Repository navigation
Replies: 9 comments 3 replies
|
Im also seeing predbat constantly "active" since upgrading to HAOS v18 |
|
Yep, me too. Calculation time seem very slow - longer than the cycle time (10 mins). I am getting " Plan data is stale (last updated >15 minutes ago)" regularly. Full reboot of hardware seems to fix it for a few hours. Edit: reverting haos to 17.3 fixes it. |
|
Likewise, same issue here |
|
@springfall2008 few reports confirming this now. Do you want me to open this as an issue? I have held off updating to HAOS v18 after reinstating my earlier snapshot. |
|
"ha os boot-slot other" to save others looking :) |
|
Ive found to fix this you now need to use the IP address (http://xx.xx.xx.xx:xxxx) instead of http://homeassistant.local:xxxx for givtcp_rest in apps.yaml |
|
I've got internal DNS running on my router and fixed it by updating all .local references within HA and apps.yaml to my .internal domain. Not sure why you'd be having specific issues with the IP though |
|
Thanks for your help everyone. I was getting the same issue with Predbat constantly trying to do an update. I fixed it by replacing the homeassistant.local with an IP address for the givtcp_rest value and that fixed it. I then went into PiHole and added homeassistant.internal to that IP address (in addition to homeassistant.local) and it worked too when configured for givtcp_rest. There must be something up with HAOS v18.0 and using homeasssistant.local. |

Uh oh!
There was an error while loading. Please reload this page.
Anyone noticed Predbat weirdness after moving to HAOS v18? My Predbat install was constantly showing calculating "active" with plans about 10 minutes behind but nothing untoward showing in the logs and no particular CPU issues.
I rolled back my VM to a prior snapshot and everything OK again. Seems to be nothing to do with Prebat versions either - currently on 8.40.14 and all is well. I will take another snapshot again shortly and attempt HAOS v18 again to confirm it.
All reactions