This topic can be used to discuss problems/experiences/questions on the openHAB 3.0 Release Candidates.
events.log
doesnât contain events after upgrade from M5 to RC1. This can be fixed by editing userdata/etc/log4j2.xml
and changing all occurrences of smarthome.event
to openhab.event
.
Error during upgrade
I have tried to upgrade from M5 to RC1 using âsudo openhabian-configâ, master and option 2 : Upgrade System - but I get the following error:
Solved
After running âsudo systemctl stop frontail.serviceâ I can now upgrade to RC1.
-
The startup wizard sets Country/Region in Regional settings but does not set the Country setting in Ephemeris (remains empty by default).
-
NEATO binding: Account is online but scanning fails.
2020-12-14 21:22:23.557 [DEBUG] [internal.handler.NeatoAccountHandler] - Attempting to find robots tied to account
2020-12-14 21:22:25.630 [DEBUG] [internal.handler.NeatoAccountHandler] - Authentication Response: {âmessageâ:âNot allowedâ}
Manually adding the robot however works.
2020-12-14 21:34:34.289 [DEBUG] [.neato.internal.handler.NeatoHandler] - Will boot up Neato Vacuum Cleaner binding!
2020-12-14 21:34:34.290 [DEBUG] [.neato.internal.handler.NeatoHandler] - Neato Robot Config: NeatoRobotConfig [refresh=60, secret=REMOVED]
2020-12-14 21:34:34.290 [DEBUG] [ab.binding.neato.internal.NeatoRobot] - Will get STATE for Robot REMOVED
2020-12-14 21:34:34.290 [DEBUG] [.neato.internal.handler.NeatoHandler] - Start automatic refresh at 60 seconds
2020-12-14 21:34:34.291 [DEBUG] [ab.binding.neato.internal.NeatoRobot] - Calling Neato WS with body: {âreqIdâ:â1â,âcmdâ:âgetRobotStateâ,âparamsâ:{}}
2020-12-14 21:34:35.473 [DEBUG] [ab.binding.neato.internal.NeatoRobot] - Result from getRobotState: {âversionâ:1,âreqIdâ:â1â,âresultâ:âokâ,âdataâ: {},âerrorâ:null,âalertâ:null,âstateâ:1,âactionâ:0,âcleaningâ: {âcategoryâ:4,âmodeâ:1,âmodifierâ:1,ânavigationModeâ:2,âmapIdâ:ââ,âspotWidthâ:0,âspotHeightâ:0},âdetailsâ: {âisChargingâ:false,âisDockedâ:true,âisScheduleEnabledâ:false,âdockHasBeenSeenâ:false,âchargeâ:96},âavailableCommandsâ: {âstartâ:true,âstopâ:false,âpauseâ:false,âresumeâ:false,âgoToBaseâ:false},âavailableServicesâ: {âfindMeâ:âbasic-1â,âgeneralInfoâ:âbasic-1â,âhouseCleaningâ:âbasic-4â,âIECTestâ:âadvanced-1â,âlogCopyâ:âbasic-1â,âmanualCleaningâ:âbasic-1â,âmapsâ:âbasic-2â,âpreferencesâ:âbasic-2â,âscheduleâ:âbasic-2â,âsoftwareUpdateâ:âbasic-1â,âspotCleaningâ:âbasic-3â,âwifiâ:âbasic-1â},âmetaâ: {âmodelNameâ:âBotVacD7Connectedâ,âfirmwareâ:â4.5.3-189â}}
2020-12-14 21:34:35.477 [DEBUG] [ab.binding.neato.internal.NeatoRobot] - Successfully got and parsed new state for REMOVED
Iâm having the same issue hereâŠ
using master of openHABian config tool and tried with option 40-41 | openHAB testing and 40-42 | Upgrade to openHAB 3
There is another process running as user openhab, @Erling screenshot shows it as 413.
Identify this process, stop it and upgrade again.
@Benjy What does the usermod do here? Is it really needed for updates ?
If your logs are full of warnings like these
Field 'firmwareStatus' could not be eliminated: Can not set final org.openhab.core.thing.firmware.dto.FirmwareStatusDTO field org.openhab.core.io.rest.core.thing.EnrichedThingDTO.firmwareStatus to null value
when you go to the settings in the UI, thatâs my fault and a fix is on the way: https://github.com/openhab/openhab-core/pull/1932
If youâre so annoyed and canât wait to get rid of these messages, do a
bundle:update org.openhab.core.io.rest.core https://github.com/ghys/openhab-core/releases/download/rc1-rest-fix/org.openhab.core.io.rest.core-3.0.0-SNAPSHOT.jar
(disclaimer: in theory you shouldnât update untested bundles like this )
After migration from MR5 -> RC1 my rules triggered on Item-State-Changes did not fire. Manual execution of the rules worked fine.
After an additional reboot, the problem went away without doing other changes.
Thanks, in my case itâs PID 553:
openhab 553 0.8 1.2 144004 49088 ? Ssl 21:33 0:11 node /usr/bin/frontail --ui-highlight --ui-highlight-preset /usr/lib/node_modules/frontail/preset/openhab.json -t openhab -l 2000 -n 200 /var/log/openhab/openhab.log /var/log/openhab/events.log
when I kill this process it starts again with another PID and then I get the same message but with the new PID.
Upgrade errors out with the same message.
Donât kill the process, do a clean
sudo service frontail stop
or whatever command it takes on your system.
I normaly stop frontail before upgrading, as I also delete the logs.
ah perfect, thanks!
After stopping the frontail service the upgrade passed fine.
Weird as itâs the first time I needed to do this.
Hi All!
Tried to upgrade from M5 to RC1 in docker by replacing with new image. It got stuck in endless loop as it says âYou are already on Openhab 3.0.0.M5â and restarts up upgrade process. If I put M5 image back it boots properly.
When the upgrade detects that the home directory is still /var/lib/openhab2
it changes it to the correct path. I hadnât realised that openHABian still uses the openhab
user for frontail (it shouldnât do (cc @mstormi)). Iâll release a workaround.
Itâs probably the Portainer issue described in the README:
@Wolfy
Yes, when using portainer you have to delete the âOPENHAB_VERSIONâ-Environmentvariable of your OpenHAB-Instance before doing the upgrade.
Right it shouldnât and Iâd think it doesnât on fresh installs in quite a long time. If you have installed openHABian/frontail before this got fixed though youâll still have the old user. Re-installing frontail via menu might help there.
6 posts were split to a new topic: Migration question
The âRC1-2â might be the version with the fix, that @ysc mentioned above, already applied (but Iâm not 100% sure)
In that case your commands would be:
wget https://openhab.jfrog.io/openhab/list/openhab-linuxpkg/pool/main/3.0.0~RC1/openhab_3.0.0~RC1-2_all.deb
sudo apt install ./openhab_3.0.0~RC1-2_all.deb
If you do a fresh install anyway, I would try it out and having a look in the logs after install - if anything looks good, youâre probably good to go.
The â-2â at the end is just for the .deb
packaging, it fixes the following install error.
The content after the install is identical and is now given to anyone who uses the milestone/testing repository. Unfortunately that does mean the same issues are present.