* Fix vendor_ouis duplicates in migration
Can't trim the duplicates efficiently due to lacking an index on oui.
* fix style
* restrict columns to make strict group by happy.
* implement LINE Messaging api
* apply code clean up
* Update Linemessagingapi.php
* fix variable type
* Match branding
* Remove extra proxy
---------
Co-authored-by: Neil Lathwood <[email protected]>
Co-authored-by: Tony Murray <[email protected]>
* MAC Vendor OUI use scheduler
Add command to update `lnms maintenance:fetch-ouis`
Show vendor column in tables if mac_oui.enabled is set to true
Improve scheduler validation handle non-standard install directories and systems without systemd
Add index to table to improve speed and improve mac->vendor lookup speed
Scheduled weekly with random wait to prevent stampeding herd issues for upstream
drop oui update from daily
* MAC Vendor OUI use scheduler
Add command to update `lnms maintenance:fetch-ouis`
Show vendor column in tables if mac_oui.enabled is set to true
* Lint fixes and better prefix detection
* update schema file
* add munging to generic stats graph include
* add graph_uptime_days config def
* make it optionally changable now
* style fix
* remove the graph_uptime_days option
* Force cast on some values before doing arithmetic
Some devices (like Delta InsightPower G3 Mini SNMP Card) return STRING instead of numerical value.
A few lines after, when doing a division on it, everything go bad ( / $divisor).
I added force casting, no more crash and the values are OK.
* Using helper cast_number() instead of intval()
* Fix style
* Add test data
* Stop using deprecated function cast_number
and increase default to 20s from 10s
20s should still be fast enough to prevent gaps, but larger installs can take longer than 10s (or even 20s) to do dispatch work.
* rework into a json app and support the old style
* working now
* add health checking
* add some datapoints to make writing alerts easy
* add a few missing item for has checks
* rework the app page
* use the right var for id9
* print the self test log if we have it
* add a graph for the max temp
* display the max temp graph
* display the health pass in the disk list if we have the data
* fix legacy handling and set app as ok
* replace id190 and id194 with the more useful maxtemp graph on the apps page
* don't print the additional info area for legacy extends
* add id232
* add id252 graph support
* properly display id232 now
* do not display SSD graphs for non-SSD disks
* for legacy extends, make sure we don't have a bad line
* add missing label for Product value
* fix metrics
* rework the metrics stuff a bit more
* typo fix
* more smart metrics cleanup
* add exit info to metrics
* style cleanup and logic checking for when disks are all now fine
* fix checking for no more health errors
* update the docs for SMART for v1 and remove slightly pointless notes about useSN
* update the docs
* initial test stuff
* add missing tests for the legacy code
* save if it it is legacy or not
* style cleanup
* update the smart legacy test
* style cleanup
* more test tweaking
* some more style cleanup
* more test fixes
* correct the disk ID for the legacy test da0 -> Z304VCFY
* more tests update
* more test cleanup
* begin cleanup of the v1 tests
* more v1 test updates
* more tests for v1
* more test cleanup
* fix exit_nonzero and add unhealthy metric
* add smart alerts
* add the metric unhealthy to the stat test
* use app data for the smart-common.inc.php