kev-0.5b declared architecture "ggmlc", which llama.cpp does not know.
It was fetched, catalogued and placed, and failed on every node — while
the answer was in the first kilobytes of the file.
internal/gguf reads a GGUF header (v1-v3) up to general.architecture,
over HTTP with a Range request, and judges it against the architectures
llama.cpp b10991 lists in LLM_ARCH_NAMES (152, copied from the source at
that tag); a vision projector is not a model on its own. An unknown
build is not refused, only not vouched for, and a header that cannot be
fetched does not refuse anything.
Checked in three places: the registry listing (from the architecture
Hugging Face reports with expand[]=gguf, so one request per repository)
shows it and withholds the fetch button; a fetch reads the header and
refuses unless "fetch anyway"; and the model service refuses to add
such a model, which covers the serve form and ctl model add.
Tried on real files: the mirror's qwen2.5-1.5b (qwen2) and bge-m3
(bert), and Qwen2.5-0.5B on Hugging Face through its redirect.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A member of an organization can be its manager (users join --org X
--role manager, or the star on the people page). /org shows a manager
their organization: usage over time by team, project or person, its
teams and projects with limits and 30-day spend, and its members, each
once with every place they belong. Read-only; administrators see every
organization. Only what was spent in the organization's projects is
shown — UsageBucketsIn filters by where a request was spent.
Found on the way:
- Usage by org, team and project, and the quotas page's spend, joined
through api_keys and so dropped every chat-page request, which has no
key row. They now read usage.project_id; keys no longer move between
projects, so the two agree.
- /org was not exempt from API-key auth and answered browsers with a
JSON error. A router test now drives the session pages through the
middleware; it fails without the fix.
Membership queries, org-scoped usage and the reworked joins pass the
PostgreSQL store tests on 16.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A person was in exactly one project, chosen at approval, and could not
use the gateway before that. People belong to several teams and to
none, so:
- memberships: a person is a member of any number of organizations,
teams and projects, or of none; a member of a team or organization
may use every project in it. users join/leave, and the people page
adds and takes out from either side. Approval no longer needs a
project.
- personal keys: a key may have an owner and no project. /profile asks
what a new key spends against; keys create --for does the same.
auth.personal sets default limits for personal use or turns it off.
- limits for people: --scope user holds a person across every key,
project and the chat page; --per-member on an org, team or project
gives each member their own share there. Resolution folds the tree,
each per-member limit for the key's owner, and the owner's own, each
in its own pool; the quotas page shows and sets both.
- the chat page asks what to spend against and checks it per message.
- usage records who spent it, so a person's budget survives a restart,
personal use and chat included; restore no longer drops rows with no
project.
Leaving a project revokes the keys made there unless another membership
still reaches it. A personal key keeps no threads or notes, which are
per project, and is refused rather than dropped into the anonymous pool.
Migration 0010 turns each person's project into a membership, adds
usage.user_id and backfills it, and makes api_keys.project_id nullable.
SQLite cannot drop NOT NULL, so the table is rebuilt; that needs foreign
keys off, which only works outside a transaction, so migrations marked
for it run on one connection with foreign_key_check before commit.
Tested on a populated SQLite store with claims, and on a copy of the
cluster's PostgreSQL store: 5 people became members, 17 keys and 3
claims intact, 145 of 577 usage rows and all 69 chat rows attributed.
The PostgreSQL store tests, including new membership cases, pass on 16.
users move is gone; join and leave replace it.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
server init now also:
- runs a model mirror beside the control plane (/srv/models, bytes on
:8081 at an address nodes reach, management API on loopback so it
needs no certificate) and adds it to server.yaml;
- makes the machine a node of its own control plane: issues its agent
certificate, writes agent.yaml labelled as the control plane, starts
llmgw-agent, and waits for it to register and install llama.cpp;
- checks for NVIDIA GPUs with nvidia-smi, and reports a card on the PCI
bus whose driver does not answer rather than ignoring it;
- fetches Qwen2.5 0.5B (Q4_K_M) into the mirror and adds it to the
catalog pinned to this node: a llama.cpp profile per GPU class found,
with a CPU profile as the fallback, then waits for it to serve.
--no-mirror, --no-node and --no-model leave these out. Every step looks
before it acts, so a rerun finishes what an interrupted one started.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Nothing in the tree could be renamed, moved or removed, and an approved
person stayed in the project they were first put in — from the web or
the CLI. Now:
- /admin/users is the tree, each project opening onto its people and
keys, with approve, move, suspend, reinstate and remove for people,
revoke for keys, and add, rename, move and remove at every level,
plus a search over every account. Forms post and get the page back,
so it works with no script; htmx 2.0.11 (0BSD, embedded, checked
against npm's integrity hash) swaps only the changed part, keeping
open projects open.
- llmgw org tree/add/rename/move/remove, users move/delete, and
keys list --all.
- Projects, teams and organizations are named by id, path
(Acme/platform/gateway) or unique name; users by id or email. With a
single project --project can be left out, instead of defaulting to
"default", which a tree built from the web does not have.
Moving a person moves the keys they made; operator keys stay with the
project. Only empty things are removed, and never a project that had a
key, since usage history refers to it.
Also: keys init no longer reparents an existing team or project;
reinstating an account with no project is refused (a rejected signup
used to come back active and fail at its first key); rejecting a
signup removes it; keys limits --budget takes whole units like the
quotas page.
Checked in headless Chrome: a move from an open project updates the
page in place, keeps the project open, and the search still filters.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
server init --postgres failed on Rocky with postgresql-setup not found.
The PostgreSQL project's packages (PGDG) provide "postgresql-server",
so dnf can satisfy that name with postgresql16-server — whose unit is
postgresql-16.service, whose tools are in /usr/pgsql-16/bin and whose
setup script is postgresql-16-setup. The code assumed the distribution's
names for all three.
The server unit is now picked from systemd's postgresql* unit files, its
data directory read from the unit's own PGDATA, and initdb run directly
as postgres when that directory is empty, with password logins over TCP.
psql, initdb and the restart use what was found. The unit names and
PGDATA values in the tests are the ones Rocky's and PGDG's packages ship.
Not yet handled: a server initialised with ident for TCP logins, one on a
port other than 5432, and several PGDG versions where the highest is not
the one on 5432.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Installs PostgreSQL with the machine's own package manager when it has
none (initdb with password auth on the RHEL family, whose default is
ident), creates the llmgw role and database with a generated password
kept in /etc/llmgw/secrets/store-dsn, and checks the address connects.
On a machine already on SQLite it stops the gateway, copies the store
so accounts and keys survive, points auth.store at the new address and
starts it again. A move that fails starts the gateway on SQLite again.
server.yaml is edited as text so its comments survive, and kept beside
it. A PostgreSQL init did not install is not resized: it may serve
something else. It refuses to move one PostgreSQL store into another.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
`llmgw server init` does what install.md's steps 2 and 3 asked of the
operator: the llmgw user and directories, a certificate authority, the
control plane's certificate for every address the machine has, a
configuration, the default project, an administrator with a generated
password, and the systemd unit — then starts it and waits for /health.
Every step looks before it acts, so a second run changes nothing, and
--host later reissues the certificate with the old names kept. It never
overwrites server.yaml or an authority, and leaves alone certificates
issued by an authority kept on another machine.
The configuration install.md showed could not start: an empty catalog
without control.enabled is refused as having nothing to serve. The one
init writes sets it, and a test holds that.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Seeing what a node did meant ssh through a jump host, to a machine whose
address nobody remembers, to read a journal only root may open. When an
engine install failed on a new node today, the reason was one line — and it
was four steps away from the page that said the install had failed.
Each node's row on the cluster page now links to its logs: the agent's own,
and every engine running there, chosen from the list the node itself sends.
The page can follow, refreshing every five seconds, and is on both the admin
listener and /admin. `llmgw ctl logs NODE [--source SHARD] [-n N]` reads the
same.
They travel over the stream the agent already holds, so they need nothing
the control plane does not already have: a request goes down, the lines come
back, and a reply that does not come in ten seconds is an agent too old to
answer. The agent keeps its last 2000 lines in memory as well as writing them
wherever it did before; the engines' last few hundred lines were already kept
and sent only when one crashed.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A node ran whatever llama-server somebody had once unpacked on it. Onboarding
checked an engine only when told its path, installed none, and a node with
nothing on it registered as ready. The one time that went wrong it went wrong
invisibly: the binary was there, executable, and could not start for want of
libgomp1, while the node reported healthy.
The control plane now says which engines every node has, at which version —
`engines:` in server.yaml, defaulting to the llama.cpp build this release was
checked against — and each agent converges on it. What only the node can
know, the node works out: the build for its processor and driver (CUDA with
its runtime when an NVIDIA driver is loaded, CPU otherwise), the OS packages
that build needs from its own package sources, and whether the result runs.
Every archive is checked against the sha256 GitHub publishes for it and
unpacked refusing any path or link that leaves the install directory. ldd
names a missing library outright; running `--version` proves the rest. Only
then is a directory marked installed, and it is checked again every ten
minutes, because an unrelated upgrade can remove a library.
An engine is reported only once installed and seen to run, or when the
agent's configuration names it on purpose — never because something happened
to be on PATH. The agent now says when it has probed, so an empty list means
"has none" rather than "too old to say": a node rebuilt from nothing is no
longer a candidate for every model at the moment it can run none of them.
Placement's refusal carries the node's own words, "installing: downloading
llama-b10991-bin-ubuntu-x64.tar.gz (16 MB)", and so do `ctl nodes` and the
cluster page, along with a warning for a CPU without AVX2 — the VM CPU type
that makes everything five times slower and nothing look wrong.
Agents also report their OS, architecture, vector extensions and NVIDIA
driver. Onboarding installs llmgw itself from the CI release for the node's
architecture, checked against the release's checksums, rather than copying
whatever binary the operator's machine had.
vLLM and SGLang are not installed yet: they are Python environments of
several gigabytes, not an archive, and still have to be named in the agent's
configuration.
Also fixes a race on the weights fingerprint, which two heartbeat loops touch
during a failover.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Two defects, both of the same kind: something the CLI knew and the rest of
the system did not.
A profile can be edited while a replica is running against the old one. The
reconciler deliberately leaves it alone — churning a replica that is serving
costs more than the imperfection — but there was no way to see it, so the
cluster could disagree with its own configuration indefinitely and read as
healthy. Profile now carries a fingerprint, a replica records the one it was
placed with, and both the API and `ctl replicas` say "stale profile" when
they differ. An empty fingerprint is not stale: it means a replica adopted
from an older agent, which is unknown rather than wrong.
Adding a model from the admin page skipped the source check `ctl model add`
has done all along, so a typo or a mirror that is down became a replica
failing on a node minutes later, with an error naming the node. The check
moves out of the CLI into a package both use.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The mirror would happily hold a four-shard model that nothing could serve. A
catalog entry names one source, the agent fetched one URL, and so a version of
several files had no way to be described — which is every safetensors model,
and every GGUF too large to ship whole.
A source ending in a slash names a version rather than a file. The agent fetches
every file in it into one directory and hands the engine the directory: vLLM
reads the shards and the config out of the folder, and llama.cpp is given the
first part and finds the rest beside it.
The trailing slash is the whole signal. A flag beside the source would be a
second place for the two to disagree, and a URL already distinguishes a
directory from a file everywhere else.
The mirror answers a version with a manifest instead of a file listing. A plain
file server gives HTML there, and something that has to know what a version
consists of would be left parsing it. It is on the byte listener, not the
management API: fetching weights is the one thing every node does, and making
it depend on the administrative port would undo the split that keeps that port
closeable.
Built from the directory rather than a file written beside it. A manifest on
disk is a second record of what is there, and the two disagree the first time
somebody copies a file in by hand — which the mirror explicitly allows.
The whole set or nothing. A directory with three of four shards loads as a
broken model rather than failing to load, so a completion marker is written
last and an interrupted fetch starts the missing files again rather than being
found and served. Files already there at the right size are not fetched twice.
Progress is reported under the version's address, not each file's: the shards
waiting on it asked for the version, and progress under a file's own URL
reaches nobody.
A filename from the manifest is refused if it contains a separator. The mirror
is a place an operator copies files into by hand, so what it reports is not
automatically safe to join onto a path.
An HTML listing where a manifest was expected says so. "invalid character '<'"
sends somebody looking at the wrong thing.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The page is served twice — on the admin listener, and at /admin behind a
session — and the history source reached only the second. The first version
built it after the admin listener's copy was already registered and passed that
one an empty NavExtras, so the page an operator reaches by SSH tunnel had no
charts and no explanation.
Built before either is registered, and handed to both.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The dashboard charted what the cluster had been billed for — requests, tokens,
spend, all from llmgw's own store — and nothing about whether it was healthy.
Saturation, queue depth, cache occupancy and replica counts were numbers in a
table at the instant the page loaded, with no way to see that one of them had
been climbing for an hour.
Four line charts now, read back from wherever the metrics were scraped to.
Read back rather than kept twice. llmgw publishes these already; what a gauge
cannot do is remember, and a ring buffer in this process would be a rival copy
of the same numbers with its own retention, lost on every restart and
disagreeing with the one an operator has on screen elsewhere. monitoring.prometheus
is empty by default and the panels are then absent rather than empty.
One axis per chart, always. Two measures of different scale belong in two
charts: a second y-axis lets any two lines be made to cross wherever the author
likes and the reader cannot tell it happened.
Four colours, handed out in fixed order, never cycled. A fifth series is summed
into "other" and drawn in grey — dropping it would make a chart that does not
add up, and inventing a hue for it would put two lines on the page that nobody
can tell apart. The fold is by timestamp, not by position: the tail's series do
not share sample times, and adding them by index slides points sideways onto
each other's clock.
A legend for every chart with more than one line, because colour alone is not
an encoding a reader can rely on.
A flat zero is drawn along the bottom rather than left blank: nothing happening
is a real answer, and an empty frame reads as a broken chart. A single sample
is not a line and says so instead of putting a dot in the corner.
One failed query fails all four panels. Three drawn and one blank invites
reading the blank one as "nothing is happening there".
NaN is dropped rather than plotted. Prometheus sends it as a string for a
sample it has no value for, Go's ParseFloat accepts it happily, and one NaN
coordinate is an SVG path that does not draw at all — taking the rest of the
line with it.
Axis maxima are rounded to a number a person would pick. An axis topping out at
0.8347 makes every gridline unreadable, and the shape is what a chart is for.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A node's registry holds an adapter for every engine whether or not its binary
exists, the agent never said which were installed, and placement carried an
Engine field it had never once read. So a vLLM replica was as happily placed on
a node with no vLLM: the node registered healthy, the shard failed minutes
later with "executable file not found" in the engine's own words, and the
reconciler tried again.
That is how fifteen gigabytes of safetensors reached this cluster's mirror
before anybody noticed it had no card and no vLLM to run them with.
Agents now probe what is installed and report it — the protocol has carried
engine_versions since it was written and nothing ever filled it in. Placement
filters on the key set, `ctl nodes` has a column, and the mirror page's serve
warning is a fact instead of a guess about cards, which would have waved
through a fleet of eight-GPU nodes that never had vLLM.
Probing resolves binaries and runs none of them. The first version asked each
engine for --version, which is fine for an engine and not fine for the wrapper
scripts people put in front of one: a script that shifts past flags it does not
recognise and then execs the server will start an inference server, at agent
startup, on every node. The agent's own test suite installs exactly such a
script and caught it. So presence is a path lookup, and the version is left to
`ctl engine-check`, which is run by somebody who meant to run the engine.
SGLang is the exception and has to execute, because a module has no path to
look up — but the command is llmgw's own one-line import rather than the
engine's launcher. Its exit code is the answer: python3 is on almost every
machine, and treating a failed import as "installed" reported SGLang on every
node in a fleet that had none, which is the failure being fixed, inverted.
Probed once, not per reconnect, and bounded by the context that registration
waits on: an agent that cannot reach the control plane is worse than one
reporting an incomplete list.
A node that reports no engines at all is not excluded anywhere — by placement,
or by the page. That is what every agent older than this reports, and emptying
a fleet on a control plane upgrade is a worse failure than the one being fixed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Fetching a model put it on the mirror and left the operator to hand-write a
catalogue entry and run `ctl model add` on the control plane — for a source the
page was already displaying, with an engine decided by the file extension it
had just shown them.
A present row now offers to serve it. The form opens on that row with the name,
the mirror's own URL and the engine filled in, and asks for the two things
nobody can infer: the context pool and how many replicas.
What it refuses is the point.
A version of several files is refused with the reason. A catalogue entry names
one source, so an entry for a sharded model is an entry for one shard, which
loads as a broken model rather than failing to load. Nothing said so before —
the mirror happily holds a four-shard model that no catalogue entry can
describe.
Safetensors on a fleet with no GPU is warned about before anything is added.
This is the failure we hit today: fifteen gigabytes of shards arrived, and only
then did looking at a node show no card, no vLLM and no SGLang. The control
plane is not told which engines a node has installed, so the check is on cards,
which it does know — a warning rather than a refusal, because a card can be
added tomorrow.
A name already declared in configuration is refused here too, matching what the
model service would say anyway, but before the form is filled in rather than
after.
Without a store there is no serve action at all: a model added at runtime would
not survive a restart, and offering it then refusing is worse than not offering.
The mirror page's dependencies became a struct. Three are optional interfaces
and two are the same shape, which at a call site is a swap nobody notices until
a page shows the wrong thing.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Mirroring anything meant knowing a repository existed, knowing which of its
thirty files were the weights, and assembling a resolve URL by hand. The mirror
page had a text box for that URL and no way to find one.
It now searches Hugging Face and lists a repository's weight files with their
sizes and published checksums, each with a button that fills the fetch form.
Only weight files. A repository is mostly a README and a tokenizer, and listing
those buries the four entries somebody came for.
The size shown is the LFS size, not the plain one. For an LFS-stored file —
every weights file — the plain field is the size of the pointer, so the other
one reports a gigabyte model as 133 bytes, and that is the number somebody
would then plan disk against.
The source URL is pinned to the commit the listing came from rather than to a
branch, so what the mirror fetches is what was on screen. A repository
re-uploaded under the same branch otherwise leaves two different sets of
weights with identical provenance and no way to tell them apart. The version
defaults to that revision for the same reason.
The registry's own failures are reported on their own line. It is somebody
else's server on the other side of the internet, and it being down says nothing
about the mirror — which is what the page is for and has to keep working
regardless. With no registry configured there is no search box at all, rather
than one that can only time out.
Gating decodes from a bool or a string, because the field is documented as the
first and is the second in practice: strict decoding fails the whole response,
so one gated model would hide the other nineteen in a page of results.
And a bug this made obvious: downloads were keyed by model and version, so the
second shard of a sharded model was refused as "already fetching" until the
first had finished. A fifteen-shard model had to be pulled one file at a time,
each waiting on the last. Keyed by file now — which is what the single-flight
was ever protecting, since the race is two writers on one .part. A version made
of several files reports as the worst of them: it is not present until every
shard is, and one shard failing makes the version unusable however many others
arrived.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Quotas can be written on an organization, a team or a project, and nothing
could show what any of them had used: usage came back per key, per user and per
model only. A quota whose consumption cannot be seen is one nobody reviews
until the day it refuses somebody — and since the previous commit made a group
quota a shared pool, it is now a number several people are spending against
without being able to look at it.
`llmgw ctl usage --by team` sums every key beneath each group, with the budget
that level sets beside it.
Beside it, and only when that level sets one. A team bounded by its
organization's budget has none of its own, and printing the organization's
against each of its teams would read as though each had that much — the exact
misreading this work exists to remove.
Joined through api_keys rather than read from usage.project_id: that column
records where the key was when the request was served, while the quota is
written against the tree. They agree until a key is moved, and a report that
disagrees with the limit shown beside it is the useless one.
Three existing reports were fixed on the way, all failing on PostgreSQL and all
passing in tests, because the tests run on SQLite:
column "k.name" must appear in the GROUP BY clause (UsageSince)
column "us.email" must appear in the GROUP BY clause (UsageByUser)
column "usage.ts" must appear in the GROUP BY clause (UsageBuckets)
SQLite picks an arbitrary row for an ungrouped column and carries on; PostgreSQL
refuses. The first two were selecting names they had not grouped. The third
grouped "ts / ?" while selecting "(ts / ?) * ?", which PostgreSQL matches as
text and so saw a bare ts — it is grouped by ordinal now, the same expression by
construction. The dashboard's usage charts have been empty on the cluster this
whole time, with the reason only in the log.
All four queries were run against the cluster's own PostgreSQL before this was
committed, because passing on SQLite is what they were already doing.
The grouping level picks from a fixed table of levels rather than being
interpolated, so an unknown one is an error and not a fragment of SQL.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Two things that were only ever confusing, and one of them stopped a command
working on the cluster.
The store type was named SQLite and kept the name when PostgreSQL was added, so
OpenPostgres returned a *SQLite and all seventy-odd methods on the production
store were spelled after the database it was not running on. Nothing broke;
people did. Earlier today I added three methods to *SQLite for a table that on
this cluster only ever lives in PostgreSQL, and the name did not once look
wrong. It is a field, which is all it ever was.
`llmgw users` and `llmgw keys` defaulted --store to a SQLite path. On a
deployment whose store is PostgreSQL every one of them failed with "unable to
open database file" while the right answer sat in auth.store in the config file
beside them; the conclusion an operator draws is that the tool is broken, and
they are not far off. They now ask the server's own configuration before
falling back, and the error names which of the three sources chose the address
— the failure was never the path, it was which one picked it.
The single-machine default still applies last, so a deployment with no config
file keeps working with no flags.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
`ctl bench` has always been able to work out what one replica can serve — it
ramps concurrency until the SLO breaks and reports the last level that held.
It printed that and nothing kept it, so the one number saying what a replica is
worth lived in whoever ran it's scrollback. Everything downstream then had to
guess: admission applies a single global default for healthy queue depth, and
sizing a model against real demand had nothing to size against.
`ctl bench --save` records it; `ctl capacity` reads it back.
Keyed by model and profile. Capacity belongs to the hardware and the engine
arguments as much as to the weights, so the same model on CPU and on an H100
are two numbers, and one key would overwrite the other on every run.
The profile is looked up from the replica rather than asked for, because the
control plane already knows which profile answers on which endpoint and a
measurement filed under the wrong one is worse than none. A target that is not
a replica — the gateway, usually — is refused with that said: benchmarking the
whole path is worth doing and says nothing about the hardware.
The conclusion is stored, not the ramp. Every level and its latencies are worth
reading once while judging whether the run was sound; keeping them would grow
without bound as a model is re-benchmarked.
Re-measuring replaces. A second run is a better answer to the same question,
not another data point, and keeping both would leave a reader choosing between
numbers with no basis for the choice.
The SLO and the date are stored with it. Capacity means nothing without the
target it was measured against, and hardware and engine builds change — a
measurement is an observation with a date on it, not a fact about the model.
A failed save is reported and not fatal: the run has already cost the GPU time.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The count is scaling.min, and until now the only way to change it was to edit
server.yaml on the host and restart — or, for a model added at runtime, to
remove it and add it back, which means having its whole definition to hand.
`llmgw ctl scale <model> --min N` instead. Only the scaling block moves:
everything else is read back from what was stored and written out untouched, so
changing a number cannot quietly become a redefinition.
The catalog is rebuilt and validated before anything is written, as in Add. A
max below a min is refused with the stored definition untouched — persisting a
rejected change would come back on the next restart as a catalog that does not
build.
The answer reports the scaling in force rather than the request. The catalog
fills in a max that was left out, and a caller echoing its own request would
print a number the cluster is not using.
A model declared in configuration is refused, for the reason Remove already
refuses one: the YAML is read on every start, so the new count would hold until
the next restart and then silently revert. A cluster disagreeing with its own
config file is worse than an error saying where to edit.
The help says what --max does, because it is not what it looks like: a model
with a minimum runs exactly that many and nothing scales it up under load, and
an on-demand model loads one replica however high its max.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
/admin/mirror, behind the same admin session as the other admin pages. What is
held, what is arriving with its byte count, what failed and why; start a fetch,
retry a failed one, remove a version.
Not mounted at all without a mirror configured. The right page for a deployment
with no mirror is no page, rather than an empty one pointing at nothing.
It reloads itself only while something is downloading. A reload at any other
moment takes with it whatever is half-typed into the fetch form, and there is
nothing to watch when nothing is moving.
A failed row offers Retry, not Remove. A fetch that failed left nothing on disk,
so the remove could only answer "the mirror holds nothing for this" — a button
whose single outcome is a confusing error. Starting the same download again is
what somebody looking at that row wants, and the mirror allows it: only a
download still running is refused.
An unreachable mirror is said outright instead of rendering as an empty table.
The two look nearly identical and mean opposite things: one is a mirror waiting
to be filled, the other is a machine to go and look at.
Dashboard takes NavExtras rather than a second bool. Two adjacent booleans at a
call site get swapped silently and are discovered by somebody clicking the
wrong link.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The mirror holds the weights, so the mirror does the downloading. A control
plane that had to pull the bytes itself and push them here would carry every
model twice over the network, through the one machine least able to spare the
bandwidth.
Two listeners, and the split is the point. Weights stay on plain HTTP where
every node can reach them: that path is eight nodes pulling forty gigabytes,
and putting a client certificate in front of it would mean re-certifying every
agent's fetcher in order to change how models are administered. Management gets
its own port, which can be given the fleet's CA — the same tlsconf the agent
stream uses, so this is a configuration rather than a new idea.
A fetch answers 202, not 200. The bytes are not here yet, and a caller that
read it as "done" would point a catalog at a file still arriving, which every
node would then copy. Progress is asked for rather than waited on, because no
client between here and the operator will hold a connection open for an hour.
A failed fetch is remembered with its reason. Forgetting it would leave the
control plane waiting on a model that is never coming, with nothing to say why.
Status falls back to reading the disk, so a restart does not report a model
that is plainly there as unknown.
Removal takes the whole version: a model split across files is useless with one
of them missing, and half a version on disk looks present and will not load. It
refuses while a download is running, because the fetch would carry on writing
and the file would reappear, looking like the delete had silently failed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The mirror was a directory and a discipline written down in a README. This
makes the three operations that matter into commands, on the mirror host,
where `llmgw keys` already sets the precedent of operator tooling that works
on files rather than through the API.
The gateway still never writes to the mirror. That property was chosen
deliberately — a mirror writable by the thing reading from it is a way to
replace a model with one every node then fetches without complaint — and a
person running a command does not take it away. `serve` states it rather than
inheriting it, so a handler added later cannot quietly make the mirror
writable.
`add` downloads to a temporary name and renames, for the reason the agent's
fetcher does: a mirror is copied to every node, so a partial file here is a
partial file everywhere. It verifies the length the server promised, reports a
checksum — the mirror is the one place a corrupt file reaches every node
rather than being caught once — and refuses to replace an existing version,
which is the whole point of the layout.
`unused` refuses to answer when it cannot read the catalog. An empty catalog
and an unreachable one produce the same list, everything, and the obvious next
step after that list is deletion. It compares by path rather than by spelling,
because a base typed with a different scheme or a trailing slash would
otherwise report a file in use as unused.
`remove` checks the catalog first and says to retire the entry before deleting
the file, because a node that has to refetch a deleted file cannot, and that
failure arrives later, on the node, looking like something else.
Verified end to end: Range requests answer 206, which is what makes an
interrupted fetch on a node resume; writes answer 405; a traversal attempt
answers 404.
Note: internal/agent/supervisor's TestSidecarRunsBeforeEngine fails under full
suite load on main as well as here — it asserts that one shell's echo lands
before another's, which the supervisor does not promise. Everything else
passes. Left alone rather than fixed in a commit about mirrors.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The agent has always logged a percentage as it fetches weights, with a comment
saying a replica pulling a hundred gigabytes and one that is wedged look
identical from the control plane. They still did: the log is on the node, and
the node is not where anyone is looking when a replica has been starting for
ten minutes.
Progress now rides on the shard status and appears under the replica in `ctl
replicas` — a line rather than a column, because most replicas are not
downloading anything and a column would be empty for all of them.
It is deliberately not a state transition. Treating each chunk as one would
re-mark the shard as starting on every report and reset the timer that decides
when a slow start has become a failed one, so an hour-long download would keep
a doomed replica alive indefinitely. It arrives on its own method for that
reason, and a real transition clears it — otherwise a finished download sits in
the listing at ninety-nine percent under a shard that has been serving for an
hour.
Watchers are keyed by source rather than by shard, because a fetch is
single-flight: eight shards starting together produce one download and all
eight are waiting on it.
No percentage is invented when the server sent no length. One that stops moving
is worse than a byte count that keeps rising, so the field is absent and the
listing says "so far".
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Five nodes fit in a table. Fifty do not, and this is built for people who
already have racks of GPUs — so the flags are how a question gets asked:
which nodes on this fabric, which of this GPU class, which are not ready,
which replicas have a shard on this machine.
Filtered here rather than in the API, so --json is narrowed the same way the
table is. A server-side filter would be two implementations of one predicate,
and the JSON one is the one nobody would notice had drifted.
An empty result says which case it is. A filter that matched nothing and an
empty fleet call for opposite next steps — one is a cluster to look at, the
other is a word to correct — and printing "no nodes" for both makes a mistyped
fabric name read as an outage.
A state matches on prefix because states are few and distinct and somebody
scanning a fleet types quickly. A pool matches either half, because it is
fabric/class and both are worth narrowing by. A replica is found from any of
its nodes, since one spanning two machines is on both.
--label takes a bare key as "carries this at all", which is how you ask for
every node with a role without caring which. The flag's help said so before
the parser did; the test caught them disagreeing.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The second half of leadership. A standby that refuses to place but still holds
an agent's stream is worse than useless: commands travel down the stream the
agent opened, so the leader's placements fail with "no agent connected" for a
node that is plainly connected.
So a standby refuses the stream, before reading the opening message, and says
why. And an agent takes a list rather than an address, trying each in turn — a
standby refusing, a stopped control plane refusing the connection and a crashed
one never answering are the same thing from a node, and the answer to all three
is somewhere else.
`server` and `servers` are both accepted, with `server` first when both are
given: every agent configuration written before there could be a second control
plane names one address, and an operator adding a standby expects the one
already working to stay preferred. With a single address the behaviour is
exactly what it was, including the backoff — which is skipped only while there
is another address that has not just been tried, because the delay exists to
stop hammering one control plane and the next one has not been hammered.
Removed a check that asked whether one field was empty, which is how the first
run of these tests failed: the list was populated and `server` was not, and an
agent configured entirely correctly was refused by its own validation.
The tests run a real agent against a real service over h2c. The first attempt
used a plain HTTP server, and the refusal arrived as an HTTP/2 framing error
that said nothing about why — which is exactly what an operator would have seen.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The first thing HA needs that a shared store does not give. Two control planes
against one store both reconcile: both see a model wanting two replicas and
none running, and both place. The model ends up with twice what it asked for,
each replica holding GPUs the other's inventory believes are free — and nothing
crashes to say so.
A lease in a table rather than a lock held by a connection. PostgreSQL has
advisory locks and SQLite has nothing like them, and this store keeps one set
of queries for both; a row with an expiry works everywhere, and it survives a
leader being powered off rather than asked to stop, which is the case that
matters. Taking and renewing are the same statement, so the two cannot disagree
about what a held lease looks like, and it is one statement, so two processes
racing cannot both believe they won.
The owner id is random rather than the hostname. A hostname would let a
restarted process match its own stale lease and believe it had been leading all
along — including through the gap that made it restart, during which somebody
else may have taken over and started placing.
A store that cannot be reached is not evidence that somebody else is leading,
so an unreachable store keeps the claim. Two leaders is worse than one, but no
leader means nothing is placed and nothing is evicted, and a brief database
blip should not stop a cluster healing itself.
Unset means leader. Every deployment that existed before this, every
single-control-plane install and every test acts unconditionally, with nothing
to configure.
The same seven assertions run against both backends, which is how the test
harness was caught inheriting a lease between subtests — it passed on SQLite,
where each subtest gets a fresh file, and failed only on PostgreSQL.
What this does not yet do is move agents. A standby holding an agent's stream
is one the leader cannot command, so failover is degraded rather than clean
until agents know more than one address.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Keys are stored hashed, so a migration that drops them cannot recreate them:
every person and every script needs a new one issued, and the old ones cannot
even be looked up to say which they were. `llmgw keys move-store` copies
instead.
Found by running it against the real cluster, twice, after the gateway had
already been stopped.
The first order came from the order the migrations create tables in, and
users is created in a later migration than api_keys even though api_keys
references it — so the copy died three tables in. The order is now checked by
a test that derives the foreign keys from the migrations rather than restating
them, so a new reference fails there instead of during somebody's migration.
Checked by putting the wrong order back and watching it fail.
Then key_claims. Its sealed column is emptied the moment a claim is used, so
every claimed row holds an empty blob in a NOT NULL column. An empty blob
scans as a typed nil slice and a genuine NULL as an untyped one — a
distinction this side can see and pgx cannot, since it encodes any nil slice
as NULL. Restored before the insert, which is the difference between a copy
that completes and one that stops four tables in.
One transaction per table rather than one for the whole copy: a single
transaction holding every row of a large usage table is how a migration runs
out of memory on the machine least able to spare it. A failure therefore
leaves a partial copy, which is why the destination must have been empty —
starting again is dropping and recreating it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
A path is still SQLite and a postgres:// URL is PostgreSQL, so every existing
configuration keeps working untouched and moving is one line changed rather
than a new setting to find. An address of the form @/path/to/file reads the
real one out of that file, which is how a DSN carrying a password stays out of
a configuration that gets copied, diffed and pasted into issues.
The driver is pgx, which is pure Go. That is not a preference: the whole binary
is built with CGO_ENABLED=0 so it runs on a node whose libc is older than the
builder's, and a cgo driver would take that away from every deployment,
including the ones still on SQLite.
Tested against a real PostgreSQL, which found two things nothing else could.
The schema_migrations table is created in Go rather than in a migration file,
so it escaped the port and still said INTEGER — 64 bits in SQLite, 32 in
PostgreSQL, and a millisecond timestamp does not fit, so the very first
migration failed to record that it had run. And the memory migration uses an
FTS5 virtual table with three triggers keeping it in step, which has no
counterpart: PostgreSQL derives a tsvector from the row, so there is nothing
to keep in step and no trigger to forget.
That leaves exactly one query that is not the same sentence in both dialects —
the note search, MATCH against @@ — and it is written out twice rather than
hidden behind a shared spelling that would mean something subtly different in
each. Everything else is one set of queries.
The text search uses the 'simple' configuration rather than English. These are
an agent's own notes, in whatever language it chose, and stemming English
rules over Turkish would quietly stop finding things.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
SetThreadRetention existed and nothing called it, so the feature was
unreachable: a project could never opt in and every request carrying a thread
was refused.
A command rather than a configuration key, because it is a decision about data
rather than about deployment. Someone has to choose to keep what a team types,
and choosing it should look like an act rather than like a default nobody
read.
--retain-hours is required. Without a number the command would create exactly
the thing the design avoids: a conversation store with no expiry, holding
everything a team ever typed because nobody chose one.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
`ctl weights --collect` answered "this control plane cannot ask nodes to
delete files" against a real cluster, with the whole suite green.
The controller is handed a dispatcherRef rather than the agent service —
an indirection that exists to break a construction cycle, and one that
satisfies the dispatcher interface by forwarding each method by hand. The
controller reaches past that interface by type assertion to find capabilities
beyond scheduling, and an assertion against dispatcherRef fails for anything
nobody remembered to forward.
Nothing could have caught it below the wiring. The dispatchers the tests use
do not implement the capability either, so its absence there is the expected
case and looks identical to this bug. The test therefore asserts on the
production types themselves: that the forwarding dispatcher carries the
capability, and that the service it forwards to has one to carry. Checked by
deleting the method and watching it fail.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Two halves of the same problem: nothing stopped a model being placed on a node
that could not hold it, and nothing ever removed the weights of a model that
had been deleted from the catalog.
Placement now has a disk check, shaped exactly like the memory one it sits
beside, with one addition: it is skipped on a node that already holds the
weights. A node holding the file needs no room for it, and without that the
node best placed to run a model — because it already has it — is the one most
likely to be rejected, since holding the file is what used the space.
Reclaiming is the dangerous half, so the whole design is about what it will
not do. llmgw deletes only files it fetched itself, and that is a calculation
rather than a heuristic: the cache name is a hash of the source, so the set of
wanted files can be computed from the catalog exactly. Anything without that
name was put there by an operator — scp, a shared mount, a copy by hand — and
is none of this system's business however unreferenced it looks. A model that
took a week to copy must not be removable by a command someone ran to tidy up.
The node checks every path again before unlinking it, and that duplication is
the point. The control plane knows the catalog; the node knows its disk. A
control plane that is wrong, newer than the agent, or tampered with must not be
able to talk a node into deleting something, so the agent refuses anything
outside its own cache directory or without the cache name — including a path
that tries to traverse out of it.
Listing is the default and deleting takes --collect. The summary says "asked to
remove", because each node decides, and one that refuses a path it cannot prove
it fetched is this working rather than failing.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Disk was the one resource nothing reported. `statfs` was already being called
on every model directory, but only to ask whether the filesystem was a network
one — the free space it returns in the same struct was thrown away. So the
only way to find out why a node was filling up was to ssh to it, which is what
this system exists to avoid.
Reported per filesystem rather than per configured directory. Free space
belongs to the disk, not the path: two model directories on one disk are one
pool of bytes, and counting them separately would let the control plane add
24 GB to 24 GB and place a 40 GB model on a node that has 24.
Shared filesystems are summed separately from local ones and left out of a
node's own total. Space on a filer belongs to everyone mounting it, so twenty
nodes would otherwise each report the same free terabyte, and a scheduler
would place twenty models into room for one.
Nothing reported means nothing is known, everywhere: an agent too old to send
storage and a node with no model directories both send nothing, and both must
read as unknown rather than as a full disk. `ctl nodes` and the dashboard show
a dash, the metric is absent rather than zero, and a missing directory — which
is what an unmounted filer looks like — is skipped rather than reported as
having no space.
Shown as free out of total, because the first is useless without the second:
8 GB free is comfortable on a 10 TB filer and nearly full on a 10 GB root
partition, and only the second stops a model being placed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
It had one verdict for a failed attempt — "This is the engine, not the model.
Choosing a bigger model will not help" — and that verdict was wrong, in a way
that sends someone off rebuilding llama.cpp when the fix is a one-line catalog
change.
Measured on the cluster. The engine renders the tool prompt correctly: the
<tools> block and the instruction to wrap the call in <tool_call> are both in
the prompt, which I confirmed through the engine's own /apply-template. The
coder models simply do not use that wrapper — qwen2.5-coder-3b puts the call
in a markdown fence, qwen2.5-coder-7b invents <function-call> — so the parser,
which looks for <tool_call>, has nothing to find and leaves it in the content.
qwen2.5-3b-instruct, same engine, same build, same flags, returns a populated
tool_calls array. So it was the model all along, and the advice against trying
another one was the most expensive sentence in the output.
A failed attempt is now two verdicts, told apart by whether the reply used the
wrapper its template asked for. Wrapped and unparsed is genuinely the engine's
parser. Unwrapped is the model, and the fix is one trained for the format
rather than a bigger one from the same family.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The metrics existed and nothing could record them. They shared a listener
with pprof, the management API and the dashboard, and pprof can dump the
process heap where API keys sit in plaintext — so a fleet could be monitored
or keep its heap private, and not both. That is not a choice anyone should
have to make, and it is why every previous answer to "how do I scrape this"
was a tunnel.
metrics_listen serves metrics and health and nothing else, so it can be
reached from wherever Prometheus lives. Metrics stay on the admin listener
too: adding a door is not the same as moving one, and anything already
scraping keeps working.
While there: admin_listen defaulted to :9090, every interface, with a
comment beside it saying not to expose it. A default that contradicts its
own documentation is a footgun, so it is loopback now and someone who wants
otherwise says so.
The handler is nil until the registry is built, and says so rather than
serving an empty page. Ordering has already caught this exact area once —
a collector registered after AdminHandler published nothing at all, silently
— and an empty scrape target looks like a working one.
deploy/monitoring/prometheus.yml has a starting configuration and the alerts
worth having: a node that has stopped reporting, a model that should be
serving and is not, a card kept out of placement by a fault, and a fleet
holding models nobody is asking for.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The feature the idle tracking was for. A cluster with more models than GPUs
can now hold every model in its catalog and only the wanted ones in its
cards: min 0 with a max above it means load on demand, and once nobody has
asked for idle_timeout the replicas stop and the cards go back.
No new field, because min 0 max 1 already reads correctly — keep none
running, allow up to one. Which made min 0 max 0 meaningful for the first
time: configured and deliberately not running. normalize() used to fill a
zero max in, on the reasonable ground that a model with no maximum cannot
run; that would now silently turn every switched-off model into an
on-demand one, including the ones already written that way in this
cluster's own configuration. A model with no scaling stanza stays off for
the same reason — defaulting it to on-demand would let any client request
start a 70B nobody intended to run.
The one real decision is what happens to the request that triggers a load.
Refusing makes the feature useless, since the first request for every model
fails; waiting without a bound turns a request into a job, and a client that
budgeted thirty seconds holds a connection for ten minutes. So it waits
briefly and then says what is happening — 503 with Retry-After and "the
model is loading", which is a different answer from "no endpoint serves
this" and produces different client behaviour. An unknown model still gets
the unknown-model error, or on-demand would turn every typo into "loading"
and the useful message would disappear.
Use keeps a model alive rather than only the request that started it,
without which the timer measures time since the load and a busy model is
evicted on schedule however hard it is working.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
42fd0aa deployed and published nothing. RegisterMetrics appends to a slice
that AdminHandler reads when it constructs the registry, and I called it
after — a comment on the function says "called before AdminHandler, because
the registry is built there", written by me, immediately above the code
that did not.
Nothing failed. The build was clean, the tests passed, the endpoint
returned 200 with fourteen kilobytes of metrics, and not one of them was a
node. Only grepping the live endpoint for what was supposed to be there
showed it.
So there is now a test that a registered collector reaches /metrics, which
is the assertion that was missing: everything around it was tested and the
wiring between them was not.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The router published twenty-two metric families about requests, endpoints,
admission and scoring — everything about traffic and nothing about the
machines carrying it. There was no way to graph GPU usage, watch a node go
unhealthy, see a fabric fill up, or answer "was the cluster full at the
time", which is the first question anyone asks after an incident. The
agents exported nothing at all.
Exported by the control plane rather than by each agent. It already holds
every number: hardware arrives at registration and state on each heartbeat,
so this is a map read away. One scrape target for a fleet of hundreds, no
Prometheus configuration per node, no firewall rule per machine, and
nothing new running on the nodes. On the same /metrics as the traffic
figures, because correlating "the cluster was full" with "requests were
queueing" should not need two systems.
A collector rather than gauges kept current by a ticker: the numbers are
derived from state that already exists, so computing them when asked is
simpler than maintaining a copy that can drift, and nothing runs between
scrapes.
GPU utilisation is new, and is the number people actually watch. Memory
used says a model is loaded; this says whether it is doing anything, and a
fleet at ninety percent occupancy and two percent utilisation is a fleet
holding models nobody is asking for.
Adding it caught two things. utilization.gpu goes last in the nvidia-smi
query and is optional, because a driver too old to know it returns a
shorter row — requiring it made an llmgw upgrade silently stop discovering
GPUs on an older node, which is far worse than not knowing how busy a card
is. And I gave the new protobuf field number 14, which ecc_errors already
had; buf lint is why that was a moment rather than a corrupted wire format.
Absent is kept distinct from zero throughout, for the same reason as
everywhere else here: an unreported utilisation shown as a flat zero reads
as an idle fleet rather than a missing measurement, and a replica nothing
has reported would be the most idle thing in the cluster.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Storage will be different at every installation — Lustre, NFS over RDMA, an
image exported with SRP or NVMe-oF, local NVMe, an rsync — so llmgw observes
it rather than managing it. An agent told which directories to look in
reports what it can see; how the files got there is nobody's business here,
and a node told nothing reports nothing and behaves exactly as every node
did before this existed.
That fact buys two things. Placement prefers a node that already holds the
model over one that must first move tens of gigabytes, as a preference and
never a filter: fetching them is a valid answer, and refusing to place would
turn a slow start into no start at all. And ctl status shows what each
fabric holds.
Per fabric because fabrics are not connected to one another. Weights in
ib-a are not weights in ib-b — the only path between them is the management
network — so a model seeded into one fabric leaves the rest of the cluster
unable to serve it without a slow copy, and nothing said so before now. One
shared filer counts the same as twenty local copies, because "can this
fabric serve it" is the question being asked.
Three distinctions the tests hold to, because each is a way to get this
quietly wrong. A node that was never asked to look must rank with one that
looked and has nothing, or enabling the feature on a single node would
drain every other. An empty directory is not weights, because a
half-finished download looks exactly like one and would send placement to a
node whose engine then fails to load. And local storage is never mistaken
for shared: a filer read as local costs an unnecessary copy, while local
read as shared would have the cluster believe a model is everywhere when it
is on one node.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Reading the tool's own output against the real fleet: it chose q8_0 as the
best fit and emitted a vLLM profile with no quantisation. q8_0 is a GGUF
quantisation that only llama.cpp serves, and from a file rather than a
repository — so following that entry loads full-precision weights, which
the same output had just listed as not fitting one card. The planner would
have caused the out-of-memory it exists to prevent, from a recommendation
that looked right.
Choosing a precision and choosing an engine turn out to be one decision, so
a Precision now carries the engine that serves it, the quantisation flag
where one is needed, and whether it lives in a file. A GGUF fit prints a
path and says to download it; a vLLM fit prints the repository and passes
--quantization so the engine does not quietly load bf16.
The arithmetic was right and the recommendation was wrong, which is the
failure mode worth having tests for: everything reported a number, and the
numbers agreed with each other.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
llmgw ctl model plan Qwen/Qwen2.5-Coder-32B-Instruct
The arithmetic this replaces is the last thing between having GPUs and
serving a model on them. Choosing tp, gpu_class and a context that will
actually fit means knowing the layer count, the KV head count, what the
weights weigh at a chosen quantisation and how much of a card the runtime
keeps — and getting it wrong is not an error message, it is an
out-of-memory twenty minutes into a load. I computed those by hand for
every model on this cluster.
Both halves are read rather than retyped: the architecture from the model's
own config.json and parameter count, the hardware from the running fleet.
Attention heads in place of KV heads is the classic slip and is worth a
factor of five on a model with grouped-query attention, so it is not a
number anyone should be entering.
Every candidate is listed, not one answer. With cards to spare you want the
full-precision fit; with one card free you want to know a quantised one
exists. Choosing between those is not the planner's job.
The tests are calibrated against files actually downloaded here — llama.cpp
reported 14.62 GiB and 8.37 GiB for the 14B at q8_0 and q4_K_M, and the
model reproduces both — and they caught two errors in my own arithmetic.
The overhead was charged per rank from the whole model's weights rather
than each rank's share, and it reserved a tenth of a card that
usableFraction had already reserved. Between them, Llama-3.3-70B at bf16
across two 80 GB cards came out as not fitting, which is a configuration
people run every day. Slack reserved in two places, each unaware of the
other, reads as caution and behaves as a bug.
Pools are grouped by class and fabric together, because a replica may never
span two fabrics: eight free cards split across two of them are not eight
cards for a model that needs eight, and merging them would plan something
that starts, passes its health checks and hangs on its first all-reduce.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Reported after watching a model appear in ctl models before it could answer
anything. Checked: clients are not told early — /v1/models listed it only
once a replica was ready, and the chat picker reads the same source. The
operator view showing it is deliberate, because a model coming up is
exactly what someone running the cluster wants to see.
But the report pointed at something real. That view distinguished "a
replica is coming up" from "nothing could be placed" by a parenthetical —
"0/1 ready (1 total)" against "0/1 ready" — and those need opposite
responses: one means wait, the other means look at explain-placement. On
four models the difference is findable. On forty it is invisible, and a
model that will never start reads exactly like one that is thirty seconds
away.
SERVING now answers the question that was being asked of the replica counts
— can a client use this right now — and STATUS says which of ready,
starting, not placed or stopped is the case.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The last piece of not restarting a control plane to try a model:
llmgw ctl model add -f qwen-coder-32b.yaml
llmgw ctl model remove qwen-coder-32b
The file is the same YAML a model takes in server.yaml, read by the same
koanf parser through the same tags, so an entry moves between the two by
being copied and a field that works in one cannot silently mean nothing in
the other.
Wired only when there is a store. A control plane without one says so
rather than accepting a model that vanishes on the next restart, which is
the kind of success that wastes an afternoon. At startup the catalog is
rebuilt from configuration plus what was stored, so models added before a
restart are still there after it — without that the whole feature would be
a slower way to lose work.
Every failure adding a model is the definition being wrong — a collision, a
missing fallback, a profile that cannot be placed — so they are 400s. A 500
would send someone to the server's logs for a mistake in their own file.
The actor comes from the caller and is kept with the row, so a catalog that
has grown over months can say where each entry came from.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Adding a model was an operator procedure with downtime: edit a file on the
control plane, restart it, wait out the settling window, watch replicas
churn while adoption catches up. I did that eight times in a day, and one
of the restarts cost two minutes of outage while stalled replicas timed
out. None of it was necessary. The reconciler already compares desired
against actual on every pass, so a new catalog is simply a different
desired.
A Catalog stays immutable — New is the only way to build one and it refuses
an invalid one — and what changes is which immutable catalog is current.
Readers take the pointer and finish their operation against it, so a swap
mid-request is invisible rather than a race. The router is told rather than
asked, the same shape as the endpoint list and for the same reason: a push
happens once per change, a pull would happen once per request.
The router still seeds from configuration as well as being pushed to. A
router with a static endpoint list and no control plane is a supported mode
and has nothing to push it a catalog; without seeding it would resolve no
model at all.
Which turned up a bug that was latent while catalogs were fixed.
reconcileModel only ever visits models that are in the catalog, so a
replica whose model had been removed was visited by nothing: it kept its
GPUs, kept counting, and kept running an engine for a name no client could
ask for. The endpoint publisher skipped such a replica with the comment
"the reconciler will tear it down", which was not true. It is now, and the
comment says what actually happens instead of what was assumed.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Two things a cluster with many nodes and many GPUs needs that were missing,
and they are the same thing seen from two distances.
Idle tracking. The control plane knew what was placed and whether it was
healthy, never whether anyone wanted it — so on a fleet with more models
than cards there was no way to answer the first question anyone asks, which
is what can I stop. The router is the only thing that knows, so it now
reports it: a timestamp written in Acquire, where a request already meets an
endpoint under a lock already held, and collected once a tick as a map. That
shape is the point. Per-request notification would be a firehose carrying
one bit — the newest timestamp wins — and this costs the same on four
replicas as on four hundred.
It travels through an interface in the direction opposite EndpointSink, for
the same reason that one exists: the control plane decides what should run,
the router knows whether anyone wants it, and neither should need to be in
the other's process to say so.
Unknown is kept distinct from idle throughout. A router that has just
restarted knows nothing, and a zero timestamp would be the oldest idle
replica in every comparison — so one restart would nominate an entire fleet
for eviction. Absent from the map, "-" in the table, negative in the JSON,
and IdleFor returns a second value rather than a duration anything could
act on by mistake.
ctl status. A per-node listing is right for four nodes and wrong for two
hundred: how full am I, what is broken, what is holding cards nobody is
using are aggregates, and reading them off a long table is a chore, which
means it gets skipped. Capacity per GPU class with usage bars, free whole
nodes per fabric — the number that decides whether a multi-node replica can
land, since free cards on the wrong fabric are not capacity for a model that
spans nodes — replica and model counts, models that cannot be placed, and
what has gone unasked-for. Derived from the same state the tables show, so
it cannot disagree with them.
Unhealthy GPUs count as neither used nor free, because counting them free is
how a cluster looks like it has room it does not.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The catalog now declares whether a model supports tools, which makes it a
claim. This is the claim that can be checked, and the one worth checking:
the gateway forwards a tools array to any model, and an engine that does
not return structured tool_calls looks exactly like a model that ignores
its tools. Every OpenAI-compatible client then behaves as though the model
is broken.
It separates the two diagnoses, which have the same symptom and different
fixes. The model ignoring its tools is fixed by choosing another model. The
model making the call and the engine handing it back as prose is fixed by
changing engine build, and no larger model will help — which is what
today's investigation established only after eliminating the gateway, the
model size, two chat templates and tool_choice in turn.
Against the live engine it says:
the model called the tool and the engine did not parse it.
{"name": "get_weather", "arguments": {"city": "Istanbul"}}
This is the engine, not the model.
engine-check asks whether an engine still accepts our flags; this asks
whether it behaves. Flags were the failure we had seen, so they were the
failure we checked for. Non-zero exit, so a rollout can be gated on it.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Placement had no idea how much memory a node has or a model needs. A CPU
profile has no card to be free, so eligibleNodes accepted every ready node:
Qwen2.5-Coder 14B, whose working set is about 24 GB, would have been placed
just as happily on one of the 8 GB nodes, which reports itself perfectly
healthy right up until the OOM killer arrives.
node_selector was yesterday's stopgap and it is the operator restating by
hand something the system can know. Agents now report MemTotal, profiles
declare min_memory_gb, and placement filters on it, naming what was short —
"has 8 GB of memory, needs 24" is actionable where "no candidates were
considered" is not. The label selector stays, because "this node is for
coding models" is a real thing to want to say that memory does not express.
Two deliberate non-exclusions. A profile with no requirement fits anywhere,
which is every profile that exists today. And a node that reports no memory
is still usable: an agent older than the field says nothing, and letting a
control plane upgrade quietly empty half a fleet is a worse failure than the
one this prevents.
ctl nodes grows a MEM column showing "-" rather than 0 for a node that did
not say, because an old agent is not a small machine.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Clicking "metrics" on the web dashboard answered "no api key was provided".
The page is served twice and its navigation assumed one of them. Metrics,
pprof and the management API exist on the admin listener; account
administration, the profile and the chat exist on the client-facing port. The
nav offered the first set in both places, so on the public port those links
fell through to the API-key middleware — which is a far more confusing way to
say "that is not here" than a 404 would have been.
Each page now offers what is actually reachable from it, and the public one
says where the rest lives rather than pretending it does not exist, including
why that port is not exposed: pprof can dump the process heap, and the heap
holds every API key in plaintext.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
