PatchMortem / Banking & BFSI

A failed patch becomes an outage. An outage becomes an audit observation.

PatchMortem is built for Indian banking, insurance, and NBFC infrastructure teams — where a missed rollback is not just downtime, it is a compliance finding.

The chain that ends in an audit observation

Enterprise patch management has a gap: the tools that deploy patches do not handle failures. SCCM, Intune, and Ansible will all tell you a patch failed. None of them tell you why, and none of them will put the server back the way it was.

In regulated environments the consequences compound. A failed patch on a production cluster means an outage. An outage during a change window means a compliance finding. A compliance finding means an RBI audit observation. The chain is well understood by every VP of IT Operations in Indian banking. The tooling to break it did not exist.

PatchMortem was built to close that gap: automated classification, automated rollback, and a cryptographic audit trail that an RBI inspector can read without logging into your infrastructure.

Built for the Indian regulatory environment

PatchMortem is designed ground-up against RBI IT Framework 2023, the SEBI Cyber Security Circular, and IRDAI Information Security Guidelines — not adapted from a Western product after the fact. All customer data is stored and processed in AWS ap-south-1 (Mumbai).

RBI IT Framework 2023Audit chain designed for Section 4.2 evidence requirements
SEBI Cyber Security CircularChange control evidence and access logging aligned
IRDAI Info SecurityApplicable to insurance and NBFC deployments
PCI DSS 6.3Compliance scope drives approval tiers automatically
Data residencyAWS ap-south-1 (Mumbai) — no cross-border replication
What BFSI teams get

Evidence, not dashboards

The three things a bank's IT operations and compliance functions actually need from a patch failure.

Compliance-aware risk tiers

PatchMortem knows the difference between a dev box and a PCI-DSS-scoped production cluster. Approval tiers, rollback policies, and notification routing adjust automatically based on compliance scope — no manual classification by your L2 team.

PCI-DSS · RBI · SEBI · IRDAI

Four-eyes approval, enforced

Any action that changes the state of a production host requires change board approval, and both approvers are written into the audit chain with timestamps. Your change control evidence is generated as a by-product of the work, not reconstructed afterwards.

CHANGE_BOARD required

One-click inspector export

Compliance exports for RBI, PCI-DSS, SEBI, and IRDAI are generated from the audit chain on demand. The inspector verifies the chain end-to-end without being given access to your infrastructure.

HMAC-SHA256 · tamper-evident

Cluster topology awareness

Most patch failures in a bank happen on the systems that matter most: failover clusters running core banking, payments, and database workloads. PatchMortem detects cluster topology before it takes any action.

WindowsWindows Failover Clustering (WSFC) — ACTIVE and PASSIVE node roles identified before remediation
LinuxPacemaker cluster state validated prior to any package action
DatabaseSQL Server Always On availability groups detected; replica health confirmed
Load balancingNetwork Load Balancing membership checked before a node is taken out of service

Quorum is checked before remediation, and peer nodes are held. A rollback should never take out the cluster it was meant to protect.

Deployment in a bank

The PatchMortem agent is a code-signed binary that collects patch state and servicing logs. It initiates outbound TLS only — there is no inbound listener and no open port on your estate. Remediation is off until you enable it, per policy, per host group.

Full detail on agent behaviour, encryption, access control, and audit chain integrity is on the security page.

How a bank evaluates this

We do not run six-month evaluations. Deploy the agent on one host group — typically 25 to 50 servers with a known failure history — in read-only mode. We triage every failure in your next patch window against your own servicing logs and hand you the root causes, with the rollback we would have executed and why.

Compare that against what your team spent on the same window, and compare the audit output against what your compliance function currently submits. If the answer is not obvious, we would rather you did not buy it.

Bring us a failure you never solved

30 minutes with an engineer on your actual patch estate. No slides. No vendor pitch.