You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
So all I'm doing is integrating at the axios level in ONE place. There is nothing downstream adding spans. This is simply ONE place where I've added axios (web-service call) instrumentation. In this instance this is a NestJS service (javascript of course) that polls the same endpoint every 30 seconds looking for notifications. So every 30 seconds it does
GET /api/v1/notifications
So what I expect is that every 30 seconds I have a brand new trace with a single span because i have no other instrumentation anywhere. So every web-service call is it's own trace.
But what I get is EVERY span is logged with the same traceId forever. So it seems there is like only one ID generated and used as a traceID from a single instrumentation point - and every single span logged from that instrumentation point will ONLY every have this one ID.
Then on top of that, every subsequent call, even though they are totally unrelated and distinct calls, the previous spanID is set as the parentID of the span... So ONE trace for every single call out of this service and one has a child, and he has a child and so on and on forever.
Here is the log of 3 spans being logged. You can see they all use same traceID somehow, and each spawns a child, even though these should all just be discrete traces - these spans have nothing to do with each other.
Can anyone explain
a) why would every single flow through an instrumnation part reuse the traceID?
b) what triggers the system to think that the previous span is the current's span's parent when they have no such relation?
So all I'm doing is integrating at the axios level in ONE place. There is nothing downstream adding spans. This is simply ONE place where I've added axios (web-service call) instrumentation. In this instance this is a NestJS service (javascript of course) that polls the same endpoint every 30 seconds looking for notifications. So every 30 seconds it does
GET /api/v1/notifications
So what I expect is that every 30 seconds I have a brand new trace with a single span because i have no other instrumentation anywhere. So every web-service call is it's own trace.
But what I get is EVERY span is logged with the same traceId forever. So it seems there is like only one ID generated and used as a traceID from a single instrumentation point - and every single span logged from that instrumentation point will ONLY every have this one ID.
Then on top of that, every subsequent call, even though they are totally unrelated and distinct calls, the previous spanID is set as the parentID of the span... So ONE trace for every single call out of this service and one has a child, and he has a child and so on and on forever.
Here is the log of 3 spans being logged. You can see they all use same traceID somehow, and each spawns a child, even though these should all just be discrete traces - these spans have nothing to do with each other.
Can anyone explain
a) why would every single flow through an instrumnation part reuse the traceID?
b) what triggers the system to think that the previous span is the current's span's parent when they have no such relation?
[Nest] 83298 - 11/16/2019, 8:19:36 PM [
{
"traceId": "39ac8b70786e7660",
"parentId": "57bfbbd396033f90",
"id": "5d3a212eef248142",
"name": "get",
"kind": "CLIENT",
"timestamp": 1573957175933000,
"duration": 209000,
"localEndpoint": {
"serviceName": "bis-app-web"
},
"remoteEndpoint": {
"serviceName": "hub-service"
},
"tags": {
"http.path": "/hub-api/notifications/all",
"http.status_code": "200",
}
},
{
"traceId": "39ac8b70786e7660",
"parentId": "5d3a212eef248142",
"id": "48146177b16174eb",
"name": "get",
"kind": "CLIENT",
"timestamp": 1573957205931000,
"duration": 223000,
"localEndpoint": {
"serviceName": "bis-app-web"
},
"remoteEndpoint": {
"serviceName": "hub-service"
},
"tags": {
"http.path": "/hub-api/notifications/all",
"http.status_code": "200",
}
}
{
"traceId": "39ac8b70786e7660",
"parentId": "48146177b16174eb",
"id": "b75e403e79d0d1b7",
"name": "get",
"kind": "CLIENT",
"timestamp": 1573957235240000,
"duration": 180000,
"localEndpoint": {
"serviceName": "bis-app-web"
},
"remoteEndpoint": {
"serviceName": "hub-service"
},
"tags": {
"http.path": "/hub-api/notifications/all",
"http.status_code": "200",
}
}
The text was updated successfully, but these errors were encountered: