Initial Stages

clusterwide-file-manifest/8-1-3/source/errs/manifest-node02-dc1.txt.err

@@ -0,0 +1 @@
+md5sum: /var/cache/salt/minion/proc/20260216150131518310: No such file or directory

clusterwide-file-manifest/8-2/source/err/manifest-node01-dc1.txt.err

@@ -0,0 +1,103 @@
+md5sum: /var/cache/salt/master/jobs/01/312f959b21aac3a07b1911b168bdd043d22a702d2f2939013fae3c868b1693/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/01/312f959b21aac3a07b1911b168bdd043d22a702d2f2939013fae3c868b1693/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/01/312f959b21aac3a07b1911b168bdd043d22a702d2f2939013fae3c868b1693/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/01/312f959b21aac3a07b1911b168bdd043d22a702d2f2939013fae3c868b1693/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/05/17d86915b811bf02578c15fc76388cdb61d2674645480d0ae9f3378c0c71fa/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/05/17d86915b811bf02578c15fc76388cdb61d2674645480d0ae9f3378c0c71fa/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/05/17d86915b811bf02578c15fc76388cdb61d2674645480d0ae9f3378c0c71fa/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/05/17d86915b811bf02578c15fc76388cdb61d2674645480d0ae9f3378c0c71fa/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/06/36f2d9524cd433a2321fb1c3427f6e8779ff7655baa42e4766c22aef79b04f/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/06/36f2d9524cd433a2321fb1c3427f6e8779ff7655baa42e4766c22aef79b04f/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/06/36f2d9524cd433a2321fb1c3427f6e8779ff7655baa42e4766c22aef79b04f/4b78f943-203d-4ae5-b4fe-dae35bd4d157/out.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/06/36f2d9524cd433a2321fb1c3427f6e8779ff7655baa42e4766c22aef79b04f/4b78f943-203d-4ae5-b4fe-dae35bd4d157/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/06/36f2d9524cd433a2321fb1c3427f6e8779ff7655baa42e4766c22aef79b04f/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/2b/06e64cb00c211a61e72ae765a8f87618692d467d271254db2ff1d6476a266a/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/2b/06e64cb00c211a61e72ae765a8f87618692d467d271254db2ff1d6476a266a/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/2b/06e64cb00c211a61e72ae765a8f87618692d467d271254db2ff1d6476a266a/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/2b/06e64cb00c211a61e72ae765a8f87618692d467d271254db2ff1d6476a266a/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/3f/123e0ca05ed360884fbc67a74834060c46da6a3216df587ee939e8722e96a0/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/3f/123e0ca05ed360884fbc67a74834060c46da6a3216df587ee939e8722e96a0/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/3f/123e0ca05ed360884fbc67a74834060c46da6a3216df587ee939e8722e96a0/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/3f/123e0ca05ed360884fbc67a74834060c46da6a3216df587ee939e8722e96a0/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/49/c0fdef6821532e4b2fa74d59945242ba23f1f583dd79d80012289150188098/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/49/c0fdef6821532e4b2fa74d59945242ba23f1f583dd79d80012289150188098/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/49/c0fdef6821532e4b2fa74d59945242ba23f1f583dd79d80012289150188098/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/49/c0fdef6821532e4b2fa74d59945242ba23f1f583dd79d80012289150188098/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/50/5b1c4e5a6093c895296969a8e8b6305b31326d05e20208d30b5b38d9ed9688/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/50/5b1c4e5a6093c895296969a8e8b6305b31326d05e20208d30b5b38d9ed9688/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/50/5b1c4e5a6093c895296969a8e8b6305b31326d05e20208d30b5b38d9ed9688/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/50/5b1c4e5a6093c895296969a8e8b6305b31326d05e20208d30b5b38d9ed9688/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/51/c0a9a331599fdb4bddd5b1597ead17ba1f96ea1799ef4133480a5ae2278929/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/51/c0a9a331599fdb4bddd5b1597ead17ba1f96ea1799ef4133480a5ae2278929/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/51/c0a9a331599fdb4bddd5b1597ead17ba1f96ea1799ef4133480a5ae2278929/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/51/c0a9a331599fdb4bddd5b1597ead17ba1f96ea1799ef4133480a5ae2278929/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/58/7a995a41151b396d1c3b570d598a56804aaea767691e11a8ab59645e9c52cd/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/58/7a995a41151b396d1c3b570d598a56804aaea767691e11a8ab59645e9c52cd/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/58/7a995a41151b396d1c3b570d598a56804aaea767691e11a8ab59645e9c52cd/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/58/7a995a41151b396d1c3b570d598a56804aaea767691e11a8ab59645e9c52cd/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/5e/cefd947d3e51090dcce6cf1fc4bbec19d2680b96982902d5146b957803b7dd/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/5e/cefd947d3e51090dcce6cf1fc4bbec19d2680b96982902d5146b957803b7dd/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/5e/cefd947d3e51090dcce6cf1fc4bbec19d2680b96982902d5146b957803b7dd/a30b5c95-29c4-4b8e-998c-2013a15d9333/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/5e/cefd947d3e51090dcce6cf1fc4bbec19d2680b96982902d5146b957803b7dd/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/1608dfce11b6c1cffa8c79cfbeb117b75d4abe85c73f231a43d0621e7f313a/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/1608dfce11b6c1cffa8c79cfbeb117b75d4abe85c73f231a43d0621e7f313a/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/1608dfce11b6c1cffa8c79cfbeb117b75d4abe85c73f231a43d0621e7f313a/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/1608dfce11b6c1cffa8c79cfbeb117b75d4abe85c73f231a43d0621e7f313a/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/2f74f122fd04ca11181d3493c2d106cb01aaa6c3bc031d555980a18fc9dd87/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/2f74f122fd04ca11181d3493c2d106cb01aaa6c3bc031d555980a18fc9dd87/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/2f74f122fd04ca11181d3493c2d106cb01aaa6c3bc031d555980a18fc9dd87/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/72/2f74f122fd04ca11181d3493c2d106cb01aaa6c3bc031d555980a18fc9dd87/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/84/b8a292785548744e8c08fa2054db5afb24ab2598e3c0b6da6282cd24596dc3/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/84/b8a292785548744e8c08fa2054db5afb24ab2598e3c0b6da6282cd24596dc3/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/84/b8a292785548744e8c08fa2054db5afb24ab2598e3c0b6da6282cd24596dc3/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/84/b8a292785548744e8c08fa2054db5afb24ab2598e3c0b6da6282cd24596dc3/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/95/93687fae345fb23193b151e69b9c91a6b453fa27f4534b6a46357a2ff9fda0/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/95/93687fae345fb23193b151e69b9c91a6b453fa27f4534b6a46357a2ff9fda0/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/95/93687fae345fb23193b151e69b9c91a6b453fa27f4534b6a46357a2ff9fda0/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/95/93687fae345fb23193b151e69b9c91a6b453fa27f4534b6a46357a2ff9fda0/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/aa/ec3b0943ecf723a0a4d126e4b7d161100308a8c5c19efab09985323011206c/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/aa/ec3b0943ecf723a0a4d126e4b7d161100308a8c5c19efab09985323011206c/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/aa/ec3b0943ecf723a0a4d126e4b7d161100308a8c5c19efab09985323011206c/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/aa/ec3b0943ecf723a0a4d126e4b7d161100308a8c5c19efab09985323011206c/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/b4/349af986e29ef410c551a4de924ffbf1afd4e4941eb355d2a9f2b254c31bab/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/b4/349af986e29ef410c551a4de924ffbf1afd4e4941eb355d2a9f2b254c31bab/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/b4/349af986e29ef410c551a4de924ffbf1afd4e4941eb355d2a9f2b254c31bab/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/b4/349af986e29ef410c551a4de924ffbf1afd4e4941eb355d2a9f2b254c31bab/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bb/e4b07a318619ee776d7a041dd1437fc687435ac1963fef0f9c06dd67dd92a8/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bb/e4b07a318619ee776d7a041dd1437fc687435ac1963fef0f9c06dd67dd92a8/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bb/e4b07a318619ee776d7a041dd1437fc687435ac1963fef0f9c06dd67dd92a8/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bb/e4b07a318619ee776d7a041dd1437fc687435ac1963fef0f9c06dd67dd92a8/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bd/d07ad5212a593b54f58ffa22b9ea2d7df84f4793e4de0f84b4405b802d6930/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bd/d07ad5212a593b54f58ffa22b9ea2d7df84f4793e4de0f84b4405b802d6930/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bd/d07ad5212a593b54f58ffa22b9ea2d7df84f4793e4de0f84b4405b802d6930/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/bd/d07ad5212a593b54f58ffa22b9ea2d7df84f4793e4de0f84b4405b802d6930/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c3/c379258b5e0a20ceae27c99701df65c5a2d9aa28ebad703da858079e0c4aa2/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c3/c379258b5e0a20ceae27c99701df65c5a2d9aa28ebad703da858079e0c4aa2/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c3/c379258b5e0a20ceae27c99701df65c5a2d9aa28ebad703da858079e0c4aa2/9398ab4f-fca3-4ad5-b0cc-d1b57326dc94/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c3/c379258b5e0a20ceae27c99701df65c5a2d9aa28ebad703da858079e0c4aa2/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c6/318e9ff9714ba3561bd8f2c85ccf9b6e42c38449ff84150e5219a40599185f/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c6/318e9ff9714ba3561bd8f2c85ccf9b6e42c38449ff84150e5219a40599185f/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c6/318e9ff9714ba3561bd8f2c85ccf9b6e42c38449ff84150e5219a40599185f/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/c6/318e9ff9714ba3561bd8f2c85ccf9b6e42c38449ff84150e5219a40599185f/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/ca/edd63c29a5fea6bd83ac08b7d55d306ba96644750e36907b690d546de3d510/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/ca/edd63c29a5fea6bd83ac08b7d55d306ba96644750e36907b690d546de3d510/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/ca/edd63c29a5fea6bd83ac08b7d55d306ba96644750e36907b690d546de3d510/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/ca/edd63c29a5fea6bd83ac08b7d55d306ba96644750e36907b690d546de3d510/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/d5/fd88747e5b396a6aa45bf614bba3d1f5ed5e68b321d6c72c6a0e543faf37f7/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/d5/fd88747e5b396a6aa45bf614bba3d1f5ed5e68b321d6c72c6a0e543faf37f7/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/d5/fd88747e5b396a6aa45bf614bba3d1f5ed5e68b321d6c72c6a0e543faf37f7/4b78f943-203d-4ae5-b4fe-dae35bd4d157/out.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/d5/fd88747e5b396a6aa45bf614bba3d1f5ed5e68b321d6c72c6a0e543faf37f7/4b78f943-203d-4ae5-b4fe-dae35bd4d157/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/d5/fd88747e5b396a6aa45bf614bba3d1f5ed5e68b321d6c72c6a0e543faf37f7/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/e8/2d9fc5b4f43343b89ee05583e1b30489dbebe28fee01b6a677126e3f5d6c68/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/e8/2d9fc5b4f43343b89ee05583e1b30489dbebe28fee01b6a677126e3f5d6c68/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/e8/2d9fc5b4f43343b89ee05583e1b30489dbebe28fee01b6a677126e3f5d6c68/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/e8/2d9fc5b4f43343b89ee05583e1b30489dbebe28fee01b6a677126e3f5d6c68/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/f4/6b7cb33be729defed16e3b50595c8da005018d7d61f30c33232e732627e5f5/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/f4/6b7cb33be729defed16e3b50595c8da005018d7d61f30c33232e732627e5f5/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/f4/6b7cb33be729defed16e3b50595c8da005018d7d61f30c33232e732627e5f5/4b78f943-203d-4ae5-b4fe-dae35bd4d157_master/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/f4/6b7cb33be729defed16e3b50595c8da005018d7d61f30c33232e732627e5f5/jid: No such file or directory
+md5sum: /var/cache/salt/master/jobs/fc/2e0a5e7b4023da0694f5b1f85a29e3be1ff28be753506d7c37e89f3cbcf76e/.load.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/fc/2e0a5e7b4023da0694f5b1f85a29e3be1ff28be753506d7c37e89f3cbcf76e/.minions.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/fc/2e0a5e7b4023da0694f5b1f85a29e3be1ff28be753506d7c37e89f3cbcf76e/9398ab4f-fca3-4ad5-b0cc-d1b57326dc94/out.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/fc/2e0a5e7b4023da0694f5b1f85a29e3be1ff28be753506d7c37e89f3cbcf76e/9398ab4f-fca3-4ad5-b0cc-d1b57326dc94/return.p: No such file or directory
+md5sum: /var/cache/salt/master/jobs/fc/2e0a5e7b4023da0694f5b1f85a29e3be1ff28be753506d7c37e89f3cbcf76e/jid: No such file or directory

clusterwide-file-manifest/8-2/source/err/manifest-node03-dc1.txt.err

@@ -0,0 +1 @@
+md5sum: /var/cache/salt/minion/proc/20260216124853879978: No such file or directory

readme.md

@@ -0,0 +1,28 @@
+Under clusterwide-file-manifest we should have directory tree of the form
+
+dir - K-Y-Z-M 
+ K-Y-Z-M/clusterwide-manifest.nmd5
+
+These are the clusterwide manifests - i.e manifests that are produced by comparing the md5s of different nodes from the same cluster,
+and producing 2 lists for each of those clusters:
+
+List 1 (matches):
+md5sum:/file/path
+
+List 2 (diffs):
+md5sum:/file/path
+
+Where List 1 is a list of all files that have the same path, and have the same path and md5sum across the cluster
+While list 2 is a list of the inverse of the above.
+
+# Stages
+
+## Stage 1 - Md5:filepath
+ - Param 1 - base path (i.e `stage1.sh /` - would produce md5sum for everything under root)
+ - Produces a file containing the md5:file on a given system
+ - Excludes directories known to contain node-specific directories
+
+## Stage 2 - Base manifest producer -
+ Takes in files produced from Stage 1 across a cluster and produces an output base-manifest
+
+## Stage 3 - Compares a stage 1 output against a known stage 2 manifest 

stages/stage1-gather-node-manifest.sh

@@ -0,0 +1,55 @@
+
+#!/usr/bin/env bash
+# manifest_gen.sh — minimal manifest generator using find + md5sum
+# Default output: /tmp/manifest-<hostname>.txt
+# Format: "<md5>  <absolute_path>"
+
+set -euo pipefail
+
+HOST="$(hostname -s 2>/dev/null || hostname 2>/dev/null || echo unknown)"
+OUT="./manifest-${HOST}.txt"
+ROOT="${1:-/}"
+
+# Default excludes (edit here)
+EXCLUDES=(/proc /sys /dev /run /tmp /var/run /var/lock /lost+found /mnt /media /cloudian* /var/lib/cassandra)
+
+ERR="${OUT}.err"
+: > "$OUT"
+: > "$ERR"
+
+ROOT="$(cd "$ROOT" && pwd -P)"
+
+{
+  echo "# manifest_version=md5sum-v1"
+  echo "# created_utc=$(date -u +%Y-%m-%dT%H:%M:%SZ)"
+  echo "# host=$HOST"
+  echo "# root=$ROOT"
+  echo -n "# excludes="
+  printf "%s " "${EXCLUDES[@]}"
+  echo
+  echo "# format=<md5>  <absolute_path>"
+} >> "$OUT"
+
+# Build prune expression
+PRUNE=()
+for ex in "${EXCLUDES[@]}"; do
+  # normalize excludes if they exist; if they don't, still keep as-is
+  if [ -e "$ex" ]; then
+    ex="$(cd "$ex" 2>/dev/null && pwd -P || echo "$ex")"
+  fi
+  PRUNE+=( -path "$ex" -o -path "$ex/*" -o )
+done
+unset 'PRUNE[${#PRUNE[@]}-1]'  # drop last -o
+
+# Deterministic ordering if available (GNU sort -z)
+if sort -z </dev/null >/dev/null 2>&1; then
+  find "$ROOT" \( "${PRUNE[@]}" \) -prune -o -type f -print0 2>>"$ERR" \
+    | sort -z 2>>"$ERR" \
+    | xargs -0 -r md5sum -- 2>>"$ERR" >> "$OUT"
+else
+  find "$ROOT" \( "${PRUNE[@]}" \) -prune -o -type f -print0 2>>"$ERR" \
+    | xargs -0 -r md5sum -- 2>>"$ERR" >> "$OUT"
+fi
+
+echo "Wrote: $OUT" >&2
+echo "Errs:  $ERR" >&2

stages/stage2-analyze-clusterwide-manifests.py

@@ -0,0 +1,134 @@
+import os
+import glob
+import json
+import argparse
+import sys
+from collections import defaultdict
+
+def parse_node_manifest(filepath):
+    """
+    Parses a single node manifest file.
+    Returns:
+        host (str): The hostname found in the header.
+        file_data (dict): Dictionary { filepath: md5_hash }
+    """
+    host = None
+    file_data = {}
+    
+    try:
+        with open(filepath, 'r') as f:
+            for line in f:
+                line = line.strip()
+                if not line:
+                    continue
+                
+                # Parse Headers
+                if line.startswith('#'):
+                    if line.startswith('# host='):
+                        host = line.split('=')[1].strip()
+                    continue
+                
+                # Parse Data (md5  absolute_path)
+                # We split by whitespace, maxsplit=1 to preserve spaces in filenames if they exist
+                parts = line.split(maxsplit=1)
+                if len(parts) == 2:
+                    md5, path = parts
+                    file_data[path] = md5
+                    
+    except Exception as e:
+        print(f"Error reading file {filepath}: {e}", file=sys.stderr)
+        return None, None
+
+    if not host:
+        print(f"Warning: No '# host=' header found in {filepath}. Using filename as ID.", file=sys.stderr)
+        host = os.path.basename(filepath)
+
+    return host, file_data
+
+def generate_golden_manifest(input_dir, output_file):
+    manifest_files = glob.glob(os.path.join(input_dir, '*'))
+    
+    if not manifest_files:
+        print(f"No files found in directory: {input_dir}")
+        return
+
+    # 1. Aggregation Phase
+    # Structure: global_registry[filepath] = { hostname: hash }
+    global_registry = defaultdict(dict)
+    all_hosts = set()
+
+    print(f"Scanning {len(manifest_files)} manifests...")
+
+    for fpath in manifest_files:
+        # Skip if it's a directory
+        if os.path.isdir(fpath): 
+            continue
+            
+        host, data = parse_node_manifest(fpath)
+        if host and data:
+            all_hosts.add(host)
+            for path, md5 in data.items():
+                global_registry[path][host] = md5
+
+    # 2. Analysis Phase
+    cluster_manifest = {
+        "meta": {
+            "total_hosts": len(all_hosts),
+            "hosts": list(sorted(all_hosts)),
+            "generated_at": str(os.path.basename(output_file))
+        },
+        "consistent_files": {}, # Files identical on ALL hosts
+        "varying_files": {}     # Files that differ or are missing on some hosts
+    }
+
+    print("Analyzing file consistency...")
+
+    for path, host_map in global_registry.items():
+        unique_hashes = set(host_map.values())
+        present_on_hosts = set(host_map.keys())
+        
+        # Condition 1: Consistent
+        # Present on ALL hosts AND has exactly 1 unique hash
+        if present_on_hosts == all_hosts and len(unique_hashes) == 1:
+            # Store just the hash, as it is the "Ground Truth"
+            cluster_manifest["consistent_files"][path] = list(unique_hashes)[0]
+        
+        # Condition 2: Varying
+        else:
+            issue_type = []
+            if len(unique_hashes) > 1:
+                issue_type.append("hash_mismatch")
+            if present_on_hosts != all_hosts:
+                issue_type.append("presence_mismatch")
+            
+            # For varying files, we need the specific details per host
+            # so the comparison tool knows what to expect where.
+            
+            # Fill in "MISSING" for hosts that don't have the file
+            full_map = host_map.copy()
+            for h in all_hosts:
+                if h not in full_map:
+                    full_map[h] = "MISSING"
+
+            cluster_manifest["varying_files"][path] = {
+                "issues": issue_type,
+                "states": full_map
+            }
+
+    # 3. Output Phase
+    try:
+        with open(output_file, 'w') as f:
+            json.dump(cluster_manifest, f, indent=2, sort_keys=True)
+        print(f"Success! Cluster manifest written to: {output_file}")
+        print(f"Stats: {len(cluster_manifest['consistent_files'])} consistent files, {len(cluster_manifest['varying_files'])} varying files.")
+    except Exception as e:
+        print(f"Error writing output file: {e}")
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Generate a cluster-wide ground truth manifest.")
+    parser.add_argument("input_dir", help="Directory containing node manifest files")
+    parser.add_argument("output_file", help="Path to write the resulting JSON manifest")
+    
+    args = parser.parse_args()
+    
+    generate_golden_manifest(args.input_dir, args.output_file)

stages/stage3-compare-sub-manifest-c.py

@@ -0,0 +1,168 @@
+import json
+import argparse
+import sys
+import os
+
+def parse_manifest_file(filepath):
+    """
+    Parses the target sub-manifest file to extract host and file hashes.
+    """
+    host = None
+    file_data = {}
+    
+    try:
+        with open(filepath, 'r') as f:
+            for line in f:
+                line = line.strip()
+                if not line: continue
+                
+                # Extract Header Hostname
+                if line.startswith('#'):
+                    if line.startswith('# host='):
+                        host = line.split('=')[1].strip()
+                    continue
+                
+                # Extract Hash and Path
+                parts = line.split(maxsplit=1)
+                if len(parts) == 2:
+                    md5, path = parts
+                    file_data[path] = md5
+                    
+    except Exception as e:
+        print(f"Critical Error: Could not read manifest file: {e}")
+        sys.exit(1)
+
+    if not host:
+        print("Warning: No '# host=' header found. Validation of 'varying' files will be impossible.")
+    
+    return host, file_data
+
+def compare_against_research(manifest_path, research_path):
+    # 1. Load the Ground Truth (Research)
+    try:
+        with open(research_path, 'r') as f:
+            research = json.load(f)
+    except Exception as e:
+        print(f"Critical Error: Could not load research JSON: {e}")
+        sys.exit(1)
+
+    # 2. Parse the Sub-Manifest (Target)
+    target_host, target_files = parse_manifest_file(manifest_path)
+    
+    print(f"--- VALIDATION REPORT ---")
+    print(f"Target Host: {target_host if target_host else 'UNKNOWN'}")
+    print(f"Files Scanned: {len(target_files)}")
+    print(f"Research Baseline: {research['meta']['total_hosts']} hosts known")
+    print("-" * 30)
+
+    # Buckets for results
+    results = {
+        "ok_consistent": [],      # Matches cluster-wide invariant
+        "ok_varying": [],         # Matches expected value for this specific host
+        "fail_consistent": [],    # Violated a cluster-wide invariant
+        "fail_varying": [],       # Violated the specific expectation for this host
+        "unknown_file": [],       # File not present in original research
+        "unverifiable_host": []   # File is varying, but we don't know this host
+    }
+
+    # 3. The Comparison Logic
+    for fpath, fhash in target_files.items():
+        
+        # CHECK A: Is it a Known Consistent File?
+        if fpath in research['consistent_files']:
+            expected_hash = research['consistent_files'][fpath]
+            if fhash == expected_hash:
+                results['ok_consistent'].append(fpath)
+            else:
+                results['fail_consistent'].append({
+                    "path": fpath, 
+                    "found": fhash, 
+                    "expected": expected_hash
+                })
+
+        # CHECK B: Is it a Known Varying File?
+        elif fpath in research['varying_files']:
+            # We need the hostname to validate varying files
+            if not target_host:
+                results['unverifiable_host'].append(fpath)
+                continue
+
+            node_states = research['varying_files'][fpath]['states']
+            
+            # Does the research know about this host?
+            if target_host in node_states:
+                expected_hash = node_states[target_host]
+                
+                if expected_hash == "MISSING":
+                    # We found a file that research says shouldn't exist on this node
+                    results['fail_varying'].append({
+                        "path": fpath,
+                        "found": fhash,
+                        "expected": "File should not exist on this node"
+                    })
+                elif fhash == expected_hash:
+                    results['ok_varying'].append(fpath)
+                else:
+                    results['fail_varying'].append({
+                        "path": fpath,
+                        "found": fhash,
+                        "expected": expected_hash
+                    })
+            else:
+                # This is a new host not in the original research
+                # We can't validate varying files for unknown hosts
+                results['unverifiable_host'].append(fpath)
+
+        # CHECK C: Unknown File
+        else:
+            results['unknown_file'].append(fpath)
+
+    # 4. Print Summary
+    # OK Sections
+    if results['ok_consistent']:
+        print(f"\n[PASS] Consistent Files Verified: {len(results['ok_consistent'])}")
+    if results['ok_varying']:
+        print(f"[PASS] Varying Files Verified:    {len(results['ok_varying'])}")
+
+    # FAIL Sections
+    if results['fail_consistent']:
+        print(f"\n[FAIL] GLOBAL CONSISTENCY VIOLATIONS ({len(results['fail_consistent'])})")
+        print("These files MUST be identical across the cluster but differed:")
+        for item in results['fail_consistent']:
+            print(f"  X {item['path']}")
+            print(f"    Expected: {item['expected']}")
+            print(f"    Found:    {item['found']}")
+
+    if results['fail_varying']:
+        print(f"\n[FAIL] NODE SPECIFIC VIOLATIONS ({len(results['fail_varying'])})")
+        print(f"These files did not match the expected state for {target_host}:")
+        for item in results['fail_varying']:
+            print(f"  X {item['path']}")
+            print(f"    Expected: {item['expected']}")
+            print(f"    Found:    {item['found']}")
+
+    # WARNING Sections
+    if results['unknown_file']:
+        print(f"\n[WARN] UNKNOWN FILES ({len(results['unknown_file'])})")
+        print("These files were not seen in the original research scan:")
+        for f in results['unknown_file'][:5]: print(f"  ? {f}")
+        if len(results['unknown_file']) > 5: print(f"  ... and {len(results['unknown_file'])-5} more")
+
+    if results['unverifiable_host']:
+        print(f"\n[WARN] UNVERIFIABLE FILES ({len(results['unverifiable_host'])})")
+        print("These files vary per-node, but this host is new/unknown to the research:")
+        for f in results['unverifiable_host'][:5]: print(f"  ? {f}")
+
+    # Exit Code
+    if results['fail_consistent'] or results['fail_varying']:
+        sys.exit(1)
+    sys.exit(0)
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Validate a sub-manifest against cluster research.")
+    parser.add_argument("sub_manifest", help="The partial manifest file (subdir scan)")
+    parser.add_argument("research_json", help="The cluster_ground_truth.json file")
+    
+    args = parser.parse_args()
+    
+    compare_against_research(args.sub_manifest, args.research_json)