To reduce the risk of the IPR DoS the client:
- Lower the timeout until the IPR will disconnect a client
- Reduce fewer surbs at a time. Large surb requests increases the
latency until all fragments in the response have been delivered. The
efficiency gains of having large surb requests dimishes quickly for
large sizes as well
* IPR v8 request/response types
* Remove signature for when we use sender tags
* Remove unused
* Address some review comments
* Update license to GPL-3.0 for IPR
Since the IPR can run as a binary, make sure it's license is GPL-3.0
* update cargo deny
* Add back support for v6
* Tidy responses
* Clippy
* Fix compilation
* Conversions
* Conversions
* Split response conversion
* request split
* Complete conversion switch
* Remove commented out code
* rustfmt
* Remove unused conversions
* Remove unused TryFrom
* use from
* fixed bearer auth for prometheus route
* basic prometheus metrics
* added rates on global values
* improved structure on the prometheus metrics
* added additional metrics for ingress websockets and egress mixnet connections
* some channel business metrics
* fixed metrics registration and added additional variants
* added counter for number of disk persisted packets
* counter for pending egress packets
* counter for pending egress forward packets
* clippy
* fixed bearer auth for prometheus route
* basic prometheus metrics
* added rates on global values
* improved structure on the prometheus metrics
* added additional metrics for ingress websockets and egress mixnet connections
* some channel business metrics
* fixed metrics registration and added additional variants
* added counter for number of disk persisted packets
* counter for pending egress packets
* counter for pending egress forward packets
* clippy
* removed mnemonic from gateway config struct
scaffolding for common mixnet listener
running verloc unconditionally in a nym-node
remove filtering by mixnode
extracted verloc to separate crate
integrated nym-node-http-server more tightly with the binary
most logic for handling forward packets
running all mixnode-related tasks natively inside nymnode
removed gateway storage trait in favour of the only concrete implementation
most logic for handling final hop packets
using nym-node owned socket listener for gateways
utility for sending plain message through mixnet + gateway fix
using common packet forwarding in both modes
nifying nym-node metrics
reproduce behaviour of the console logger
cleaned up cli args
redesigned gateway tasks startup procedure
removing dead code
scaffolding for old config v6
config migration
implemented MixnetMetricsCleaner
* clippy
* require entry/exit for wireguard
* removed dead code in migration code
* updated config template
* use custom user agent for verloc queries
* fixed premature shutdown of gateway tasks
* hidden nym-api flag to allow illegal node ips
* experiment: final hop handing with wireguard
* added additional startup logs
* typo
* fixed legacy stats endpoint data
* additional logs
* apply review comments
* fixed local testnet manager
* Add ipv6 in configs
* Make v4 latest
* Fix linux
* IPv6 prefix in config
* Fix template of private ip
* Fix clippy
* Fix v6 cidr
* Move from 2001:db8::/32 to fc00::/7 addresses
* Fix version number on conversion
* added explicit SP suffix to started tasks
* added 'GatewayTopologyProvider' that always injects itself into the network
* use the new topology provider to bypass described bootstrapping problem
benedettadavico