We were running on AppEngine but recently moved over to Managed VMs. For some reason AppStats is no longer available? We just get a 404 not found error when browsing to our appstats URL. Is appstats not supported on Managad VMs? If not, is there a way of isolating poorly performing endpoints within our application?
One way to isolate poorly performing endpoints is to use the advanced filter search in the GCP Logs Viewer. It is a little hard to find at first.
To get there, in your Google Cloud console, navigate to Logging for your project. At the right of the text box for "Filter by label or text search" you will see a small dropdown arrow. Click that and select "Convert to advanced filter". This will allow you to write your own sql-ish query where you can find requests that took longer than n to complete.
For example, add the following to the filter:
protoPayload.latency>"0.300s"
This will return a list of all requests that took longer than 300 milliseconds to process. If you have Cloud Trace enabled, you can click on the request response time to see the timeline for the individual service calls.
Related
I have an GAE app PHP72, env: standard which is hanging intermittently (once or twice a day for about 5 mins).
When this occurs I see a large spike in GAE dashboard's Traffic Sent graph.
I've reviewed all uses of file_get_contents and curl_exec within the app's scripts, not including those in /vendor/, and don't believe these to be the cause.
Is there a simple way in which I can review more info on these outbound requests?
There is no way to get more details in that dashboard. You're going to need to check your logs at the corresponding times. Obscure things to check for:
Cron jobs coming in at the same times
Task Queues spinning up
By log sampling I mean that not all the logs are saved to Stackdriver but just a sample of them. Useful for high traffic applications.
I've searched but the only mention I see is here in the context of App Engine Flex and OpenAPI.
There is also an option to sample logs at query time here.
Log sampling is also mentioned as an option for Trace clients here but that doesn't seem applicable to App Engine.
Stackdriver does not have built in support for sampling logs but you can specify a log exclusion filter which will allow you to remove log entries based on a criteria.
If you choose to exclude everything and specify a sampling criteria I think you can achieve this:
The link below explains this in a bit more detail:
https://cloud.google.com/logging/docs/exclusions
I want to check latencies of RPC every day about CakePHP Application each endpoints running in GKE cluster. I found it is possible using php google client or zipkin server by reading documents , but I don't know how easy to introduce to our app though both seem tough for me.
In addition, I'm concerned about GKE cluster configuration has StackDriver Trace option though our cluster it sets disabled.Can we trace span if it sets enable?
Could you give some advices?
I succeeded to send gcp's trace api in php client via REST. It can see trace set by php client parameters , but my endpoint for trace api has stopped though I don't know why.Maybe ,it is not still supported well because the document have many ambiguous expression so, I realized watching server response by BigQuery with fluentd and DataStudio and it seem best solution because auto span can be set by table name with yyyymmdd and we can watch arbitrary metrics with custom query or calculation field.
I like that I can use the Logs API (described here: https://cloud.google.com/appengine/docs/java/logs/) to programatically access and display app & request logs as I see fit--it's great.
Now that I'm using Managed VMs on AppEngine you can see on the Admin Console Logs Viewer that there are a ton of additional logs--including in my case a custom log which I found I could include in the viewer (decribed here: https://cloud.google.com/appengine/docs/managed-vms/custom-runtimes#logging).
My question is: Is there any way I can use the Logs API (or other pipelines already built?) to access these logs? My Managed VM module includes several components which could produce logs that I want to view:
App logs -- I can get these! No problem here.
Custom log files created by background processes I kick off in _ah/start (like "my_custom_1.log" in the screenshot)
STDERR & STDOUT from my background processes
Relevant Managed VM logs (e.g. for when an instance was restarted due to bad health... other system events like normal restarts?)
Basically I want "the total picture" at the instance level. Anyone tried to tame Managed VMs in this way with success? I'm not looking forward to rolling my own solution. And I wouldn't even know where to start on the problem of capturing STDERR and STDOUT. Any help appreciated.
There is a difference between App Engine logging and Google Cloud logging. Some of the Managed VM logs go to both, but much of it only goes to cloud logging.
Until recently there was not an API to read Cloud logs, only to write them. However, there is a new v2 beta API: https://cloud.google.com/logging/docs/api/introduction_v2
To do things at an instance level, entries in Cloud logging should have metadata set to denote which VM they came from. Both of these values seem to vary on logs from my VMs:
compute.googleapis.com/resource_name
compute.googleapis.com/resource_id
I have a web app and I include google analytics. My active users seems to of spiked and I'm incredibly paranoid that I'm somehow double counting my analytics.
Is there any way to see if I'm doing this?
As Nebojsa mentioned, you can inspect source and search for ga.js or analytics.js to see if it's in your application twice.
Look through your source code to see if you have the partial rendering in multiple places (ex. header and footer)
Setup another Google Analytics account and test locally if its double counting your visits. See this post for setting up GA on localhost
Use the Google Analytics Tag Assistant to verify that everything is setup correctly. It will tell you if there are any implementation problems, including multiple tracking codes. It also helps with Adwords, re-marketing and other Google product scripts.
Use the Google Analytics Debugger. This would probably be the most helpful to determine if a single hit is being double counted as it walks you though every single function call the analytics urchin makes.
just open source in the browser and look-up for code of analitics...par example
_gaq.push(['_setAccount', ...