Display Disappeared from CMS 1.7.6

During upgrading some devices today from Xibo 1.7 R56 to the our Android White Label 141 the display completely disappeared from the CMS. We have never seen this happen before.

What happened: We removed Xibo R56. Installed 141 and attempted to connect to the CMS. The device told us that it was unable to connect to the CMS.(Note this particular location does have some sort of network issue from time to time. Unfortunately, it is not our network.) We decided to reboot the device, but upon reboot the device never came back to life. Several power cycles and nothing. (We see this from time to time and just re-flash the device and all is good again. We believe it is from the units getting their power pulled and reconnected, incorrectly)

When we went to flag the unit in the CMS, it was nowhere to be found. All the other units we have upgraded to the while label never disappear from the CMS. Is there any known reason for this to happen?

I would post a log, but I can’t tell what the date and times are in the log becuse they show up like this:

2016-21-24 11::

Bug :bug:

If someone can tell me what to look for I might be able to do that.

The only way displays are deleted is if they are manually removed - if your display really have gone, then that is the only answer.

However it might be that the display has become disconnected from its display group - you could check that in the database if you have access.

You would look in the display table for your missing display, if its there then you’d look in the lkdisplaydg for the same displayId. I’m expecting you to not find anything. If that is the case you’d add a displaygroup record with isDisplaySpecific = 1 and then create a link record between the two.

Dan,

I am the only one that has access to the particular server where the display disappeared from. I certainly did not remove the display from the system. I looked at the database and I also do not see it there.

We will just re-add the display, but still strange that it happened in the first place.

Any idea why the log date and times are screwy?

Its odd because the only delete statement for a display happens during the manual delete process - i.e. there is only one delete statement and its only in one place, if you see what I mean?

Regarding the times - what do you have as your date format in settings? If you are using international dates then please remember that the format changes. (although in 1.8 this has been reworked and resolved)

Really strange.

Dates:

International Dates is on, and the format is

YYYY-MM-d HH:mm:ss

If we change this, then the dates on the Forcast.io do not show as expected. So I guess for now it is a either or. We really love the vast number of improvements in 1.8. Not that 1.7 was bad.

Great to hear!

I can look at that:

1 Like