fshy

fshy / Storage

Clojars

clojars.org · Storage · 1 record · JSON

Clojars (repo.clojars.org) Storage

fails: quotacountry-phoneno cardphoneRper-object-cap

clojars.org · registry · storage/clojars · JSON · compare

Free tierNo published byte quota on the site pages read. A 2020-era project commit records an 81 MB uberjar being rejected as 'too large for clojars', and a later commit adds a client-side 'uberjar size guard' - so there is a per-artifact ceiling in the tens-of-MB, historically around 100 MB.
unit MB; metered on stored-bytes
Pricenone stated
Per-object capper-object-cap: ~100 MB (100000000 bytes)
Retentionper-object-cap: ~100 MB, from third-party evidence only (81 MB jar rejected). retention: indefinite (Maven-layout artifacts are immutable once deployed)
EgressFree. repo.clojars.org is S3/CloudFront-backed ('Server: ... NoSuchKey' XML errors expose S3 origin) and serves without metering.
Identitycountry-phone derived from the record's prose, not from an identity field, and no signup was executed: n the site pages read. A 2020-era project commit records an 81 MB uberjar being rejected as 'too large for clojars', and a later commit adds a client-side 'uberjar size guard' - so t
Limits
  • per-object-cap per-object-cap: ~100 MB source
per-object-cap: ~100 MB, from third-party evidence only (81 MB jar rejected). retention: indefinite (Maven-layout artifacts are immutable once deployed). request-rate: not documented. Write path is authenticated lein deploy with a Clojars token.
Bindingper-object-cap the per-object cap sets the shard size and therefore the object count; it bounds the total more tightly than any headline quota
Interfacerest, rclone rest
Toolinglein deploy clojars, clj -T:build deploy / deps-deploy, mvn deploy with the clojars repo, and plain HTTPS/Maven layout GETs. No rclone backend.
Composes withunlimited artifacts (group/artifact/version) x unlimited versions x unlimited Clojars accounts (signup with an SSH key, free). At ~100 MB/artifact that is 100 MB x N; 20 TiB would need ~200,000 artifacts. Trivially repeatable in principle, brutally inefficient.
ReliabilityRead path serves real jars (verified 3,224 and 16,324 bytes on real artifacts). The cap is the story and it is poorly documented - the only hard numbers come from a 2020 project commit and forum traffic. Confirming the current number would need an authenticated deploy probe.
NotesNo published byte quota on the site pages read. A 2020-era project commit records an 81 MB uberjar being rejected as 'too large for clojars', and a later commit adds a client-side 'uberjar size guard' - so there is a per-artifact ceiling in the tens-of-MB, historically around 100 MB.
Evidence rungreported One page read moves it to documented.
Checked2026-10-21 (inherited); stale on 2026-11-04
Parsed numbers
  • $.limits[0].value limit:per-object-cap 100.0 MB
Evidence (4)
datesourcewhat it proved
2026-10-21repo.clojars.org/metosin/reitit/0.7.2/reitit-0.7.2.jarprobe: GET -> 200, 3,224 bytes
2026-10-21repo.clojars.org/hiccup/hiccup/1.0.5/hiccup-1.0.5.jarprobe: GET -> 200, 16,324 bytes
2026-10-21github.com/IntraCel/core/commit/568d4758f5994ac0aa4b7d015af7'the uberjar task produced an 81MB jar file which was too large for clojars'
2026-10-21github.com/pokle/clojr/commit/7c8b2b9b755f751b2f50cbe9e1fa97'add uberjar size guard to deploy'
More fields (1)
{
 "dead_end": {
  "failure_class": "quota",
  "why": "No published byte quota on the site pages read. A 2020-era project commit records an 81 MB uberjar being rejected as 'too large for clojars', and a later commit adds a client-side 'uberjar size guard' - so there is a per-artifact ceiling in the tens-of-MB, historically around 100 MB.",
  "retry_when": null,
  "first_recorded": "2026-10-21"
 }
}
Unknown or withheld values (7)
  • $.extras.dead_end.retry_when process
  • $.extras.dead_end.why partly-held
  • $.free_tier.notes process
  • $.free_tier.stated partly-held
  • $.notes partly-held
  • $.per_object_cap.notes not-extracted
  • $.retention.notes not-extracted