<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
<title>production-stack: releases and problems | Archestack</title>
<link>https://archestack.com/best-of-ai/projects/production-stack/</link>
<description>New releases (breaking changes and security fixes flagged), newly reported problems and fixed issues for production-stack, from its public GitHub data.</description>
<language>en</language>
<lastBuildDate>Sat, 07 Nov 2026 00:00:00 GMT</lastBuildDate>
<atom:link href="https://archestack.com/best-of-ai/projects/production-stack/feed.xml" rel="self" type="application/rss+xml"/>
<item>
<title>production-stack vllm-stack-0.1.14</title>
<link>https://github.com/vllm-project/production-stack/releases/tag/vllm-stack-0.1.14</link>
<guid isPermaLink="false">archestack-watch-release-best-of-ai/production-stack-vllm-stack-0.1.14</guid>
<pubDate>Sat, 07 Nov 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;production-stack vllm-stack-0.1.14 was released on 2026-11-07. Its notes do not flag a breaking change or a security fix; that does not mean the upgrade is safe.&lt;/p&gt;</description>
</item>
<item>
<title>production-stack vllm-stack-0.1.13</title>
<link>https://github.com/vllm-project/production-stack/releases/tag/vllm-stack-0.1.13</link>
<guid isPermaLink="false">archestack-watch-release-best-of-ai/production-stack-vllm-stack-0.1.13</guid>
<pubDate>Tue, 29 Sep 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;production-stack vllm-stack-0.1.13 was released on 2026-09-29. Its notes do not flag a breaking change or a security fix; that does not mean the upgrade is safe.&lt;/p&gt;</description>
</item>
<item>
<title>New problem reported: Static-discovery /v1/models drops backend metadata — no way to advertise max_model_len</title>
<link>https://github.com/vllm-project/production-stack/issues/1049</link>
<guid isPermaLink="false">archestack-watch-problem-best-of-ai/production-stack-1049</guid>
<pubDate>Mon, 24 Aug 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;Open on GitHub, 1 reaction: &lt;a href=&quot;https://github.com/vllm-project/production-stack/issues/1049&quot;&gt;Static-discovery /v1/models drops backend metadata — no way to advertise max_model_len&lt;/a&gt;&lt;/p&gt;</description>
</item>
<item>
<title>Fixed: bug: CRD VllmRuntime : increase shm size</title>
<link>https://github.com/vllm-project/production-stack/issues/899</link>
<guid isPermaLink="false">archestack-watch-fixed-best-of-ai/production-stack-899</guid>
<pubDate>Tue, 28 Jul 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;&lt;a href=&quot;https://github.com/vllm-project/production-stack/issues/899&quot;&gt;bug: CRD VllmRuntime : increase shm size&lt;/a&gt; closed by &lt;a href=&quot;https://github.com/vllm-project/production-stack/pull/994&quot;&gt;feat(operator): support shmSize on VLLMRuntime for tensor parallelism (#899)&lt;/a&gt;.&lt;/p&gt;</description>
</item>
<item>
<title>production-stack vllm-stack-0.1.12</title>
<link>https://github.com/vllm-project/production-stack/releases/tag/vllm-stack-0.1.12</link>
<guid isPermaLink="false">archestack-watch-release-best-of-ai/production-stack-vllm-stack-0.1.12</guid>
<pubDate>Fri, 24 Jul 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;production-stack vllm-stack-0.1.12 was released on 2026-07-24. Its notes do not flag a breaking change or a security fix; that does not mean the upgrade is safe.&lt;/p&gt;</description>
</item>
<item>
<title>Fixed: bug: Router - Stale Prometheus endpoint health metrics when engines become unavailable in service discovery</title>
<link>https://github.com/vllm-project/production-stack/issues/888</link>
<guid isPermaLink="false">archestack-watch-fixed-best-of-ai/production-stack-888</guid>
<pubDate>Thu, 14 May 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;&lt;a href=&quot;https://github.com/vllm-project/production-stack/issues/888&quot;&gt;bug: Router - Stale Prometheus endpoint health metrics when engines become unavailable in service discovery&lt;/a&gt; closed by &lt;a href=&quot;https://github.com/vllm-project/production-stack/pull/936&quot;&gt;[Bugfix][Router] Fix stale Prometheus endpoint health metrics (#888)&lt;/a&gt;.&lt;/p&gt;</description>
</item>
<item>
<title>Fixed: bug: Router does not authenticate with vllmApiKey to /v1/audio/transcriptions in v0.1.10</title>
<link>https://github.com/vllm-project/production-stack/issues/910</link>
<guid isPermaLink="false">archestack-watch-fixed-best-of-ai/production-stack-910</guid>
<pubDate>Thu, 23 Apr 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;&lt;a href=&quot;https://github.com/vllm-project/production-stack/issues/910&quot;&gt;bug: Router does not authenticate with vllmApiKey to /v1/audio/transcriptions in v0.1.10&lt;/a&gt; closed by &lt;a href=&quot;https://github.com/vllm-project/production-stack/pull/914&quot;&gt;[Bugfix][Router] Fix router auth for transcription proxy&lt;/a&gt;.&lt;/p&gt;</description>
</item>
<item>
<title>Fixed: bug: Istio compression messes up the router-engine connection and user receives decompressed body with content-encoding: &quot;gzip&quot;</title>
<link>https://github.com/vllm-project/production-stack/issues/827</link>
<guid isPermaLink="false">archestack-watch-fixed-best-of-ai/production-stack-827</guid>
<pubDate>Sat, 21 Feb 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;&lt;a href=&quot;https://github.com/vllm-project/production-stack/issues/827&quot;&gt;bug: Istio compression messes up the router-engine connection and user receives decompressed body with content-encoding: &amp;quot;gzip&amp;quot;&lt;/a&gt; closed by &lt;a href=&quot;https://github.com/vllm-project/production-stack/pull/836&quot;&gt;fix: filter hop-by-hop headers from streaming responses&lt;/a&gt;.&lt;/p&gt;</description>
</item>
<item>
<title>Fixed: bug: Router log level is unchangeable, and default in log.py is DEBUG</title>
<link>https://github.com/vllm-project/production-stack/issues/817</link>
<guid isPermaLink="false">archestack-watch-fixed-best-of-ai/production-stack-817</guid>
<pubDate>Fri, 13 Feb 2026 00:00:00 GMT</pubDate>
<description>&lt;p&gt;&lt;a href=&quot;https://github.com/vllm-project/production-stack/issues/817&quot;&gt;bug: Router log level is unchangeable, and default in log.py is DEBUG&lt;/a&gt; closed by &lt;a href=&quot;https://github.com/vllm-project/production-stack/pull/832&quot;&gt;fix: make --log-level CLI argument actually control router log levels&lt;/a&gt;.&lt;/p&gt;</description>
</item>
</channel>
</rss>
