[APM] "View job" link from latency charts leads to a malfunctioning page

[dgieselaar]

When clicking on the "View job" link from the overview latency chart, the user sees this screen:

This is not very helpful, and it also triggers an infinite loop of requests. At first, the errors return 200s, but eventually they error out as ERR_INSUFFICIENT_RESOURCES.

I'm not sure if our link is wrong.

[elasticmachine]

Pinging @elastic/apm-ui (Team:apm)

[dgieselaar]

Spoke to the ML folks and the link is broken:

• pause should be either 'true' or 'false'
• entities are not being passed because transactionType is empty. We can probably fix that by getting transactionType from APMServiceContext.

[dgieselaar]

The link on the service map should also be fixed, and possibly other places.

[cauemarcondes]

is it something that ML team recently changed? @sqren fixed some of the ML links on this PR #85976, maybe we missed that.

[dgieselaar]

@cauemarcondes I'm not sure about pause. But we are not passing service.name and transaction.type to the ML link helper function, so that's something that we should fix.

[cauemarcondes]

I fixed to pass the service name and transaction type. But it triggers an infinite loop because we set pause:true, value: 0, this is the default setting defined on kibana advanced settings. On APM we hardcoded pause to false because of this issue, maybe we should do the same here?! cc. @formgeist

[dgieselaar]

@cauemarcondes I think the ML folks will be able to help you here. I just spoke to Dima about it.