Hydros Maven or Neptune?

[BeanAnimal]

There are several people reporting that if they lose connection they go into reboot mode with yellow LEDs.

 

[n2585722]

There are several people reporting that if they lose connection they go into reboot mode with yellow LEDs.

yellow LED's are not a reboot mode and only indicate they cannot connect to the cloud. Here is a video of a X4 with no internet connection . It is not rebooting but just giving a visual indication that it cannot connect to the cloud.

 

[n2585722]

There are several people reporting that if they lose connection they go into reboot mode with yellow LEDs.

Here is another video of no internet for an extended time period.



I got the videos by unplugging my cable modem. The controllers go through a period of breathing the green yellow and then a short period of breathing red. My guess is my router is overloaded with internet request and that is causing the short periods of the no wifi indication from the controllers. As you can see they all do not do it at the same time. Basically breathing red is no wifi at all, breathing orange is issues with the router granting access. breathing green yellow is no cloud access. Breathing blue on several of the controllers is everything is good. The breathing blue will only happen during the times setup in the options and does not happen at night.

 

[BeanAnimal]

Is there a reboot log somewhere - or other system level debug logging?

 

[n2585722]

Is there a reboot log somewhere - or other system level debug logging?

Yes there is one. I have had a reboot on one yesterday and one that goes back as far as 10/21 since the last reboot and that was probably when I rebooted the collective. The one from yesterday I probably did when changing some of the settings. If for some reason a controller fails update changes I will reboot it from the app and then retry update changes.

 

[jrill]

There are several people reporting that if they lose connection they go into reboot mode with yellow LEDs.

Not something I've experienced with 8 controllers. And not something I've read about which is odd since I try to follow Hydros related stuff in several places

 

[areefer01]

There are several people reporting that if they lose connection they go into reboot mode with yellow LEDs.

Maybe link those posts or tag them here so they can reply.

 

[Sisterlimonpot]

Here's the post and thread in question:

Maybe link those posts or tag them here so they can reply.

The devices will function with whatever they were doing for the current schedule but will eventually start rebooting on a 1-2 minute loop when they fail to get a connection from the server. This was brought to my attention via a support ticket and was honestly shocked that this was the solution. It means that if CoralVue ever has an extended outage period, or goes out of business, within 30 days (certification expiring time that ultimately causes the yellow breathing lights) all these devices are toast and people will be holding a big bag of useless bricks. When it enters this "reboot loop mode" with flashing yellow led's, it will constantly reboot all of your equipment plugged into it as well and restart any schedules. It caused so many issues with my tank that I almost ripped it all out in frustration before eventually finding a solution. While you wait on that though, you can't even put your tank in feed mode and the only way to shut equipment down is to physically unplug things. I honestly feel like none of this equipment has been tested in a "worst case" failure kind of scenario and it's just waiting to drop on everyone. Those kind of scenarios always come as such a shock to people that think "this will never happen to me" which is the vibes I get from CoralVue currently.

The issue has been frustrating reporting to CoralVue because they refused to acknowledge that it is an issue and anyone who brings it up quickly get's their threads locked on their facebook group. I've looked into parameter monitoring for some of the open source platforms and eventually want to move to one of those given that CoralVue has no interest in some of the things the community has brought up like this and an actual API integration.

 

[areefer01]

Here's the post and thread in question:

Thanks. Wasn't meant to be a challenge if that makes sense. I'm aware of another thread on another platform talking about the level sensor.

 

[n2585722]

Here's the post and thread in question:

It has to be more than just loosing the internet. I lost internet on Wednesday for 6 hours and the only reason I was aware of it was the graphs had a 6 hour gap. He tank was fine. I also had a 3 hour internet outage on Thursday and no issues. It was about 3 years ago but I lost the internet for 5 days straight and the controllers continued to run as they were setup to do. The only way to communicate with then was to use my phone with mobile access and star the app and choose bluetooth mode. Then I could see all the inputs and outputs. I could start a mode or override output to off, auto or on if needed in bluetooth mode. So it would do what was describe just by loosing internet I would have had issue without internet for those 5 day or even more recently with the 6 hour or 4 hour outage. The controllers do not like no wifi though but even then They would still try and control things but may reboot occasionally to try and reconnect to wifi. A reboot only takes a few seconds though.Breathing green yellow is just an indication on the controller that it cannot connect to the cloud but not a reboot. I posted a couple of links to the lights on the controllers when there is no internet access.

 

[n2585722]

There are several people reporting that if they lose connection they go into reboot mode with yellow LEDs.

I made a video of a collective reboot. The X4 in the video is the wifi master of the collective. The video starts before the reboot and runs until after the reboot complete. Breathing blue is normal operation for this time of day.

 

[BeanAnimal]

It has to be more than just loosing the internet. I lost internet on Wednesday for 6 hours and the only reason I was aware of it was the graphs had a 6 hour gap. He tank was fine

Could it be that there are combinations of user controlled parameters with (or without) the loss of internet that could cause race conditions or other watchdog reset events?


I am interested in seeing the ticket referenced by the other poster.

 

[SpyC]

I made a video of a collective reboot. The X4 in the video is the wifi master of the collective. The video starts before the reboot and runs until after the reboot complete. Breathing blue is normal operation for this time of day.

Thank you for taking the time to make these videos and share your experience with numerous hydros controllers.

 

[n2585722]

Could it be that there are combinations of user controlled parameters with (or without) the loss of internet that could cause race conditions or other watchdog reset events?


I am interested in seeing the ticket referenced by the other poster.

The watchdog may cause a reboot if a wifi signal is lost for a given time frame to reboot the wifi radio control to make sure that is not the issue.

 

[BeanAnimal]

The watchdog may cause a reboot if a wifi signal is lost for a given time frame to reboot the wifi radio control to make sure that is not the issue.

That is certainly a design flaw if the controllers reboot to try and regain wifi!

Maybe that is what is being discussed in these instances?

 

[n2585722]

That is certainly a design flaw if the controllers reboot to try and regain wifi!

Maybe that is what is being discussed in these instances?

I think if I was away from home I would want some kind of attempt to retry the connection because without it you would not have any communications while away and would have to wait until you got back to fix it yourself. Besides it only takes a few seconds to do a reboot. I am sure there is some kind of timer that has to elapse between reboot attempts if there is more than one attempt. I just posted a video of a collective reboot from before the start till after it completed and it shows the LEDs on 4 of the controllers. The larger one is normally the wifi master of the collective. If for some reason it does loose wifi then one of the others will take over that roll so if a single controller is lost only the inputs and outputs on that controller are lost to the collective so it would be more like loosing a module on the other brand controllers and not the head unit.

 

[leepink23]

I recently had no internet for 2x days due to a line being cut in our subdivision. I didn’t experience any rebooting issues as described.

 

[BeanAnimal]

I would not want a controller rebooting to try to regain wifi. To that end, I think wifi/ethernet should not be on-board. But alas, I am not the designer.

So it would appear that people are conflating "internet" being down to "wifi" being down. That answers a good bit, but still leaves me concern with regard to unexpected reboots of a life support system or the repercussions of such.

 

[n2585722]

I would not want a controller rebooting to try to regain wifi. To that end, I think wifi/ethernet should not be on-board. But alas, I am not the designer.

So it would appear that people are conflating "internet" being down to "wifi" being down. That answers a good bit, but still leaves me concern with regard to unexpected reboots of a life support system or the repercussions of such.

I have frequently rebooted controllers on mine without any issues. They will recover from a reboot very quickly and the reboot will only be the controller that decides it needs a reboot via the watchdog and not the whole collective. I have even done it when I was not home before but I have even changed settings before when not at home. The one thing I will not do without being there when I do it is updating firmware. I have downloaded and so it would be ready to update when I did get home. Also I make sure I have plenty of time after an update just in case. So far I have not had anything issues with the updates but better safe than sorry.

 

[BeanAnimal]

I have frequently rebooted controllers on mine without any issues. They will recover from a reboot very quickly and the reboot will only be the controller that decides it needs a reboot via the watchdog and not the whole collective.

There is a difference between attended and automatic. And, it sounds like if the WiFi itself is down, then all devices enter reboot mode. I don’t care how fast they come back up, this is not acceptable behavior to me.

Whatever process is running is terminated and the device reboots. There is so much to unpack here.