RE: [ima-sdk] Test!

[IMA SDK]

Hello R Baudisch,

 

Michael here from IMA SDK Team. Thank you for reaching out to us.

 

It appears that you have opened up a new thread without enough context. Would you be able to confirm if you are having any issues or have any concerns with the SDK?

Regards,

Michael Angelo IMA SDK Team

ref:_00D1U1174p._5004Q2aR6cf:ref

[R Baudisch]

I'm still in trouble with omweb and IMA sometimes:
https://github.com/InteractiveAdvertisingBureau/Open-Measurement-JSClients/issues/25

[IMA SDK]

Hello Rene,

 

Thank you for responding back to us. It appears that you did not receive my last response so I am reposting it here.

 

I asked if it would be possible for you to share any test page where we can reproduce the issue? Kindly include the navigation steps we need to make in order to replicate it using the test page. After submitting, we will raise it to the rest of our team to get their insights as well.

 

If you are having trouble receiving private responses from us, please check your spam messages.

Regards,

[R Baudisch]

Hi Michael,

I also raised this as an issue to the IAB:
https://github.com/InteractiveAdvertisingBureau/Open-Measurement-JSClients/issues/25

Within a few reloads you should be able to get the error on e.g.:
https://www.bild.de/news/ausland/news-ausland/mallorca-droht-streik-chaos-zum-ferienstart-easyjet-ryanair-und-lauda-air-betrof-80473646.bild.html

It is some play-together that results in this errors here I think.
In this case the videoplayer used is bitmovin, I used videojs7 and had the same error, so it doesn't seem to be player related.

MfG
René

[R Baudisch]

Let me add, the messages leading to the errors you will see on this bild.de page, are coming from dfp display ads.

[IMA SDK]

Hi René,

I work along with Micheal. The issue has been flagged to a wider team. As per checking, the team were able to replicate the reported behavior and is now under investigation. We'll update this thread the soonest the team share their findings.

Regards,

Teejay Wennie IMA SDK Team

ref:_00D1U1174p._5004Q2aR6cf:ref

[R Baudisch]

Hi there,

any news on this?

Best 

René

[IMA SDK]

Hi René,

Looking into the this it seems to be an implementation issue.

Adding the following lines to the adsRequest resolved the error:

 adsRequest.omidAccessModeRules = {};
 adsRequest.omidAccessModeRules[google.ima.OmidVerificationVendor.OTHER] = google.ima.OmidAccessMode.DOMAIN;

You get more info on this from the OM SDK HTML5 guide. It looks like the tag linked is using doubleverify, which can be specified as a vendor as well (see the OMID verification vendor values).

Regards,

Aryeh IMA SDK Team

ref:_00D1U1174p._5004Q2aR6cf:ref

[R Baudisch]

Hi, I do not use IMA so I can't access google.ima.OmidVerificationVendor.OTHER.

I'm doing adrequests by https://imasdk.googleapis.com/pal/sdkloader/pal.js 

[IMA SDK]

Hi René,

I brought your comments to my team. We will get back to you as soon as possible.

[R Baudisch]

Please let me ad, it does not depend on where the ad is coming from, the error also happens without pal.js included.

I have implemented OMID into a videojs7 fork also containing the MOL-VAST-PlugIn.

Through an adcall I then got some VPAID creatives from DFP that loaded an imabridge.js (or similar).

This imabridge installs OMID but OMID is already on the page.

While then trying to send messages to the omid framework errors occur, 

mainly in the formerly first registered OMID framework.

These ads are brought by SSPs like the TradeDesk, so it doesn't matter what adserver initiated the call stack,

it could be GAM by IMA, by PAL, could be Xandr, SmartAdServer, that doesn't mind, as the TradeDesk is connected to all of them.

[IMA SDK]

Hello René,

 

Thank you for responding back to us. Our team have shared their feedbacks below:

 

I believe this error message starting with "The source window of session client post messages" is usually harmless, so it should not be a cause for concern for the publisher.

 

It is fixed in the latest version of OM for Web so that it will no longer print unless there is a real problem, and I have yet to see any example of it indicating a real problem. However, that version of OM for Web (1.3.35) has not been fully deployed to all omweb-v1.js URLs. For example, the fix is not yet included in the omweb-v1.js hosted by Google for use by IMA, but I'm hoping it will be in the next few days.

 

Unlike IMA, PAL does not include any omweb-v1.js script, so I am not sure which version of omweb-v1.js this publisher is using. Do they host omweb-v1.js themselves, or is some ads SDK on the page including it? If I could see the full URL of omweb-v1.js that is printing the error messages, I can determine when it might be fixed.

 

But the main point is that the error message is almost always a false alarm, so the publisher can ignore it unless they are seeing issues with their viewability metrics as measured by whichever measurement provider they are using.

 

Let me know if you have further questions.

Regards,

Michael Angelo IMA SDK Team

ref:_00D1U1174p._5004Q2aR6cf:ref