* If messageId changes and network ID doesn't, remove 'old' message_id
* Boot nodes/connection when no recent messages received for it
* Separate task needed for soketto recv to avoid cancel-safety issues with new interval
* Wee tidy up
* cargo fmt
* Add some logging around node adding/removing
* Another log info msg
* a bit of tidy up
* bump stale node timeout to 60s
* adding dual deployment gitlab-ci
* adding the vault tempalte to the build stages
* fix typo
Co-authored-by: James Wilson <james@jsdw.me>
* fix typo
Co-authored-by: James Wilson <james@jsdw.me>
* adding variables to deploy-production stage
* fix
Co-authored-by: James Wilson <james@jsdw.me>
* Make soak tests work again with genesis hash subscribing
* cargo fmt
* derive hex from hash
* actually compile code
* fmt
Co-authored-by: David Palm <dvdplm@gmail.com>
* Refactor `Tooltip` into function component with hooks
With using hooks, we don't need to maintain the checking in sCU and also
don't need to assign the value of `ref` manually.
* Show block number beside the tooltip instead of its children
* test running tests
* Add delay so that core knows about node before feed subscribes to limit chance of race
* move delaye to the right place
* Don't do expensive docker step untill we will be pushing the image
* docker test skipped as hoped, so push to 'true'
* just remove docker steps in github CI entirely since they aren't needed by anything (gitlab CI does this stuff now)
* run CI on pull requests too to catch PRs from forks
* A second attempt to avoid the flume memory leak, since the first didn't actually work
* No luck with second flume fix, so revert to futures::mpsc in a few places)
* cargo fmt
* Add a comment to cover use of into_stream
* Tweak rolling_total test to also confirm capacity doesn't go nuts
* Use Jemalloc
* Avoid flume's into_stream and use a workaround for now
* cargo fmt
* Improve comments now that there's an issue to point to
* Handle subscription by hash in the frontend
* Forward-ported backend changes
* Fix unit tests
* Remove unused `chains_by_label`
* fmt
* Updated but failing E2E tests
* subscribe by genesis hash in tests
* fmt
* Copy `BlockHash` instead of returning a ref
* Pin chains by genesisHash
Co-authored-by: James Wilson <james@jsdw.me>
* adding values-parity.yaml to the project again
* removing the custom values file
* replacing the namespace with variable
* removing the default configuration for frontend nginx
* fixing the CI
* Upgrade TS from 2.9.2 to 4.4.2
* Avoid mutating a readonly array
`list` is a readonly array but `sort` will mutate the array in-place.
In newer version of TS, it can catch this unexpected behavior.
To avoid mutating a readonly array, we can sort with a new array with
either `[...list]` or `list.slice()`.
* Fix arguments of `window.location.reload`
The `reload` function type definition in `lib.dom.d.ts` takes 0
arguments now. So, we cannot pass a boolean to it.
Also, based on the documentation on MDN, Firefox is the only browser who
supports passing a `forceForget` boolean to `reload` function.
So, I wonder it should be fine to remove the boolean argument and align
the behavior to other browsers.
Ref: https://developer.mozilla.org/en-US/docs/Web/API/Location/reload
* Beef up error reporting of IP and location info
* Tidy up error reporting after some manual testing of it
* Don't cache erroneous locations; try again when asked again
* cargo fmt
* adding values-parity.yaml to the project again
* removing the custom values file
* replacing the namespace with variable
* removing the default configuration for frontend nginx
* Only remove the feed that disconnected to not break the rest...
* use multimap struct to avoid sync issues between feed and chain
* add a remove test, too
* cargo fmt
* fix name of test
* move multimap to common so we can doctest it and add 'unique' to name
* cargo fmt
* Return old key if value moved to make uniqueness more obvious
arshamparity