between devices
* Most temp sensors don't have low limits, so adding
* update test data
Co-authored-by: Tony Murray <murraytony@gmail.com>
* Fix routing engine discovery on standalone JunOS devices
In #9426, the sensor discover code was replaced with YAML.
Unfortunately it treats the FRU table as standalone OIDs, and did not
implement the filtering that was present in the original code.
This change causes LibreNMS to fully walk the FRU table, which allows us
to silence alerting on devices with an 'unknown' status. Without this
change, standalone devices have a spurious routing engine with state 1,
which triggers false harware failure alerts.
As we're changing the name of the sensors, this causes a loss of sensor
history.
This has been brought up on community a couple of times:
https://community.librenms.org/t/juniper-qfx-routing-engine-state-alert/6152https://community.librenms.org/t/spurious-failed-routing-engine-from-some-junos-devices/6208
* Update test data
* Update test data
* Update test data
* Update test data
* This syntax works, but fails linting
* Create addhostCliTest.php
Added test for addhost.php cli tool
* Added test for addhost function
Added tests and fixed code in order for the function to be able to run with default parameter
* Create addhostTest.php
added tests for addhost function
* Update addhostCliTest.php
Changed from testing addhost.php to Artisan command 'device:add'
* remove delhost exec
* remove delhost
* Remove isn't needed, transactions are used.
* Remove unneeded, transactions are used.
Co-authored-by: Tony Murray <murraytony@gmail.com>
* Global Settings - Alert Default Settings
* Revert "Global Settings - Alert Default Settings"
This reverts commit a1af62b146.
* Widget Eventlog - Sensors link and overlib Image
* travis fix
* handling via AppServiceProvider
* collect function
* Travis ...
* static list of Sensor Types
* adding all sensor graph_types
* remove collect
* correct getTypes - remove change to lower case
* class sensor_descr switch
* class sensor_descr switch - revert
* Use standard sensor type names
* move types to attribute list
* going back to changes from 9eee7fad58
Co-authored-by: Tony Murray <murraytony@gmail.com>
* Global Settings - Alert Default Settings
* Revert "Global Settings - Alert Default Settings"
This reverts commit a1af62b146.
* remove in, not_in cause of non functional
* Delete ports via eloquent event
Chunk delete during purge all operations so we don't use too much memory.
* protect against missing device
* fix whitespace
* fetch less from the database when deleting a device's ports
fix output
apparently GD now returns false when getting the bounding box of empty strings, this causes jpgraph to throw an incorrect error. Work around that by setting all possible text to a single space character.
* Add Advantech logo.
Taken from https://commons.wikimedia.org/wiki/File:Advantech_logo.svg
with minor edits (removed width and height).
* Add Advantech MIB files.
* Add Advantech definitions.
* Add suppor for version and hardware polling.
* Add advantech snmp recording.
* Add sensors for SFP ports
Each SFP will have temperature, voltage, tx bias, Tx/Rx power.
Also update snmp recording to include new data.
* Rename ADVANTECH-EKI-PRONEER-MIB_1_01.mib to ADVANTECH-EKI-PRONEER-MIB
* Rename advantech-common-mib.mib to ADVANTECH-COMMON-MIB
* Update advantech.yaml
* Update advantech.svg
* Create advantech.json
Co-authored-by: PipoCanaja <38363551+PipoCanaja@users.noreply.github.com>
* Added support for ICT Modular Power System and ICT Sine Wave Inverter
* Added SNMP test data for ICT Modular Power System and ICT Sine Wave Inverter
* Fixed styling errors and refactored state sensor for ICT Sine Wave Inverter
* Fixed newly created styling issues from refactoring ICT Sine Wave Inverter files.
* Fixed further styling errors from previous changed to appease Travis CI
* Removed unnecessary math from ICT Sine Wave Inverter voltage sensor
* Cleaned up unnecessary lines in ict-mps/ict-swi sensors. Moved from snmp_get to snmp_get_multi in ict-mps/ict-swi os pollers.
* Added test data for ICT MPS and SWI
* Fixed issue with load sensor not discovering
* Update snmpget to snmp_get
Co-Authored-By: Jellyfrog <Jellyfrog@users.noreply.github.com>
* Updated test data post-fixing sensor detection issue on ICT Sine Wave Inverters
* Moved ict-swi and ict-mps from legacy to yaml sensors
* Removed legacy code for ict-mps and ict-swi sensors
* Last commit mistakenly removed OS polling from ict-mps and ict-swi
* Fixed formatting for OS polling and updated test data for ict-mps and ict-swi
Co-authored-by: Jellyfrog <Jellyfrog@users.noreply.github.com>
* honoring config option `force_ip_to_sysname` in bill creation and editing
* refactored variable name to match it in edit.inc.php
* forgot one variable
* and one more :S
* Update new_bill.inc.php
* Update edit.inc.php
* Update new_bill.inc.php
* Update edit.inc.php
* Update edit.inc.php
* Update edit.inc.php
Co-authored-by: Neil Lathwood <gh+n@laf.io>
Co-authored-by: Tony Murray <murraytony@gmail.com>
* perform json_encode() on memcached app $response data to match update_application() string parameter
* return raw snmp_get result for memcached response data during update_application()
* ipv6 support for BGP routes in VRP devices
* reverse bgpPeers array to allow VRF aware BGP
* codeclimate
* codeclimate
* codeclimate, finally
* Split VRP out of bgp-peers for readabilty
* Split VRP out of bgp-peers for readabilty
* tests
* CodeClimate
* improve function description
* improve function description
* v6data
* rebase
* space
* ipv6 support for BGP routes in VRP devices
* reverse bgpPeers array to allow VRF aware BGP
* codeclimate
* codeclimate
* codeclimate, finally
* Split VRP out of bgp-peers for readabilty
* Split VRP out of bgp-peers for readabilty
* tests
* CodeClimate
* improve function description
* improve function description
* v6data
* rebase
* space
* fixed corner case
* ascom support
* tests
* typo
* Correct missing 'mpls' in misc/os_schema.json
* remove mpls for now
* tests
* Update ascom.svg
* Don't disable already disabled modules
As I usually state, not much gain in disabling these here. Would be better to optimize the modules if you think they take too long on devices that don't have support.
* refactor polling code
Co-authored-by: Tony Murray <murraytony@gmail.com>
