Data Instrumentation

Hey there!Our team noticed today that we have had a huge increase in page views over the last month from visitors in Singapore with an unusual user agent (Mobile Safari on Android). I did a reverse IP lookup and found it belonged to petalbot-114-119-130-251.petalsearch.com - which appears to be the web crawler for Huawei’s Petal search engine.This leads me to believe that Amplitude’s servers (or library) is not excluding this bot from its data. Currently I’m filtering the 114.119.xxx IP range out of our metrics to fix this, but curious if there is a better approach or something that Amplitude should do to fix this at their level?We are not seeing any similar issues in Google Analytics, and a quick search reveals that Mixpanel encountered and fixed a similar problem about a year ago: https://github.com/mixpanel/mixpanel-js/issues/316I could also try updating our Amplitude JS library but I’m not sure if that will help or not - anyone else experiencing this issue?  

Currently we are doing experiment using Amplitude ID as our bucketing key. We learned from https://help.amplitude.com/hc/en-us/articles/115003135607-Track-unique-users-in-Amplitude#h_4caf2577-336a-4b32-8a6a-c7245b5f2364 that a user would be assigned to same Amplitude ID after anonymous events. However, after we did some tests, the actual behavior was Amplitude would create a new Amplitude ID and than merged them. It may cause variant jumping when the first a anonymous user logging in. How can we solve this problem?    

Hi Team, We had 3 questions that needed clarification:App install and uninstall event instrumentation → Can someone guide us step by step on how to instrument this? or an example of a code snippet? User update behavior → When a user updates the app, does the amplitude ID, user ID and device ID hold even after the update? Log out → Similar to 1, can someone share a guide to instrument this? Looking forward to your reply.

Since iOS 16 was released, we’ve been seeing a growing number of crashes related to CFBundle and CFDictionary, including with the Amplitude SDK. We are completely stumped why this is happening and have been unable to locate other reports of similar crashes. Is anyone else seeing crashes like this?Amplitude SDK is Amplitude-iOS (4.10.0) installed via CocoaPods.

Hello, We have a kids app (iOS and Android). We used React Native SDK (Legacy) but suddenly we stopped to receive any events. We have't had any app updates.We decided to update SDK and use React Native SDK (Beta) but Google Play rejected our app because location permissions: [!] Google Api Error: Invalid request - You have declared that your app targets under 13 age groups, but your app asks for location permissions.I don't see an opportunity to disable COPPA Control in Beta version as it was in legacy SDK. Could you help me how I can fix this issue? 

I’d like to transfer all project data (events, user propeties etc.) from one Amplitude account to another. But the project that I’m going to transfer has more then 80 million events. Since Amplitude considers these events as new, there will be an excess of 10 million events (I have a Starter plan). Will Amplitude consider this exceeding 10 million 8 times and block my account, or will it be considered as a one-time exceeding 10 million events?

I have two charts: one showing sign ins and another usage of functionality once signing into the product. I have a user that appears on the functionality chart that does not show in the sign-in chart. This is impossible as they must have signed in to use that functionality. The segmented by/performed by/grouped by sections are both the same in each chart with the only difference being the event. Any ideas on what could cause this? 

My application triggers an event when there is an API error. One of my services was disabled, and that caused many retries, each retry triggered an event, so in the end, each user could have thousands of errors, and that ended up being a couple millions of events.Is there a way to clean up all these events to prevent them from counting towards my event quota?

Hi all, We have four projects in our Amplitude account. Has anyone been able to see event data from multiple projects in their BigQuery Export… I’m only seeing event data from one Project. Is that expected? How do you filter data sets in BigQuery by Project (e.g. App Product 1 UAT, App Product 1, App Product 2, App Product 2 Devel)?

Resurrected users - which user count as resurrected? What he should o to have this label? 

Hi all, anyone has experience integrating Amplitude with Zendesk? We would like to track the who views our Zendesk article which currently not provided by Zendesk. Thank you.

In Segmentation, if I pick any event and set “Historical Count”:”1st” there’s a noticeable difference between unique users and total events. Any ideas why? It seems that if a user can only be counted once anyway (because it's the user's first time firing the event), that the unique and total might be the same. Any ideas? Thanks. 

Hi Amplitude community, I wonder of this use case. Let’s say that I need to know if a user a subscriber and want to put it to user_properties. this is the case A user do::Mon: user joined Wed: user subscribed Sat: user unsubsribedif let’s say I user HTTP API V2 to update the user properties called `is_subscribe`I don’t want the user to have any `is_subscribe` so not doing anything update `is)subscribe` to a string: “yes” (here is my question) -- I want to revert to the first state So for 3, what value that I have to use for HTTP API V2 ya? I know that the value of non existed properties is (none). With that being said, if I update `is_subscribe` to a string “(none)”, will this be merged to Amplitude value of <non-existent> / <not provided>? Thanks!

Hi!I try to solve a problem. The amplitude does not receive events. The problem concerns android and ios app version. The problem occurs when passing frontend application events. There are a couple of backend events that the amplitude accepts. Events are not blocked, not filtered, not hidden. Connection software has not changed and not update. All others analytics systems(not amplitude) receive and display events correctly. Last 3-4 days "Total Event Volume" graphic is in "flat". Application is available to user  

I got this message. Amplitude can’t recieve any data. The issue persists for 1 day.So I found this articleI check amplitude status, but there is no downtime for 1 day. Is there any bug in my project? I need help.  

Hello!I have a problem in Amplitude. We have user for whom on ios events are not tracked  And also some other user for whom client events are partly not tracked  Most of these users are on ios. What might be the possible problem?

Hi team,Can I set the `time` key to be newer than `now`?For example, `now` is 2022/Dec/20, I would like to set the `time` for my event to be 2022/Dec/30.As I tested, the furthest timestamp is 3 days from `now`, it’s 2022/Dec/23. And I can see my event in the Event Explorer.Please help. 

TL;DR:It seems like there would be an easy to to find the data for a specific event, where the properties for the event are defined in the segmentation chart above. -- Example:I have an event called “points_transaction” where the property transaction_source can be from various sources (A, B, C, D, E...). I am trying to find the (raw) data for an event where transaction_source = “B”, but I might be doing it the hard way. Create a segmentation chart Click a point on the graph Select “View User Streams” Pick any user Filter by event = “points_transaction” It's not possible to filter for an event + property (where source = “B”), so finding the specific transaction I’m looking for can take a while as I go through them one at a time, looking for source = “B”  Look at the event properties and note the time stamp for the desired event.  Click user lookup in top right Find the event again on the user’s event stream under user lookup.  Switch event details view from “Info” to “Raw” to view the

 In my react web app, I only installed amplitude marketing-analytics-browser SDK.(amplitude-ma-ts)But when i saw user look up, there is events which have library properties ‘amplitude-ts’ attached screenshot is what i tested. I fired 2 events consecutively in the same environmenet(same device+same browser). But their library properties are not same. Is this bug? or intended?   

Shopify’s client-side integration looks to be sending the email in lowercase while the server-side integration does no case transformation.This is a problem because:Amplitude User IDs are case-sensitive. If you set a user ID in a different case, Amplitude will track two separate profiles for the same user.

Hello, When I login to my domain I get an error that says “We failed to talk to the server. Reloading the page might fix the problem.” (screenshot attached)and I am unable to do anything else. I had no issues getting to my dashboard a few days ago but now I am unable to do so and am stuck on this page.  I have tried on multiple devices, multiple internet browsers, and am still unable to get pass this issue.  Please advise. Best,Jacob 

Hi!When I create “simple” GA4 destination syncs, everything looks fine. “All events” or one particular event - both options are working.But when I try to make sync a bit more complicated, I can’t change Status from disabled to enabled, or update mappings. UI shows me this: Sync failed to update. Please try again or contact support.And POST request to this URL: https://analytics.amplitude.com/t/graphql/org/MYORGNUMBER?q=dataIOUpdateSync returns internal server error:{ "errors": [ { "message": "Internal Server Error" } ], "data": { "dataIOUpdateSync": null }}Also, if I change default mappings from the device_id & user_id to something else, I can’t go back to the defaults :). I.e., if I chose City instead of device_id, I can’t find & set back the device_id - it’s just missing in the property list.

Hey everyone,When trying to view our users in this funnel we only see 12 IDs instead of 42.Why is this the case?Thank you for your help. 

Dear community, I have the following question:We have created an event property named duration according to this description in order to measure the time that users spend in our app on specific content types.Now we would like to create a histogram in order to see the distribution of the value of those, but summed up, meaning that I need a PROPSUM distribution chart where I would see how many users spent how much time in our app in a given time period. I found this demo chart that comes close to it, but only shows the distribution of the values at a certain time and not the SUM of those. Is there any workaround to get there? I know I could do this with the Session Chart maybe if I set start and end event differently and select length distribution, or maybe the funnel chart with Time to Convert but neither looks like a very clean way to do it. Would be great if anyone could help, many thanks in advance! Best, Asja 

I’m using this dashboard REST API to export a data table. The endpoint returns the values and the row labels but not the column labels (ie. the event names and segment names). It instead contains a 'valuesMetadata' entry that’s a list of objects with ‘metricIndex’, ‘segmentIndex’ which have integer values 0, 1, …Is there a way to get the metric name and segment name from these indexes. The indexes don’t seem related to any global ID as they always start from 0 irrespective of the events chosen in the data table.