Dashboard down?

The GraphRyder dashboard is not down, but it reports zero annotations, so it cannot build the co-occurrences graph.

The APIs from edgeryders.eu seem to work as normal: https://edgeryders.eu/opencare/annotations

@Jason_Vallet | @melancon

Back to normal but…

I actually had the issue since Friday; the server could not managed to get the annotations in time and was returning a “request time-out”. I tried again several times during the week-end and this morning and finally ended up with a response.

This behaviour also occurs when a manual request is submitted for instance:

jason@vallet-laptop:~$ wget  https://edgeryders.eu/en/opencare/annotations

–2017-05-09 10:38:47--  https://edgeryders.eu/en/opencare/annotations

Resolving edgeryders.eu (edgeryders.eu)… 2a01:488:66:1000:57e6:1559:0:1, 87.230.21.89

Connecting to edgeryders.eu (edgeryders.eu)|2a01:488:66:1000:57e6:1559:0:1|:443… connected.

HTTP request sent, awaiting response… 200 OK

Length: unspecified [application/json]

Saving to: ‘annotations’

    [    <=>                                                        ] 5,611,479   7.62MB/s   in 0.7s   

2017-05-09 10:40:58 (7.62 MB/s) - ‘annotations’ saved [5611479]

return a result in more than two minutes but the same request a few second later:

jason@vallet-laptop:~$ wget  https://edgeryders.eu/en/opencare/annotations-pretty

–2017-05-09 10:41:17--  https://edgeryders.eu/en/opencare/annotations-pretty

Resolving edgeryders.eu (edgeryders.eu)… 2a01:488:66:1000:57e6:1559:0:1, 87.230.21.89

Connecting to edgeryders.eu (edgeryders.eu)|2a01:488:66:1000:57e6:1559:0:1|:443… connected.

HTTP request sent, awaiting response… 200 OK

Length: unspecified [application/json]

Saving to: ‘annotations-pretty’

    [    <=>                                                        ] 6,868,395   8.43MB/s   in 0.8s   

2017-05-09 10:41:18 (8.43 MB/s) - ‘annotations-pretty’ saved [6868395]

only takes a second.

Because I performed many requests on Friday on the ovh server where GraphRyder is hosted, I thought that the server had locked me out to avoid a DDoS attack but the behaviour showed in the example above is not consistent with such action.

Let’s hope it is only a one time thing due to a bad alignment of the stars but if such occurrence were to appear on a regular it would become troublesome.