| Version 8 (modified by , 17 years ago) ( diff ) |
|---|
Inventory Grid Service
Table of Contents
Inventory Gridservice
Main purpose of Inventory OMF service is to expose the inventory information to other services as well as experiments (experiment scripts). In addition to obvious node set aliases (say based on device type and ordinal number) it should allow us to query the inventory based on other criteria like "at least 5 nodes with memory > 500M & a bluetooth radio & an Atheros 802.11 card".
- It should have an (optional) query qualifier (with interface to CMC) to return only the functional node set.
- It should have …
It is important to consider the Inventory database schema part of the experimental interface. The advantage of building a web service for Inventory is that it can provide commonly used queries that incorporate data from other OMF web services (namely the CMC, as in "available" in "give me all available intel nodes"). At the same time, it will be easier for experiment scripts to build custom SQL queries that implement complex criteria than to utilize a more abstract web service. Furthermore, arguably, SQL is RESTful in the first place.
Service Configuration File
Lives in /etc/gridservices2/available/inventory.yaml
inventory:
testbed:
default:
# Name of the MySQL Database Server
host: host-name
# Username and password to access the database server
user: the-db-username
password: the-db-password
# Name of the database to use
database: new_new_inventory
Inventory Gathering
Inventory, is the "experiment" that uses a special much larger image, which can have a relatively large number of drivers for a wide variety of devices. Inventory is used to do the slower more difficult job of generating informative scans of the pci and usb busses.
As of 10/16/2007, we run the inventory every Wednesday during maintenance. Plan is to, in addition, run Inventory whenever there is a change in hardware configuration. There exists a full inventory package that inserts "all" information into the database.
The gathering procedure uses:
- lspci
- lsusb
- dmidecode
Known Device IDs
Devices that are discovered by gathering procedure are uniquely identified by either USB ids or PCI ids. The ids for most important devices are listed in the ID table:
| Bus type | Vendor ID | Vendor | Device ID | Description |
| USB | 0x050d (1293) | Belkin | 0x0084 (132) | Bluetooth Transciever F8T003v2 |
| -"- | -"- | -"- | 0x0131 (305) | Bluetooth Transciever with trace filter (F8T001 ?) |
| -"- | 0x0a5c (2652) | Broadcom Corp. | 0x2101 (8449) | A-Link BlueUsbA2 Bluetooth |
| -"- | 0x0a12 (2578) | Cambridge Silicon Radio, Ltd | 0x0001 (1) | Bluetooth Dongle (HCI mode) |
| -"- | 0x0403(1027) | Future Technology Devices International, Ltd | 0x6001 (24577) | ZigBee Mote (FT232 USB-Serial (UART) IC) |
| -"- | 0x04b4 (1204) | Cypress Semiconductor Corp. | 0x8613 (34323) | CY7C68013 EZ-USB FX2 USB 2.0 Development Kit |
| -"- | 0xfffe (65534) | -"- | 0x0002 (2) | CY7C68013 EZ-USB FX2 USB 2.0 (USRP and NoiseGenerator) unprogrammed |
| PCI | 0x11ab (4523) | Marvell Semiconductor | 0x4320 (17184) | 88E8055 Marvell Yukon PCI E Gigabit Ethernet Controller |
| -"- | 0x8086 (32902) | Intel Corp. | 0x4223 (16931) | Intel PRO/Wireless 2915ABG Network Connectio |
| -"- | 0x168c (5772) | Atheros Communications | 0x0013 (19) | Atheros AR5212, AR5213 802.11a/b/g Wireless Adapter |
| -"- | 0x1106 (4358) | VIA Technology | 0x3122 (12578) | VT8623 CastleRock AGP 8X Controller (VGA) |
| -"- | -"- | -"- | 0x0571 (1393) | Bus Master IDE Controller |
| -"- | -"- | -"- | 0x3177 (12663) | VT8235 PCI to ISA Bridge |
| -"- | -"- | -"- | 0x3104 (12548) | VT6202 USB 2.0 Enhanced Host Controller |
| -"- | -"- | -"- | 0x3038 (12344) | USB&UHCI Controller |
| -"- | -"- | -"- | 0xb091 (45201) | VT8633 PCI-to-PCI Bridge (AGP) |
| -"- | -"- | -"- | 0x3123 (12579) | VT8623 CPU to PCI Bridge |
| USB | 0x3f0 (1008) | Hewlett-Packard | 0x0024 (36) | KU-0316 Keyboard |
We should look into turning inventory image into PXE image and avoid imaging phase completely!
Inventory Database
Inventory database lives on internal1 and consists of 6 tables:
- device_kinds
- device_tags
- inventories
- locations
- motherboards
- nodes
- peripherals
- testbeds
device_kinds table
| Field | Type | Null | Key | Default | Extra |
| id | int(11) | NO | PRI | NULL | auto_increment |
| inventory_id | int(11) | NO | NULL | ||
| oui | varchar(8) | YES | NULL | ||
| bus | varchar(16) | YES | NULL | ||
| vendor | int(11) | NO | NULL | ||
| device | int(11) | NO | NULL |
device_tags table
| Field | Type | Null | Key | Default | Extra |
| tag | varchar(64) | NO | NULL | ||
| peripheral_id | int(11) | NO | NULL |
motherboards table
| Field | Type | Null | Key | Default | Extra | Description |
| id | varchar(64) | NO | PRI | UUID of the motherboard | ||
| node_id | varchar(64) | YES | UNI | NULL | Link to 'id' in nodes table | |
| sn | varchar(16) | NO | UNI | manufacturer serial number of the motherboard | ||
| hd_sn | varchar(16) | NO | UNI | Hard drive serial number | ||
| cpu_type | varchar(X) | YES | NULL | CPU Type | ||
| cpu_speed | int(11) | YES | 0 | CPU speed in MHz | ||
| memory | int(11) | YES | 0 | Memory size in MB | ||
| hd_size | int(11) | YES | 0 | Hard disk size in bytes | ||
| updated_on | timestamp | NO | CURRENT_TIMESTAMP | |||
| updated_by | varchar(64) | NO |
(NOTE: 'node_id' is NULL when this motherboard is not installed on any node, i.e. new parts that just got in, or stored extra/spare parts)
We could also move the hard-drive info in a separate table if we allow hard-drive swapping between motherboards.
nodes table
| Field | Type | Null | Key | Default | Extra | Description |
| id | varchar(64) | NO | PRI | UUID of the node (i.e. the chassis). | ||
| chassis_sn | varchar(16) | NO | UNI | Manufacturer serial number of the node's chassis | ||
| location_id | varchar(64) | YES | UNI | NULL | Link to 'id' in 'locations' table | |
| updated_on | timestamp | NO | CURRENT_TIMESTAMP | |||
| updated_by | varchar(64) | NO |
(NOTE: 'location_id' is NULL when this chassis is not installed at any location, i.e. new parts that just got in, or stored extra/spare parts)
locations table
| Field | Type | Null | Key | Default | Extra | Description |
| id | varchar(64) | NO | PRI | UUID of the location |
| x | int(11) | NO | 0 | |||
| y | int(11) | NO | 0 | |||
| z | int(11) | NO | 0 | |||
| unit | int(11) | NO | 0 | |||
| testbed_id | varchar(64) | NO | 0 | Link to 'id' in 'testbeds' table | ||
| updated_on | timestamp | NO | CURRENT_TIMESTAMP | |||
| updated_by | varchar(64) | NO |
testbeds (resources) table
| Field | Type | Null | Key | Default | Extra | Description |
| id | varchar(64) | NO | PRI | UUID of the testbed | ||
| domain | varchar(4) | NO | UNI | |||
| control_ip | varchar(12) | NO | UNI | |||
| data_ip | varchar(12) | NO | UNI | |||
| cm_ip | varchar(12) | NO | ||||
| latitude | int(11) | NO | 0 | |||
| longitude | int(11) | NO | 0 | |||
| elevation | int(11) | NO | 0 | |||
| updated_on | timestamp | NO | CURRENT_TIMESTAMP | |||
| updated_by | varchar(64) | NO |
Notes
The design goal of this schema is to allow the double use of the Inventory database as:
- a source of information for user experiment scripts
- a 'real' hardware inventory giving operators information on which piece of hardware (chassis, motherboard) is used (or not) in which testbed/location.
The entries in the testbeds, locations, nodes tables are manually created and updated by operators, when:
- a new testbed is being deployed
- a new location is added to the testbed (e.g. physical place-holder creation on a sandbox testbed for future addition of a third node)
- a new purchased chassis (i.e. empty node box) is delivered, or mounted to a new location, or switched from a location to another one
We do not expect these events to happen very often, thus it should be ok to make the operator responsible for creating/updating the related entries. (furthermore he/she could also use some scripts to do this job…)
The entries in the motherboards table are also manually created upon delivery of a new purchased motherboard. The only field that needs to be manually filled by the operator is the node_id, which will happen when the operator installs a new motherboard inside a node/chassis. All the other fields are automatically populated by the Inventory process (i.e. the scripts in the inventory package).
The interfaces and devices tables are created and updated as in the previous schema.
These schemas have a tendency to change, so assume the actual database is more authoritative than this document.
