Brocade SAN basics: installation, management, ports and FC logins

01Inventory the chassis, software and supported ports

1. Record exact model, serial, installed FOS release, firmware partitions and physical port layout. Select its hardware manual and the matching software/release notes.
2. Check licensed/entitled ports, installed optics, power/cooling status and any boot or hardware issue. A disabled entitlement or unsupported optic is not a zoning problem.
3. For directors, include chassis blades, both control processors and the supported active/standby HA state. Use the director’s console and installation procedures rather than assuming a small fixed switch.

version
firmwareshow
chassisshow
switchshow
license --show
sensorshow

02Complete physical installation and cable identity

Port numbers in this guide are examples. A director can use slot/port notation, and a port index used by a diagnostic can differ from the visible physical port label. Record both before operating on a port. The switch’s management cable does not carry this FC workload.

1. Use the exact model’s rack/rail, grounding, lifting, airflow and power requirements. Confirm cold-aisle intake and hot-aisle exhaust for the fitted airflow configuration.
2. Provide the intended independent power feeds and management/serial access. Inspect boot/component health before attaching workload links.
3. Verify the approved SFP type, wavelength, speed and fiber reach for every link. Clean/inspect connectors and route cable within the supported bend radius.
4. Label both ends with fabric, switch/port, peer device/port and cable ID. Compare A/B fabric labels before connection.
5. Attach only the planned host and storage links. Keep a planned ISL isolated until software, domain, zoning and security merge checks pass.

03Set management addressing from a retained console

1. Record current management configuration and follow the interactive prompts for the supported address family and network design.
2. For directors use the model-specific chassis and CP management procedure; standalone ipaddrset is not a complete director addressing recipe.
3. Test a new authorized SSH session and the approved management route before ending console work. Preserve a recovery path if routing or access policy is wrong.

ipaddrshow
# New isolated standalone switch only, with serial console retained:
ipaddrset
ipaddrshow

04Establish identity, time and secure administration

Use a meaningful switch name, unique authorized accounts/RBAC, SSH and the organization’s supported AAA, logging and SNMPv3 design. Retain a tested administrative recovery path. Do not copy default passwords or disable security controls to make a first login easier. Switch renaming can produce fabric notifications and belongs in the planned commissioning window.

1. Select the actual timezone and approved reachable time sources; verify synchronized time rather than merely a configured address.
2. Send a controlled management/logging test and confirm events arrive with correct identity/time.
3. Test intended account permissions and allowed management sources. Keep credentials out of exported public examples and change records.

tstimezone
tsclockserver --show

# Optional commissioning changes after policy review:
switchname brocade-a
tstimezone --interactive
tsclockserver "192.0.2.123;192.0.2.124"
tsclockserver --show

05Select the intended Virtual Fabrics context

Where Virtual Fabrics is in use, a zoning or port operation acts in a particular logical switch. setcontext selects this session’s existing FID; it does not move a physical port or create a new logical fabric. Verify the prompt/context and actual port membership before changing anything. A chassis-wide operation has wider scope than a logical-switch command.

1. Confirm whether VF is enabled, which logical switches exist and which FID owns the intended ports.
2. Choose that existing context and inspect switchshow/fabricshow. If VF is not deployed, retain the correct native/default context rather than inventing FID 20.
3. Do not enable/disable VF or move production ports as part of this inspection. Those are disruptive architecture changes covered separately.

fosconfig --show
lscfg --show
# Only when the intended existing logical switch is FID 20:
setcontext 20
switchshow
fabricshow

06Inspect and enable only the designated endpoint link

1. Verify port 0 is the actual host connection and has the intended current configuration, licensed capability and supported optic.
2. Review the default-zone policy and planned zoning before enabling the endpoint. Unzoned behavior must match the isolation design.
3. Enable only the approved port when needed, then inspect negotiated speed, online state and F_Port role. An ISL or Access Gateway uplink needs a different role/topology review.
4. If it does not come online, diagnose peer state, speed/optic compatibility, cable route and licensing before adding zones.

switchshow
portshow 0
portcfgshow 0
sfpshow 0

# Conditional: only an approved unused/correctly connected host port:
portenable 0
portshow 0
switchshow

07Verify local and fabric-wide device identities

1. Match the local host login to app01’s observed HBA port WWPN. A similar WWNN or another server’s port identity is not interchangeable.
2. Locate the intended target LIF WWPNs. NPIV can register multiple logical identities through one physical target link.
3. Use directory and fabric evidence to distinguish local attachment from a remote device reached through an ISL. Verify the current FID each time.

portloginshow 0
nsshow
nsallshow
nodefind 10:00:00:90:fa:aa:00:01
portshow 4
portshow 5

08Record health, errors and monitoring status

1. Read current hardware and MAPS observations in the context of the configured policy, time and connected workload.
2. Collect port counters twice with an interval to identify new errors. A lifetime nonzero count can predate commissioning; a quiet zero-traffic link does not prove loaded operation.
3. Ensure monitoring receives the expected switch identity and alarms. Broader MAPS actions, fencing/quarantine and policy changes require a deliberate supported design.

sensorshow
mapsdb --show
mapsdb --show history
mapsdb --show congestion
porterrshow
portstatsshow 0
errdump

09Back up the correct scope and verify the destination

Use an approved SCP/SFTP receiver, unique switch/context filename and interactive credentials. Confirm the remote file exists and is protected. For VF/chassis recovery, a current-context export alone is not sufficient: the advanced guide explains separate -vf and -all backups with chassis permissions. Do not confuse configdownload, which restores/changes configuration, with a backup command.

1. Record switch/FID, timestamp, destination, filename and intended restoration scope.
2. Retain effective/defined zoning and endpoint/port evidence alongside the configuration.
3. Confirm an authorized recovery owner can locate the backup and the matching release/model restoration procedure. A successful transfer message without a retrievable file is not acceptance.

configupload

10Finish the basics before authorizing storage access

A registered HBA is not a mapped LUN. Continue with the zoning guide, then verify storage-side initiator login, LUN masking/mapping and host multipathing. Save the completed worksheet so a later incident can compare actual identity and path changes.