From divaexchange:

I2P community related updates:

a) a locally installable I2P(d) testnetwork is available in the public domain. Here is the link to github: https://github.com/h-phil/i2pd-testnet-kubernetes . I2Pd (C++) is supported, I2P java not. The author of the testnet might or might not fix that issue. So: PRs are welcome for those who need Java I2P support. This kurbernetes-based testnet (which scales very well) is a side effect of the I2P de-anonymization study (academic work) which has been finished mid of January 2024 (see below).

b) Independent (of the I2P[d] developers) de-anonymization study (academic research, sponsered by diva.exchange, done at Lucerne University of Applied Science, Switzerland): results in a nutshell “there have been no patterns found, using passive network surveillance technologies, to relate a Lease Set to a Router Info - hence it was not possible to de-anonymize an I2P service provider by just using mass surveillance technology”.

Remark 1: the study is NOT trying to identify/exploit bugs within the I2P software to de-anonymize service providers within the I2P network (in this context, this is not interesting for the researchers - the overall I2P architecture is the interesting part).

Remark 2: there are two areas of “I2P de-anonymization” research sponsored by diva.exchange - one is focussing on “taking over I2P tunnel control by harrassing network participants” the other one is focussing on “de-anonymization using network surveillance technologies in combination with pattern recognition”.

The study will be published sooner or later on academic channels. To get notified, follow here: https://x.com/@DigitalValueX or https://social.diva.exchange/@social (mastodon)

c) I2P Docker / I2P-SAM news: docker images (https://hub.docker.com/r/divax/i2p) and a complete I2P SAM library (https://github.com/diva-exchange/i2p-sam) are updated regularily. The docker images and the SAM library are used in the academic context. They are well tested and reliable.

​