Gladys not broken anymore?

Hello!
It seems Gladys Plus is down again for me :confused:

Here are the details:




Do you mean « again » — do you get this error often?

I checked my analytics; I don’t have the impression there was any interruption at those times. Are you sure it’s not your local instance that was having trouble responding (aggregation running, Gladys Plus backup in progress that locks your database, etc.)?

Especially since in your screenshot we can clearly see that all requests are responding successfully, so it doesn’t look like a Gladys Plus downtime.

I assume the log you’re showing comes from your local instance; it looks more like a message parsing problem.

Hello @pierre-gilles !

Indeed I’ve already had this issue and it was back the next morning.
Locally it was running fine :frowning:
However, after my message I checked on my phone just to be sure and it was OK for gplus while on the work PC it was KO :frowning_with_open_mouth:.
This morning it’s still KO on the

Ok, are you on a supported browser (Firefox, Chrome, or Safari)?

I looked at the code, I think I understood

Is your office PC’s clock set correctly?

The error you see is that the message is seen as « expired » by the Gladys instance, the signature is too « old ». It’s a security measure to prevent « replay attacks » ( Attaque par rejeu — Wikipédia )

If your PC’s clock is wrong, Gladys will think this request comes from the past and is a « replayed » message by an attacker.

For info, a message expires after 5 minutes, so there must not be a time difference of more than 5 minutes between your PC and your Gladys instance, check that :slight_smile:

2 Likes

I am indeed on Chrome.

Indeed it must come from that because I had to make some changes to simulate different timezones..

I’ll check tomorrow to see if everything is ok :slight_smile:
Thanks for taking your precious time

1 Like

Hello @pierre-gilles I confirm that the « issue » came from there, thanks and sorry to have taken your time!

1 Like