getting bad data for USA-MA58767-X xmltvid 56905

[karog]

For the last two days I have been getting bad data for lineup USA-MA58767-X xmltvid 56905, The Discovery Channel HD. I find several gaps in the data: one show's start time + duration is well before its successor's start time by anywhere from 30 min to a few hours. I don't see this with the DD XML data. Here is one example from the /schedules for this stationid 56905, first the metadata and then two shows with the time gap.

"metadata": {
    "modified": "2014-11-17T16:24:11Z",
    "md5": "Ot6txAbpmWcsWcDBnPG9FQ",
    "startDate": "2014-11-17",
    "endDate": "2014-11-30",
    "days": 13
}

"Programs": [
    {
        "programID": "EP015676680096",
        "airDateTime": "2014-11-18T02:00:00Z",
        "duration": 7260,
        "md5": "8Sb24Yri2lNTziQVZ7ePWA",
        "new": true,
        "audioProperties": [
            "cc",
            "stereo"
        ],
        "videoProperties": [
            "hdtv"
        ],
        "ratings": [
            {
                "body": "USA Parental Rating",
                "code": "TV14"
            }
        ]
    },
    {
        "programID": "SH017330220000",
        "airDateTime": "2014-11-18T08:00:00Z",
        "duration": 1800,
        "md5": "raBur2xKtjBI8sDquObMXg",
        "audioProperties": [
            "cc",
            "stereo"
        ],
        "videoProperties": [
            "hdtv"
        ]
    }

]

[rkulagowski]

Confirmed in the raw data; I'm opening an upstream ticket.

[rkulagowski]

This has been confirmed as an upstream ingest error that was affecting multiple stationIDs; they state that the core issue has been addressed and that any gaps will be cleared by 2014-11-21. I have code on the server side which will now find the gaps automatically after we run our ingest.

If I don't find gaps during our ingest on 2014-11-21 I will close this ticket.

[karog]

Thanks for following up on this. Today is the first day since reporting this problem that I have not seen any gaps.

[rkulagowski]

I'm still seeing gaps in our upstream schedule data (although not necessarily for this stationID) so I will leave this ticket open for now.

[karog]

Yes, today I saw gaps on several stationIDs as well and was going to report it. I am glad your gap monitoring caught it.

[rkulagowski]

The gap issue should be resolved in API 20141201.