kirill lykov

Current master afa2ae2986cd704c0d7cb22bfa7238cd9e4372cc shows different trend: Looks like problem with "fur" has been resolved by this PR: https://github.com/solana-labs/solana/pull/27461 The problem with memory growth is there. I've profiled the moment of...

Another question is why memory grow on one machine but not on the other. This is unclear

To exclude that the problem is due to impossible super majority, I've tried the same with 5 nodes: ```bash cd net/ ./gce.sh create -n 4 -c 1 ./init-metrics.sh ./net.sh start...

A note about `store_create` and memory peaks: if we look into time of memory peaks and at create_store count, they are coincident:

If we use TpuClient instead of ThinClient, it leads to uniform traffic distribution among nodes and this lower the memory growth trend (but it is still present). Surprisingly, it also...

> Are you on an intel mac? I'm not seeing these tests fail on my arm mac, FYI. Just ran off of [c04747d](https://github.com/solana-labs/solana/commit/c04747dd66534ac20102cc26e5f74ff2af3280c9) No, I'm on M1 Pro. I've tried...

interestingly, some tests are failing always (`nonblocking::quic::test::test_quic_server_staked_connection_removal`), while others (`recvmmsg::tests::test_recv_mmsg_multi_iter_timeout`) are failing sometimes

Spin off https://github.com/solana-labs/solana/issues/27917

Changes are split into many small PRs: 1. https://github.com/solana-labs/solana/pull/26067 2. https://github.com/solana-labs/solana/pull/26068 3. https://github.com/solana-labs/solana/pull/26513 4. https://github.com/solana-labs/solana/pull/26623 5. https://github.com/solana-labs/solana/pull/26661 6. https://github.com/solana-labs/solana/pull/26670 7. https://github.com/solana-labs/solana/pull/26829 8. https://github.com/solana-labs/solana/pull/27030 9. https://github.com/solana-labs/solana/pull/27151 TODO: PR which adds flag...

Ongoing work: adding durable nonce run to bench-tps runs