Manual sets, SP Switch Router manuals | Lucent Technologies 9076

About this Guide

Manual sets

Manual sets

This section provides a list of relevant GRF manuals. A second list includes IBM system manuals that contain information speciﬁc to the SP supercomputer.

SP Switch Router manuals

The SP Switch Router Adapter media card is described only in the SP Switch Router Adapter Guide. The SP Switch Router and other media cards are described in the GRF manual set.

The GRF 1.4 Update 2 documentation set consists of the following manuals:

•GRF 400/1600 Getting Started - 1.4 Update 2

•GRF Conﬁguration and Management - 1.4 Update 2

•GRF Reference Guide - 1.4 Update 2

•GRF GateD Manual - 1.4 Update 2

•SP Switch Router Adapter Guide - 1.4 Update 2

Router manuals and software upgrade release notes for the 9077 are available in PDF format from this Web site:

http://www.rs6000.ibm.com/resource/aix_resource/sp_books/sra /index.html

To check which software release your SP Switch Router is running, use the getver command:

super> getver

Current Revision: 1.4.20.ibm Version: default super>

IBM SP system manuals

IBM information speciﬁc to the SP Switch Router Adapter card appears in:

•IBM RS/6000 Scalable POWERparallel Systems:

Planning, Volume 1, Hardware and Physical Environment, GA22-7280

•IBM RS/6000 Scalable POWERparallel Systems:

Planning, Volume 2, Control Workstation and Software Environment, GA22-7281

•IBM Parallel System Support Programs for AIX: Installation and Migration Guide, GA22-7347

•IBM Parallel System Support Programs for AIX: Administration Guide, SA22-7348

•IBM Parallel System Support Programs for AIX: Diagnosis Guide, GA22-7350

•IBM Parallel System Support Programs for AIX: Command and Technical Reference, Volume 1, SA22-7351

•IBM Parallel System Support Programs for AIX: Command and Technical Reference, Volume 2, SA22-7351

•IBM Parallel System Support Programs for AIX: Messages Reference, GA22-7352

xiv	SP Switch Router Adapter Guide - 1.4 Update 2

Image 14

Lucent Technologies 9076, 9077 16S, 9077 04S Manual sets, SP Switch Router manuals, IBM SP system manuals

Contents

SP Switch Router Adapter Guide Lucent Technologies Contents Chapter Conﬁguring the SP Switch Router Adapter Chapter Appendix a Part Numbers Appendix C Viii SP Switch Router Adapter Guide 1.4 Update Figures Figures Tables Tables About 1.4 Update How to use this Guide IBM SP system manuals SP Switch Router manualsManual sets Documentation conventions Convention Meaning IP routing publications Introduction to the SP Switch Router Adapter card What is the RS/6000 SP Switch Router ? SP Switch Router Cables included in your system SP Switch Router systems for IBM sitesSP Switch cable Pcmcia 520MB disk Redundant AC power suppliesRedundant supply safety Ethernet cable Upgrading system memory RAM Overview of the SP Switch Router Adapter card Face plate diagram Inserting a media card into the SP Switch Router Media card components Card insertion procedure ESD requirements OFF Error SP Switch Router Adapter card LEDsLED activity during boot OFF TX HB LED activity during normal operations STATE1 Bottom two LEDs amber go on during SP Switch Router Adapter card speciﬁcations Element Value Assigning ﬁlters Tcpdump Snmp on the SP Switch Router Adapter card SP Switch Router Adapter dependent node MIB support SP Switch Router Adapter media card states Snmp Snmp conﬁguration overview Snmp activity during media card start up Conﬁguring the SP Switch Router Adapter Introduction to installation and conﬁguration Pre-installation assumptions Location of relevant information Order of information Installing an SP Switch Router Adapter card Installation overview Installing the Pcmcia spinning disk Managing Pcmcia slots Installation steps Panic dumps sent to external ﬂash device ﬁle entries should now look like the following Save all changes and reboot Do not damage the connector ends Attaching SP Switch Router cablesSP switch cable Ethernet cable Procedure for attaching cables to card and SP Switch Keep the plastic cap on Determining the switch connection for a dependent node Conﬁguration required on the SP system Procedure Sources of conﬁguration information Multiple frames for multiple system connections SP system a Overview of steps Step-by-step media card conﬁgurationConﬁguration ﬁles and their uses SP Switch Router requires a speciﬁc conﬁguration ﬁle Check Snmp in the SP Switch Router system Members Unspecified Subagents Put Snmp changes into effect ALL GET SET Trap Method 2 Optional, edit /etc/grifconﬁg.conf Assign IP addressesMethod 1 Recommended, use SP Snmp Manager Netmask Interface nameInternet address Broadcast / destination address Argument ﬁeld Default MTU valuesPutting grifconﬁg.conf additions into effect MTU discovery facility Specify Icmp throttling Change proﬁle settingsSpecify card-level parameters Card proﬁle Specify different dump settings Specify different executables Configuring the SP Switch Router Adapter Look at the SP Switch Router Adapter card settings Change executables for all dev1 cards Load proﬁle Change dump defaults for all dev1 cards Dump proﬁle Dump vectors read-only Configuring the SP Switch Router Adapter Method 2 Optional, edit /etc/grdev1.conf How to run the commandRun dev1conﬁg to create grdev1.conf Contents of /etc/grdev1.conf Parameter deﬁnitions Extension Node Identiﬁer ibmSPDepNodeName read-only Conﬁguration State ibmSPDepConfigState read-only Reset card to install ﬁles Saving conﬁguration ﬁles Verify SP Switch Router Adapter card from router Verify media card operation using ping Media card states Verify switch node connectivity using pingCheck media card status using grcard Reset media card using grreset # grreset all -h # grreset slot -h Checking connectivity to the SP system Procedure Configuring the SP Switch Router Adapter Monitoring and Management Tools ﬂashcmd SP Switch Router command overviewCsconﬁg Grcard GetverGrarp Grﬁns Grrmb GrresetGrms Grroute Mountf GrstatGrwrite Setver Route SP Switch Router Unix toolsPing Traceroute Grroute.conf ﬁle Using the netstat command Netstat -rn Netstat -rs Netstat -in Netstat -an Netstat -s Layer 3 statistics Obtaining layer 2 and 3 statistics grstatOptions List of IP stats Multicast packets attempted to route Layer 2 statistics List of layer 2 stats SP Switch Router Adapter card maint commands Preparing to use maint commands Find transmit tx binary version maint Sample maint commandsFind hardware and software version numbers maint Display conﬁguration and status maint Maint 5 display switch statistics Maint 4 display media statistics Filtering commands maint 50-58 Maint 6 display combus statistics Conﬁgure UDP packet discards maint 89 List where ﬁlters are assigned maintList the ﬁlters per media card maint Display ARP table maint 189 Flush the ARP cache maint 189 Display switch route table maint 189 Switch route not found Where to ﬁnd the user guide Checking for hardware problems grdiagWhat is tested Stopping or halting grdiag When a media card does not boot after grdiag Switch receive error can indicate hardware problem Media card dumps SP Switch Router dumpsSystem dumps Use grdinfo to collect logs Data collection utility grdinfo Grdinfo -card=slot all SP Switch Router example Grdinfo -all ============================================================ Monitoring and Management Tools Monitoring and Management Tools SP Switch Router logs Accessing a log ﬁle Sample gr.console log For example Sample gr.boot log Sample entries in the gr.boot log Sample messages log Sample entries in the messages log Burning in media card ﬂash memory Part Numbers Parts list model 04S GRF-AC-CB16 Parts list model 16SGRF-AC-SWB16 GRF-AC-AC16 Publication numbers IBM manuals IBM publication number Manual title Log Messages Alphabetical list of messages Log Messages Message descriptions Access Fifo Sync Error from RC, int1=%dAccess Fifo Sync Error from TC, int1=%d ACK Words 0x%08x, 0x%08x, 0x%08x, 0x%08x, 0x%08x, 0x%08x Board Conﬁguration timed-out, retrying ARP added IP= %s, SW-node=%d, state=%dBad Hdr/Svc received, Ret code=%d CPU ready msg received from TX-CPU Conﬁguration ParametersConﬁguring transmit side Discarding msg, unknown msgtype 0x%x Descriptor Sync Error from RC, int1=%dDescriptor Sync Error from TC, int1=%d ERR! Duplicated buffer %d Hot interrupt detected, ier1=0x%8x, ier2=0x8x Expect Nodeinit but received Stat/Err RequestExpired IP buffer received %d Initializing Main Task Loopback routes found for IP adr 0x%x, sw-node IOSTB3RX SET TOD service message receivedInitializing RX Subsystem data structure Lost of STI clock, Tbic Status = 0x%x RX Clock is valid NetStar GigaRouter %s RX Interface InitializingNetStar GigaRouter %s TX Interface Initializing RX-CPU Loaded msg received from TX-CPU RX-TBIC permanent errors detected IER1=0x%x, IER2=0x%x RX send Sendtod to the switch. or RX Reading TBICs TODRX-TBIC outage errors detected IER1=0x%x, IER2=0x%x RX got Tbic Inited fron TX Sending if Reset message to TX-CPU RX-TBIC transientt errors detected IER1=0x%x, IER2=0x%xSend TOD service message received Sending Grid Init Sending if Init for if 0x%x State machine changes from xxx to yyy Sending mib-2 trap, type = %d, state = %dSending params to TX Switch route entry not found, node = %d Status/Err service message receivedSwitch received with errors. Descriptor 0x%8x Svc Msg rxed, svccmd=0x%x %s, nodecmd 0x%x %s TB4 segment received in wrong state Switch Route table loaded, %d entriesTB4 segment received in error. TB4-HDR word%d 0x%x Tbic Init msg sent to TX-CPU TBSI-RX TBICs TOD 0x%x, 0x%x Tbus Parity Error from RC, int1=%dTbus Parity Error from TC, int1=%d TBSI-TX Unexpected SVC threshold interrupt Timeout waiting for Cfgdn bits clear, stat0=%x Timeout waiting for FPGAINITCOMPLETE, stat0=%x Timed out resolving ARP. If %d IP %d.%d.%d.%d RX Setting the RC in operational mode, stat=0x%xTX-CPU Access Fifo busy, discarding IP packet TX-CPU Conﬁg Params msg received from RX-CPU TX-PROR startnode %d, endnode %d, # entries %d TX-CPU Tbic Init msg received from RX-CPUTX Port is connected, stat=0x%x TX pseudo STI clock is valid TX-TBIC transient errors detected IER1=0x%x, IER2=0x%x TX-TBIC outage errors detected IER1=0x%x, IER2=0x%xTX-TBIC permanent errors detected IER1=0x%x, IER2=0x%x Un-expected descriptor from RC, desctype=%d Log Messages Network Conﬁguration Examples Conﬁguration requirements Example 2 Multiple cards, single partition ARP table Outgoing trafﬁc coming from SP processor nodes Conﬁguration tasksIncoming trafﬁc going to SP processor nodes Recovery procedure if an SP Switch Router Adapter card fails Example 3 Multiple cards, multiple SP partitions Obtaining new machine code SP Switch Router as an IBM productSupport for code installation IBM License Agreement for Machine Code Index ARP Index GRF IBM Index MTU Snmp Index Index