Commit Graph
11 Commits
Author SHA1 Message Date
Ilker ManapandClaude Opus 5.5 9e46ee83e4 Merge C24 KV-index routing from vLLM's cache events
Conflicts with the LoRA work: vLLM's arguments, the engine profile, router
metrics and the scheduler's request all keep both; agent.pb.go regenerated
from the merged proto; the two test fakes of vLLM given separate names.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-05 01:24:25 +02:00
Ilker ManapandClaude Opus 5.5 117f417ef5 Precise KV-index routing: vLLM KV cache events through the agent, routed by leading blocks held
Plan §7.5 tier 2, opt-in per profile with kv_events. The agent starts vLLM
publishing KV cache events to a loopback port it listens on (pure-Go zmq4,
msgpack in both of vLLM's encodings), with sha256_cbor hashing and a fixed
PYTHONHASHSEED; it mirrors each engine's blocks and sends net changes up the
existing stream, with a full copy on every new one and a reset on gaps,
restarts and clears. The control plane hands them to the router, for the
sending node's own replicas only. The router tokenizes through a replica's
/tokenize, hashes the tokens as vLLM does (checked against cbor2/hashlib
golden values), and scores replicas by the share of leading blocks held,
falling back to the approximate tree. New metrics beside
llmgw_prefix_decisions_total. Untested against a running vLLM.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-05 01:21:11 +02:00
Ilker ManapandClaude Opus 5.5 2a1e0e6c4c LoRA adapters on vLLM, loaded on demand with replica affinity: max_loaded_adapters
A vLLM model can carry PEFT adapters, registered at start, or with
max_loaded_adapters held at most N per replica: the router loads an adapter
through vLLM's runtime LoRA API when a request first wants it, sends the
adapter's requests to a replica that holds it while that one is not
saturated, loads each cold adapter once, and unloads the least recently used
idle one to make room. llama.cpp keeps loading every adapter at start.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-05 01:14:38 +02:00
Ilker ManapandClaude Opus 5.5 3b05ec0538 Have engines listen on the management address, and take SGLang's key off its command line
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-10-04 10:36:42 +02:00
Ilker ManapandClaude Opus 5.5 62657f4c2a Release v0.10.4: CPU threads, tracing, and the rest of the review
Release / notes (push) Successful in 3s
CI / test (push) Successful in 4m57s
Release / build (push) Successful in 4m11s
CPU replicas start on half the machine's CPUs, not all of them: two busy
llama.cpp servers each spinning on all 8 vCPUs fell to 0.8 tok/s apiece,
where at 4 threads they keep 5.5. llamacpp.threads sets it per node.

Inference requests can be traced over OTLP/HTTP (tracing.endpoint), with
admission, schedule and upstream spans and traceparent sent to the engine.

Security: onboarding no longer puts the cluster root in the system trust
store and removes it where it was; the mirror's mutual-TLS API can be
limited to named certificates (--api-client); init.log no longer keeps the
administrator's password; a warning when auth is off on a public listen;
the recorded actor is sanitised and sudo-aware; Ray's dashboard is off, and
the remaining engine gaps are documented.

reconciler.go and agent.go are split by topic, the status mapping of each
failure is tested, and the plan's checklist is brought up to date.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 04:00:35 +02:00
Ilker ManapandClaude Opus 5.5 9853936e0b Give every engine a key only the router holds
Engines listen on every address their node has, with no key. Checked on the
live cluster: from a neighbouring node, a request straight to an engine's
port was answered — past the gateway's authentication, quotas and
accounting, by anyone who could reach the node.

Each replica's engine is now started with a key and refuses requests without
it. The key is derived from the replica id with an HMAC over a secret kept
in the store, rather than stored per replica: the router has to send the
right key to replicas this process never started — adopted after a restart,
or placed by the other control plane — and a function of a shared secret
gives that without a table to keep in step. A key taken from one node opens
that replica only. The secret is made on first use and copied with the rest
of the store when moving to PostgreSQL; a new one would lock the router out
of every running engine at the moment of the move.

The key reaches llama.cpp and vLLM through their environment
(LLAMA_API_KEY, VLLM_API_KEY) rather than their command line, which every
user on the node can read; SGLang only takes it as an argument. The agent's
readiness check, the router's proxy and its metrics scrape send it, the
client's own Authorization still never reaches the engine, and `ctl bench`
fetches the key of the replica it is pointed at from the admin API.

Engines started before this keep running without a key until their replica
is next started. With no store there is nowhere to keep the secret, and
engines start without one, with a warning.

Also formats two files that went in unformatted.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-30 00:38:29 +02:00
Ilker ManapandClaude Opus 5.5 4c5a2fd6da Install vLLM on every GPU node that can run it, as llama.cpp is
vLLM was installed by hand and named in each agent's configuration. The
agent now installs it: a Python 3.12 environment built by uv, which is
itself a GitHub release checked against its checksum, from PyPI resolved as
it stood a day after that vLLM was published and installed with
--require-hashes. The resolved list is kept in the environment. The
environment is built relocatable under a scratch name and renamed into place
once it runs, like an archive.

The default is 0.30.0, whose flags were checked against the ones llmgw
builds. It is built for CUDA 13, so it needs driver 580 and a Turing GPU or
newer, and 25 GB free while it installs; a node lacking one says which, and a
node with no GPU skips it without being shown as a fault. gcc comes with it
because Triton compiles its launcher at the first model load. SGLang
installs the same way when the control plane names a version. llmgw doctor
reports whether vLLM can be installed on the node.

Checked by installing 0.30.0 this way on a test node: 7.7 GB, torch 2.13 for
CUDA 13.0, importing cleanly; the node has no GPU, so vLLM was not started.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-29 13:15:54 +02:00
Ilker ManapandClaude Opus 5.5 45e3f52da3 Launch multi-node vLLM its own way by default, not through Ray
An empty launch_mode meant ray: a Ray head and joins on every node, a
dependency to install, and a path never checked against a real cluster.
It now means mp - every node runs vllm serve with its rank and the
leader's address, the workers headless - which is how vLLM documents a
multi-node launch today and needs nothing but vLLM. ray is still there
for a vLLM too old for mp's flags. The example agent configuration,
doctor and the reference say so.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-29 09:13:59 +02:00
Ilker ManapandClaude Opus 5.5 9ee21dc814 Fix what a multi-node launch would have got wrong on real hardware
vLLM's mp launch gave workers no rank and the leader's address as their
own VLLM_HOST_IP; every node now runs serve with --nnodes, --node-rank,
--master-addr and --master-port, and the workers --headless, as vLLM
documents it.

NCCL_SOCKET_IFNAME and GLOO_SOCKET_IFNAME were the control plane's one
interface name for every node; the agent already worked out its own -
the one holding its management address - and now uses it.

On RoCE, NCCL's default GID index is a v1 link-local entry on most hosts,
which does not route. The agent finds the RoCE v2 entry holding an IPv4
address and passes it as NCCL_IB_GID_INDEX; InfiniBand gets none.

A worker whose only process needs the leader - a Ray join - started
without waiting for it, since the wait was honoured for sidecars alone;
it found no head, exited and failed the replica. It now waits.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
2026-09-29 00:17:31 +02:00
ilkerandClaude Opus 5 1d1a6aebb0 Phase 4: make a multi-node replica able to start
CI / test (push) Failing after 53s
Placement and environment generation had landed early, which made multi-node
look further along than it was. A replica spanning nodes could not come up at
all, and all three reasons came down to assumptions that only hold on one node.

Not every rank serves HTTP. The agent health-checked every shard at its own
port, but only the leader answers; workers timed out, reported Failed and took
the replica with them. Engine.ServesHTTP now says which ranks serve, and a
worker is judged by staying alive. That is deliberately weaker evidence: the
replica's real readiness is the leader answering, which only happens once every
worker has joined.

A shard is not always one process. Engine.Commands returns a sequence — every
command but the last is a sidecar kept alive for the shard's lifetime and torn
down with it, and the last one's exit is the shard's. vLLM under Ray is why: the
leader runs a Ray head and then the server against it, while a worker runs only
the join and no vLLM process at all. RayCommand had been written and tested but
never called, so launch_mode: ray would have run vllm serve against a cluster
that did not exist. Sidecars are killed with the shard, because a Ray head left
behind holds the port the next replica needs.

Start order was backwards. Workers went first, on the theory that the head
should wait for them. Both engines put the rendezvous point on rank 0, so a
worker started first has nothing to connect to. vLLM's server blocks until the
Ray cluster has the GPUs it needs, so the leader can start immediately and waits
on its own. A command can now also require that rendezvous point to be listening
before the next starts, bounded by a timeout so a head that never appears fails
the shard instead of hanging it.

Stop was also killing only the engine, leaving sidecars behind.

None of this is verified against real engines — it is tested with stub
processes. The Ray sequencing, and which launch mode a given vLLM version wants,
can only be confirmed on hardware.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 06:20:03 +02:00
ilkerandClaude Opus 5 6a13e3dddf Phase 2 (1/2): node discovery, engine adapters, process supervisor
CI / test (push) Successful in 45s
Everything here works without GPUs and is covered by tests. The control plane
connection is not built yet, so llmgw agent itself is still a stub and the phase
exit criterion is not met.

- discovery: IB from /sys/class/infiniband, parsing the formats sysfs actually
  uses and telling native InfiniBand from RoCE; GPUs via nvidia-smi, the
  cgo-free path, with health flags for uncorrected ECC and memory already held
  by another process. A node without IB is not an error, it is a single-node
  island.
- engines: vLLM and SGLang command and environment construction, kept free of
  process handling so it stays testable. The NCCL environment is the part the
  plan marks critical: NCCL_IB_HCA lists only the chosen fabric's adapters,
  because leaving it open lets NCCL pick an adapter on a fabric where the peer
  is unreachable, and the job then hangs instead of failing.
- validation before launch: no mixed GPU types in a tensor-parallel group, the
  node/GPU/TP arithmetic has to add up, and fp8 is refused below compute
  capability 8.9 rather than silently falling back to a slower path.
- supervisor: process groups, because engines fork workers that hold GPU memory
  and outlive a signal sent only to the leader; a bounded log tail, since engine
  output is the only diagnosis when loading fails; Starting separate from
  Loading, because a 70B model legitimately takes minutes; and orphan adoption
  through pid files fingerprinted with the command line, since pids get reused
  and adopting an unrelated process is worse than leaking the original.
- readiness is two questions: /health says the process is up, /v1/models says
  the weights loaded. Routing on /health alone sends traffic into a loading model.
- llmgw agent inventory reports what the scheduler would see on a node.

One plan correction: GPU class sizes come from the driver in GiB, which does not
always match the marketing name. An L40S is sold as 48GB and reports 44.99 GiB,
so its class is l40s-45g, not l40s-48g. Consistency matters more than matching
the box, because that number decides whether a model fits. Plan §2 and §5.1
updated.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-16 02:01:59 +02:00