For the next outage on Cobalt, which recovery procedure should Ivo use, and which outage gate led to the delivery-mode change?

--- directory: viking://resources/formal/m_d7a9d1f22107f233/ source: kind: local uri: /app/.openviking/native/temp/upload/upload_0932bc3f237e4b04b2fc5d6e87fcfa27.txt generated_by: component: SemanticProcessor trigger: resource_ingest freshness: total_entries: 1 sampled_entries: 1 unsampled_entries: 0 pending_child_changes: 0 --- # m_d7a9d1f22107f233 This directory contains a brief meeting note about lost updates associated with outages and a tested recovery option. It records Mara’s report that BATCH-0-741 lost updates during GATE-0-519 outages, the team’s test of REPLAY-0-632, and the meeting’s decision not to approve a change in delivery mode. The note is relevant to tracking BATCH-0-741, GATE-0-519, REPLAY-0-632, lost updates, outages, and delivery mode. No specific audience or required prior knowledge is stated. ## Directory Coverage There is **1 direct entry**, and **all entries are represented** in the summaries provided. ## Quick Navigation What do you want to learn? - → Read the meeting report, outage context, replay test, and delivery-mode decision: [m_d7a9d1f22107f233.md](viking://resources/formal/m_d7a9d1f22107f233/m_d7a9d1f22107f233.md) ## Detailed Description ### [m_d7a9d1f22107f233.md](viking://resources/formal/m_d7a9d1f22107f233/m_d7a9d1f22107f233.md) This brief meeting note records an update-loss issue, a recovery test, and the decision about whether to change delivery mode. It is dated 2026-08-08 and identifies the reported issue as lost updates in BATCH-0-741 during GATE-0-519 outages. The note attributes the report to Mara. It states that the team tested REPLAY-0-632 and that this test restored the missing updates. It does not specify how the test was performed, provide further technical details about the outage or recovery process, or describe the underlying systems beyond the named identifiers. These limits mean the note supports tracking the reported incident and test outcome, but does not provide a detailed technical account of either. The meeting decision is also explicit: no change of delivery mode was approved. The note therefore records a successful recovery test alongside a decision not to approve a delivery-mode change at that meeting. It does not state why the change was not approved or whether a later decision was made. For retrieval, the main terms associated with this entry are BATCH-0-741, GATE-0-519, REPLAY-0-632, lost updates, outages, and delivery mode. The note may be useful to readers following these update and outage issues; the source does not identify a particular audience or specify prior knowledge.

2026-08-08. Ivo reported that BATCH-1-741 lost updates during GATE-1-519 outages. The team tested REPLAY-1-632, which restored the missing updates. No change of delivery mode was approved at this meeting.

2026-08-01. Ivo started project Cobalt. They chose BATCH-1-741 for batch delivery. The initial retry policy was RETRY-1-108.

--- directory: viking://resources/formal/m_f7ff6441bdd98e53/ source: kind: local uri: /app/.openviking/native/temp/upload/upload_07358da9e4124ed690805a9b15a09280.txt generated_by: component: SemanticProcessor trigger: resource_ingest freshness: total_entries: 1 sampled_entries: 1 unsampled_entries: 0 pending_child_changes: 0 --- # m_f7ff6441bdd98e53 This directory contains an operational update dated 2026-08-08 about lost updates associated with outages and a tested recovery approach. The note reports that BATCH-1-741 lost updates during GATE-1-519 outages and that testing REPLAY-1-632 restored the missing updates. It also records an important decision outcome: no change of delivery mode was approved at the meeting. The directory is useful for readers tracking the reported update loss, the replay-based recovery, and the delivery-mode decision. The identifiers are recorded without further technical explanation. ## Directory Coverage There is **1 direct entry**, and **all entries are represented** in the overview. ## Quick Navigation What do you want to learn? - What update loss was reported and what recovery was tested? → [m_f7ff6441bdd98e53.md](viking://resources/formal/m_f7ff6441bdd98e53/m_f7ff6441bdd98e53.md) - What was decided about changing the delivery mode? → [m_f7ff6441bdd98e53.md](viking://resources/formal/m_f7ff6441bdd98e53/m_f7ff6441bdd98e53.md) ## Detailed Description ### [m_f7ff6441bdd98e53.md](viking://resources/formal/m_f7ff6441bdd98e53/m_f7ff6441bdd98e53.md) This brief operational meeting or incident note records an update dated 2026-08-08. It reports that Ivo identified lost updates in BATCH-1-741 during GATE-1-519 outages. The note then describes a tested recovery approach: REPLAY-1-632 restored the missing updates. It does not provide further technical details about the identifiers, the outage conditions, or how replay was performed. The note also distinguishes the recovery result from a delivery-mode decision. Although replay restored the missing updates, the meeting did not approve a change of delivery mode. Thus, the recorded outcome is that the tested replay approach recovered the missing updates, while the delivery mode remained without an approved change in this meeting note. The document does not state whether a change was considered elsewhere or whether any later decision followed. This file is suited to readers who need a concise record of the reported loss and recovery, or who are checking the stated meeting outcome. Its searchable terms include BATCH-1-741, GATE-1-519, REPLAY-1-632, outages, lost updates, replay, and delivery mode. These terms reflect the subjects named in the note; the note itself does not expand on them with additional system descriptions, implementation guidance, or explanations.

# Summary On 2026-08-22, Ivo reviewed Cobalt. # 2026-08-22 (Saturday) ChatLog: **user**: 2026-08-22. Ivo reviewed Cobalt. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-1-907, superseding RETRY-1-108. No budget owner has been appointed.

# Summary On 2026-08-22, the user specified that the recovery procedure validated on 2026-08-08 be used for the next outage. # 2026-08-22 (Saturday) ChatLog: **user**: 2026-08-22. Mara reviewed Larch. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-0-907, superseding RETRY-0-108. No budget owner has been appointed.

# Summary GATE-1-519 outages were recorded on 2026-08-08. The user asked to retain the successful recovery procedure from that incident. # 2026-08-15 (Saturday) ChatLog: **user**: ...The cause was the GATE-1-519 outages recorded on August 8; keep the successful recovery procedure from that incident.

# Summary On 2026-08-22, retry policy RETRY-1-907 superseded RETRY-1-108. # 2026-08-22 (Saturday) ChatLog: **user**: 2026-08-22. Ivo reviewed Cobalt. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-1-907, superseding RETRY-1-108. No budget owner has been appointed.

--- directory: viking://user/formal/memories/events/2026/08/15/ generated_by: component: MemoryUpdater trigger: memory_update freshness: total_entries: 3 sampled_entries: 3 unsampled_entries: 0 pending_child_changes: 0 --- # Events Overview - [On 2026-08-15, Ivo approved switching Cobalt from BATCH-1-741 to STREAM-1-286, superseding the August 1 delivery decision.](./cobalt_stream_switch.md) - [GATE-1-519 outages were recorded on 2026-08-08. The user asked to retain the successful recovery procedure from that incident.](./gate_outage_recovery.md) - [On 2026-08-15, Mara approved switching Larch from BATCH-0-741 to STREAM-0-286, superseding the 2026-08-01 delivery decision.](./larch_source_switch.md)

# Summary The recovery procedure was validated on 2026-08-08. # 2026-08-22 (Saturday) ChatLog: **user**: 2026-08-22. Mara reviewed Larch. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-0-907, superseding RETRY-0-108. No budget owner has been appointed.

2026-08-08. Ivo reported that BATCH-1-741 lost updates during GATE-1-519 outages. The team tested REPLAY-1-632, which restored the missing updates. No change of delivery mode was approved at this meeting.

2026-08-08. Mara reported that BATCH-0-741 lost updates during GATE-0-519 outages. The team tested REPLAY-0-632, which restored the missing updates. No change of delivery mode was approved at this meeting.

2026-08-01. Ivo started project Cobalt. They chose BATCH-1-741 for batch delivery. The initial retry policy was RETRY-1-108.

2026-08-01. Mara started project Larch. They chose BATCH-0-741 for batch delivery. The initial retry policy was RETRY-0-108.

Kestrel-0 signed return policy. Ordinary goods have a 7-day return window. The complete exception list consists of two clauses: custom-made goods use exception CUSTOM-0-7c2; quality defects use exception DEFECT-0-9b4 and are exempt from the 7-day limit. Both exception identifiers must be retained when listing all exceptions. The old informal draft listed only DRAFT-0-2d1 and has no authority.

User Input: 2026-08-22. Ivo reviewed Cobalt. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-1-907, superseding RETRY-1-108. No budget owner has been appointed. Agent Response: Recorded.

User Input: 2026-08-15. Ivo approved switching Cobalt from BATCH-1-741 to STREAM-1-286. This supersedes the August 1 delivery decision. The cause was the GATE-1-519 outages recorded on August 8; keep the successful recovery procedure from that incident. Agent Response: Recorded.

User Input: 2026-08-22. Mara reviewed Larch. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-0-907, superseding RETRY-0-108. No budget owner has been appointed. Agent Response: Recorded.

The delivery choice made on August 15, 2026, is to remain in force.

User Input: 2026-08-15. Mara approved switching Larch from BATCH-0-741 to STREAM-0-286. This supersedes the August 1 delivery decision. The cause was the GATE-0-519 outages recorded on August 8; keep the successful recovery procedure from that incident. Agent Response: Recorded.

User Input: 2026-08-08. Ivo reported that BATCH-1-741 lost updates during GATE-1-519 outages. The team tested REPLAY-1-632, which restored the missing updates. No change of delivery mode was approved at this meeting. Agent Response: Recorded.

2026-08-08. Ivo reported that BATCH-1-741 lost updates during GATE-1-519 outages. The team tested REPLAY-1-632, which restored the missing updates. No change of delivery mode was approved at this meeting.

2026-08-08. Mara reported that BATCH-0-741 lost updates during GATE-0-519 outages. The team tested REPLAY-0-632, which restored the missing updates. No change of delivery mode was approved at this meeting.

2026-08-01. Ivo started project Cobalt. They chose BATCH-1-741 for batch delivery. The initial retry policy was RETRY-1-108.

2026-08-01. Mara started project Larch. They chose BATCH-0-741 for batch delivery. The initial retry policy was RETRY-0-108.

Kestrel-0 signed return policy. Ordinary goods have a 7-day return window. The complete exception list consists of two clauses: custom-made goods use exception CUSTOM-0-7c2 quality defects use exception DEFECT-0-9b4 and are exempt from the 7-day limit. Both exception identifiers must be retained when listing all exceptions. The old informal draft listed only DRAFT-0-2d1 and has no authority.

On 2026-10-16, the user said to keep the delivery choice made on August 15, 2026.

User Input: 2026-08-22. Ivo reviewed Cobalt. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-1-907, superseding RETRY-1-108. No budget owner has been appointed. Agent Response: Recorded.

User Input: 2026-08-15. Ivo approved switching Cobalt from BATCH-1-741 to STREAM-1-286. This supersedes the August 1 delivery decision. The cause was the GATE-1-519 outages recorded on August 8; keep the successful recovery procedure from that incident. Agent Response: Recorded.

The user instructed that the August 15 delivery choice be kept.

User Input: 2026-08-22. Mara reviewed Larch. Keep the August 15 delivery choice. For the next outage, use the recovery procedure validated on August 8. The retry policy is now RETRY-0-907, superseding RETRY-0-108. No budget owner has been appointed. Agent Response: Recorded.

User Input: 2026-08-15. Mara approved switching Larch from BATCH-0-741 to STREAM-0-286. This supersedes the August 1 delivery decision. The cause was the GATE-0-519 outages recorded on August 8; keep the successful recovery procedure from that incident. Agent Response: Recorded.