fernando/clusterlogs
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

memory(MEM-20260824-C5E50A): Reparar mon.mcps-hpc-01 en crash-loop y 2 PGs inconsistentes nuevas en Ceph

Browse Source
This commit is contained in:
fernando
2026-08-24 14:49:00 -06:00
parent 2c13427788
commit 3bf89b9888
2 changed files with 136 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
data/generated/site_state.json
+58 -3
View File
@@ -1,14 +1,40 @@
{
"version": 2,
"generated_at": "2026-08-24T18:25:00Z",
"source_sha256": "73d66ea39e4def01d7ccf1993b29bbfd832e893c7926a96dc842bc02d70f6044",
"generated_at": "2026-08-24T20:49:00Z",
"source_sha256": "c7dbbf4afcb1ad2c038fd2a96e982297753b12e091129fe9e0fc0d4e12cc07d8",
"stats": {
"active_admins": 1,
"pending_reviews": 4,
"pending_reviews": 5,
"open_tasks": 1,
"active_projects": 1
},
"review_queue": [
{
"entry_id": "MEM-20260824-C5E50A",
"title": "Reparar mon.mcps-hpc-01 en crash-loop y 2 PGs inconsistentes nuevas en Ceph",
"description": "El mon del login node (mcps-hpc-01) llevaba ~16h en crash-loop por corrupción de su store RocksDB local; se sacó del monmap y cephadm lo redesplegó con store fresco. En paralelo aparecieron 2 PGs nuevas inconsistentes (HEALTH_ERR), diagnosticadas y reparadas sin ambigüedad.",
"date": "2026-08-24T20:49:00Z",
"author_id": "gen",
"risk": "high",
"tags": [
"almacenamiento",
"ceph",
"mon",
"salud-cluster"
],
"systems": [
"ceph",
"mcps-hpc-01"
],
"permalink": "/memorias/mem-20260824-c5e50a-reparar-mon-mcps-hpc-01-en-crash-loop-y-2-pgs-inconsistentes-nuevas-en-ceph/",
"content_sha256": "e3ca2349c7ebbfc769db94409776e82677c818d10de0124881a2b1d1f4f96233",
"reviewed_by": [],
"missing_reviewers": [
"gen"
],
"required_count": 1,
"reviewed_count": 0
},
{
"entry_id": "MEM-20260820-4D10C2",
"title": "Reparar 2 PGs inconsistentes nuevas en Ceph (HEALTH_ERR, mcps-hpc-01)",
@@ -112,6 +138,35 @@
}
],
"recent_memories": [
{
"entry_id": "MEM-20260824-C5E50A",
"title": "Reparar mon.mcps-hpc-01 en crash-loop y 2 PGs inconsistentes nuevas en Ceph",
"description": "El mon del login node (mcps-hpc-01) llevaba ~16h en crash-loop por corrupción de su store RocksDB local; se sacó del monmap y cephadm lo redesplegó con store fresco. En paralelo aparecieron 2 PGs nuevas inconsistentes (HEALTH_ERR), diagnosticadas y reparadas sin ambigüedad.",
"date": "2026-08-24T20:49:00Z",
"author_id": "gen",
"author_fingerprint": "SHA256:JhN1Tt4jYqWoJIznDJ6QkMAiOR6HFHr3xLz9aBJqmPg",
"risk": "high",
"review_required": true,
"review_policy": "all-active",
"reviewer_snapshot": true,
"reviewers": [
"gen"
],
"tags": [
"almacenamiento",
"ceph",
"mon",
"salud-cluster"
],
"systems": [
"ceph",
"mcps-hpc-01"
],
"slug": "mem-20260824-c5e50a-reparar-mon-mcps-hpc-01-en-crash-loop-y-2-pgs-inconsistentes-nuevas-en-ceph",
"permalink": "/memorias/mem-20260824-c5e50a-reparar-mon-mcps-hpc-01-en-crash-loop-y-2-pgs-inconsistentes-nuevas-en-ceph/",
"source_path": "content/memorias/2026-08-24-mem-20260824-c5e50a-reparar-mon-mcps-hpc-01-en-crash-loop-y-2-pgs-inconsistentes-nuevas-en-ceph.md",
"content_sha256": "e3ca2349c7ebbfc769db94409776e82677c818d10de0124881a2b1d1f4f96233"
},
{
"entry_id": "MEM-20260820-4D10C2",
"title": "Reparar 2 PGs inconsistentes nuevas en Ceph (HEALTH_ERR, mcps-hpc-01)",