When the client side metrics were added constLabels
got missed out. Without this, we cannot do the high
level grouping by rings.
https://mattermost.atlassian.net/browse/MM-62901
```release-note
NONE
```
* MM-61865: Add mobile client content load network metrics
* added new common label
* renaming from MobileClientContentLoad to MobileClientNetworkRequests
* content_load_group => network_request_group
* refactor more NetworkRequest-* changes
* replace contentLoadGroup to networkRequestGroup
* new metrics elapsedTime
* Refactor urlCount to totalRequests
* add averageSpeed metric
* replace contentLoadGroup with networkRequestGroup
* use h.Labels vs commonLabels for network_request_group
* add agent
* add effective latency metrics
* add total parallel requests & total sequential
* mocks generated by mockery
* did a bit of cleanup and sorting
* formatting
* updated the AcceptedNetworkRequestGroups
* cleanup and sorting
We create a custom histogram metric that logs the userID
when the observed value is greater or equal to the last bucket value.
This allows us to start tracking the slowest users of a system
while at the same time not polluting the Prometheus metrics
by storing a userID for every observation.
https://mattermost.atlassian.net/browse/MM-61887
```release-note
NONE
```
Page load is one of the metrics that we track and present
to MLT. However, in its current form, it is not very
actionable because it also contains the network latency.
We split the whole metric into these parts:
startTime
|
responseStart = TTFB
|
responseEnd = TTLB
|
domInteractive = Start of processing phase
|
loadEventEnd = Load complete
This gives us better visibility into exactly
which phase in the load process is slow.
I have experimented with other metrics like
- domContentLoadedEventStart
- domContentLoadedEventEnd
- domComplete
and observed that they do not have sufficient
gaps in the timespan to have any relevance.
Additionally, I have moved TTFB from being a
web vitals metric to being tracked from the performance
metrics to remain consistent with the other navigation
metrics measured.
Lastly, I took this chance to improve some of the
validation errors that we threw to include more
context into the input that was passed and why
does it fail.
This also meant that I had to change the tests
to check for error strings rather than direct
errors which is a bad thing, but I don't think
it's worth the effort trying to have named error
variables for all of them.
https://mattermost.atlassian.net/browse/MM-61886
```release-note
NONE
```
* Add metrics for mobile versions snapshots
* Add notifications disabled and fix lint
* Address feedback
* Verify all references to JobTypeActiveUsers
* Fix typos
* Improve platform values
* Add test and MySQL support
PR: https://github.com/mattermost/mattermost/pull/27204 fixed the issue
where `invalidateWebConnSessionCacheForUser` was not broadcasted across
the cluster. However, we became too conservative and started to invalidate
the session cache as well.
But the session cache only contains the session details for the user
and the team membership. There is no channel membership related info.
Therefore, we avoid that and improve performance as well.
https://mattermost.atlassian.net/browse/MM-60478
```release-note
NONE
```
Co-authored-by: Mattermost Build <build@mattermost.com>
This came up as part of the Redis work. I noticed that a lot of new
invalidation events were missing. This is important to understand
where invalidations are coming from.
https://mattermost.atlassian.net/browse/MM-59826
```release-note
NONE
```
Co-authored-by: Mattermost Build <build@mattermost.com>
* Improve mocking of imported resources in unit tests
We have Webpack configured so that, when code imports an image or other resource, the code gets the URL of that image. Jest now matches that behaviour which is needed because React Testing Library would previously throw an error.
* Polyfill ResizeObserver in all unit tests
* Ensure haveIChannelPermission always returns a boolean value
The previous code could sometimes return undefined. While that should behave the same in practice, it can cause React to print prop type warnings
* MM-58535 Add region label to LCP metrics
* MM-58535 Upgrade web-vitals and add INP attribution
* Change new labels to use snake_case
* Remove replaceGlobalStore option from renderWithContext
I was going to add this in case any tests failed with this option set to false, but after running those tests, that's not the case. I'm going to remove this as an option since it seems more likely than not that anyone using RTL would prefer to have this on.
* Add platform information for push notification metrics
* Address feedback
* Add the client platform returned by the devices to the normalize function
* Add "no platform" platform label to distinguish from unknown
* MM-57882 Add metric for the time it takes to open the Threads list
* Clean up mark because the starting mark may be missing
* Pass global threads load duration to Prometheus
* Update mocks
* Ensure your own posts are never ACKed
* Don't ACK mobile websocket notifications
* Add counter for the unsupported Desktop Apps
* Count only push messages when checking for acks
* Fix generated
* Add tests, fix comment
* Fix help string
* Check for nil session
* [MM-57066] Add metric counters for notification events
* Some small changes
* Account for Metrics() sometimes being nil
* Fix test (again)
* Fix more tests
* A few changes from testing - added success counter
* Missed a mock
* Lint
* Add feature flag for notification monitoring
* Expose metrics under a source available license
* do not assume Cluster()
* allow metrics if licensed or dev
* temporary vet override
* simplify BULID_TAGS handling
* auto clean old imports.go
* fix license listener
* e2e test metrics & license semantics
* update from enterprise
* switch back to mattermost-govet/v2@new now
* update metrics from upstream
* Update license_spec.js
Co-authored-by: Saturnino Abril <saturnino.abril@gmail.com>
* Update license_spec.js
Co-authored-by: Saturnino Abril <saturnino.abril@gmail.com>
* Update e2e-tests/cypress/tests/integration/channels/enterprise/metrics/license_spec.js
Co-authored-by: Saturnino Abril <saturnino.abril@gmail.com>
* Update e2e-tests/cypress/tests/integration/channels/enterprise/metrics/license_spec.js
* split up specs
* require/delete license earlier in e2e test
* expanded expect to debug failures
* more logging
* Revert "more logging"
This reverts commit 0bc513fd926c07a0c6050f4b4c742e856d8dc435.
* e2e: try deleting license first
* update from enterprise
* toggleMetricsOn to work around license delete
* eslint
* ensure admin before deleting license
* update from enterprise
* updates from enterprise
* fix cypress logging
* temp: log at DEBUG for Cypress tests
---------
Co-authored-by: Saturnino Abril <saturnino.abril@gmail.com>
Co-authored-by: Mattermost Build <build@mattermost.com>
