Tru64 Command CheatSheet
TRU64 Command CheatSheet
1. System Version
sizer -v
2. Unix Name and hostname
uname -n; uname -a
3. Number of CPU(s)
sizer -p
4. Memory state
vmstat -P
5. Swap state
swapon -s
6. Filesystems
df -k df -k -t advfs
7. CPU numbers, speed
# * Number of CPUs with psrinfo
psrinfo -n
# * Number of CPUs with sizer
sizer -p
# * Number of CPUs with dbx (debugger)
echo "p ncpus" | dbx -k /vmunix /dev/mem 2>/dev/null | tail -1
echo "p rpb->rpb_numprocs" | dbx -k /vmunix /dev/mem 2>/dev/null | tail -1
# * CPUs type with psrinfo (speed, model, cache)
psrinfo -v
# * CPUs type with hwmgr (only for version 5)
hwmgr get attr -cat cpu -a sub_category
hwmgr get attr -cat cpu -a sub_category -a speed
# * CPUs speed (MHz) with dbx (see also "rpb" and "rpb -r" with kdbx)
SPD=$(echo "pd rpb->rpb_counter/1000000;quit" | dbx -k /vmunix /dev/mem 2>/dev/null | tail -1)
echo "CPU SPEED = ${SPD}"
8. System Type
# * System Type (only for version 5)
SYS_TYPE=$(hwmgr get attr -cat platform -a name | grep -v : |sed 's/ name = //')
echo "${SYS_TYPE}"
# * System Type with kdbx (kernel debugger)
KERN=$(sizer -b)
SYS_TYPE=$(echo sum | kdbx -k /${KERN} /dev/mem | grep cpu
SYS_TYPE=$(echo "${SYS_TYPE}" | cut -f2 -d: | cut -f1 | cut -b2-)
echo "${SYS_TYPE}"
# * System type with dbx
KERN=$(sizer -b)
STR="pd (char *)rpb + rpb->rpb_dsr_off + ((struct rpb_dsr *)\
((char *)rpb + rpb->rpb_dsr_off))->rpb_sysname_off + sizeof(long)"
SYS_TYPE=$(echo "${STR}" | dbx -k /${KERN} /dev/mem 2>/dev/null | tail -1 | cut -f2 -d'"')
echo "${SYS_TYPE}"
# * System Type with snmp_request (SNMP query)
COM=$(grep ^community /etc/snmpd.conf | tail -1 | awk '{print $2}')
snmp_request 127.0.0.1 ${COM:-public} get 1.3.6.1.2.1.1.1.0
9. Patches and installed software: “setld” e “dupatch”
# * Patch Kit summary
dupatch -track -type kit -nolog#
# * Manual Patches
dupatch -track -type patch -nolog | grep "Patch C"
# * Delete a list of patches (divided by product: UNIX and Cluster)
dupatch -delete -product Tru64_UNIX_V5.1B -patch "C 00362.02" "C 00926.00" "C 00927.00" \
"C 00469.00" "C 00564.02" "C 00453.00" "C 00407.00" "C 00409.01" \
-product TruCluster_V5.1B -patch "C 00127.00" -name "HP" -note "delete manual patches PK3 from the cluster"
# * Patches/Package dependences
cd /usr/.smdb.; /usr/lbin/depord -I package_name
# * Truly "active" Patches
grep -v none /usr/.smdb./OSFPAT*.inv | awk '{print $1}' | sort -u | cut -f1 -d:
# * Software/patches installed
ls /usr/cluster/members/{memb}/.smdb.
# or (like manpages says)
ls /usr/.smdb./*.lk
10. Verify that permissions owners of “files” installed on the system are:
# correct and with a valid checksum (see also: manpage fverify) ls -l ls /usr/.smdb./*.lk | sed 's/lk/inv/' | while read F; do /usr/lbin/fverify -np <$F; done 2>&1
11. Last PatckKit (PK) Installed
# (dupath -track -type kit -nolog -> is the same, but slower)
PATCH=$(grep AS0 /var/adm/patch/log/event.log | cut -f2 -d'>' | cut -b2- | tail -1)
expr ${PATCH##*AS%%-*}
# * Baseline (in version 4.0g there isn't)
grep AS0 /var/adm/patch/log/event.log | cut -f2 -d'>' | cut -b2- | tail -1 |\
awk -F"AS" '{print $1}' | rev | cut -b -3 | rev
# * Manual Patch (by hand)
grep ^KITNAME /var/adm/patch/log/event.log | grep -v AS | cut -f2 -d'>' |\
cut -b2- | sort -u
# * Manual Patch only for the last PatchKit
BN=$(grep AS0 /var/adm/patch/log/event.log | cut -f2 -d'>' | cut -b2- | tail -1 |\
awk -F"AS" '{print $1}' | rev | cut -b -3 | rev)
grep ^KITNAME /var/adm/patch/log/event.log | grep -v AS | cut -f2 -d'>' | cut -b2- |\
sort -u | grep $BN | tr '\n' ',' | sed s/,$//
12. Users
# NOTES: see the manpage of: "usermod" # some commands are only valid for system with C2 # (Security ENHANCHE - "rcmgr get SECURITY") # # Unlock a User usermod -x administrative_lock_applied=0 username # Lock a User usermod -x administrative_lock_applied=1 username # * Enable a user when he is in DISABLE state # In "10" days the user must login usermod -x grace_limit=10 username # * Enforce user password change usermod -x passwd_must_change=1 username # * Get/Query user status passwd -q username passwd -qa # -> all users (LK=Lock, PS=OK, NP=no password) # * Change HASH field to "*" like shadow password sed 's/:[^:]*:/:*:/' /etc/passwd
13. C2 (Security ENHANCHED)
# * Is the system in C2 (Security ENHANCHED) rcmgr get SECURITY # * Delete all users from SIA db (C2 db) that are not present in "/etc/passwd" /tcb/bin/authck -p convuser -d # * Some Checks grpck pwck siacfg ckfsec # * Editing of the default values of SIA DB # (edauth may be located at: "/tcb/bin/edauth") VISUAL=/usr/bin/vi edauth -dd default VISUAL=/usr/bin/vi edauth username # * Get/Show user "C2" attributes edauth -g username edauth -g # show all users # * Get/Show default values edauth -g -dd default # NOTES: see prppasswd, default manpages
14. Filesystem/Files
# * Show files in "size" order
cd /; find . -mount -type f -exec du -k {} \; | tee /tmp/df1
sort -rn /tmp/df1 | tee /tmp/df2; mv /tmp/df2 /tmp/df1
more /tmp/df1
15. Advance Filesystem (AdvFS)
# Under the path: "/sbin/advfs/"; there are AdvFS utility
#
# * Repopulate "/etc/fdmns".
# May be useful when you do a storage migration.
# Ex. - All disks of storage 'A' will be migrated to a storage 'B'.
# (see manpage of advscan)
/sbin/advfs/advscan -ga
/sbin/advfs/advscan -g
/sbin/advfs/advscan dsk101
/sbin/advfs/advscan -g dsk101
# For the real fix be carefull (pay ATTENTION!)
/sbin/advfs/advscan -r
/sbin/advfs/advscan -f domain_name
# * Files in ".tags" directory (.tags files)
#
# tag 1 is th fragment file.
# (the biggers is, worst in performance)
cd /mount_point/.tags
ls -li 1 # Check the size of frag file
showfile -x 1 | grep -i extentcnt # Check number of extent
# * Disable frag file
# [ With this operation files will be create with a size that is a
# multiply of 8192 (8KB). A file of 100 byte will reserve 8KB
# The file "/mount_point/.tags/1";
# handle all "pieces" of small files, this to use at best the
# Filesystem space ]
chfsets -o nofrag fset_name # Disable frag file
# * BMT file (Bitfile Metadata Table -> Metadata)
# The BMT is M-10
# AdvFS volumes begin from M-6 and increment by a factor of six
# (M-6, M-12, M-18, ...)
cd /mount_point/.tags
showfdmn domin_name # How many disks
ls -ls M-6 M-12 M-18 M-24 M-30 M-36 # Check the size (ex. 6 dsks)
showfile -x M-6 M-12 M-18 M-24 M-30 M-36 | grep -i extentcnt # Check extent
showfile -x M-10 # Check extent of BMT
# If BMT (M-10) is too much "fragmented" you may create a new FS with
# different values.
# -p (the initial size of BMT)
# -x (size of extent for BMT)
# * AdvFS statistics and Buffer cache
advfsstat -b # Buffer cache
advfsstat -l 0 # Lock stats basic
1 # detailed
2 # very detailed
advfsstat -v 2 #
advfsstat -v 3 #
16. How to create a particular AdvFS Filesystem
mkfdmn -p 256000 -x 4096 dsk69c particular_domain # It will take a lot, because it will create 256000 pages of 8192 (or 8184) bytes # 2 GByte of preallocate bitfile metadata table. # # * How to Increment AdvFS performance vfast -o defragment=enable,balance=enable,topIObalance=enable my_domain vfast status my_domain
17. Defragment of AdvFS
# Show defragment status of domain/Filesystem defragment -nv my_domain # * Defragment for 30 minutes # with -T "exactly" of 30 minutes (it is not true at all ...) defragment -t 30 -v my_domain # * Defragment of ALL domains (in parallel) for 30 minutes defragcron -T 30 -p
18. Kernel parameters (sysconfig)
# * List of dynamic subsytems loaded during boot /sbin/init.d/autosysconfig list # * List of subsytems (dynamic and static) loaded and configured sysconfig -s # * List of subsytem sorted by: dynamic and static sysconfig -m # * Load a dynamic kernel module onto the system (ex envmon) sysconfig -c subsys # * Unconfigur a dynamic kernel module sysconfig -u sybsys # * Set automatic load of a dynamic kernel module at boot # ATTENTION (use sysconfig -c subsys) # See: SUBSYSTEM_LIST in "/sbin/init.d/autosysconfig" /sbin/init.d/autosysconfig add subsys # * Delete automatic load of a kernel module at boot /sbin/init.d/autosysconfig delete subsys # * Get/Query kernel parameters sysconfig -q subsys [parameter1] [parameter2] ... # * Get/Query configurable values of a kernel parameter/s sysconfig -Q subsys [parameter1] [parameter2] ... # Meaning: # type=? typp: INT, UINT, LONG, ULONG e STRING # op=? Opertions: # C = Configurable (if there isn't R -> needs reboot) # R = Configurable at Runtime; # Q = Query # min_val=? min value # max_val=? max value # # * Get/Query a kernel parameters stored in: "/etc/sysconfigtab" sysconfig -d subsys [parameter1] [parameter2] ... # * Get/Query a kernel parameters stored in: "/etc/sysconfigtab" sysconfigdb -l [subsys] # * Set/Modify a Runtime kernel parameter sysconfig -r sybsys parameter=value # * Manpage of a specific subsystem (for kernel parameters) man sys_attr_subsys # * Notes: # - When you load subsys lag (man lag) you can't do the unconfigure # - To make static a dynamic module you need to rebuild de kernel # Modify by hand configuration file (section options) see: doconfig # - To make permanent the value of a parameter after a reboot, you must # edit the file: "/etc/sysconfigtab" (pay Attention!) # Manpages suggests to use: "sysconfigdb" # - On a cluster keep parameters in sync (...with some exceptions)
19. SNMP Simple Network Management Protocol
# * Information of the system: "SYSTEM" with COMMUNITY "public" # Change "public" if you have a variant (see: "/etc/snmpd.conf") snmp_request SYSTEM public get 1.3.6.1.2.1.1.1.0
20. Show Firmware firmware version
snmp_request SYSTEM public get 1.3.6.1.4.1.36.2.18.22.1.2.10.1.3.1
21. Event Management (evm)
evmget | evmsort | evmshow -T "%Y/%m/%d %T " -t "[@priority] @@" | more
22. System Temperature
# Load kernel module if not present (see sysconfig tips) # (sysconfig -c envmon; sysconfig -q envmon; sysconfig -u envmon) sysconfig -q envmon
23. Cluster and cluster alias
# * Cluster status
clu_get_info -raw
# (version 4)
cnxshow
# * Cluster alias information
cluamgr -s all # -> Configuration
cluamgr -a alias=alias_name,leave,rpri=0 # -> Disable an alias
clua_active alias_name # -> 0 if active, 1 otherwise
cluamgr -r start # -> Restart of alias routing
cluamgr -r info # -> Show some more information (Not documented!!!)
# * Cluster Alias Bilanced connections
cluamgr -a selp=10,rpri=10,alias=DEFAULTALIAS
# this on node two
cluamgr -a selp=5,rpri=5,alias=DEFAULTALIAS
# (cluamgr -a selw=3,selp=10,rpri=5,join,alias=DEFAULTALIAS)
# Service in "/etc/clua_services" must be in_multi
# * Configuration of a single member
# See: directory /.local.. -> cluster/members/{memb}
24. Memory channel (MC)
# * Show members of a cluster orderd by ID (versione 4.x)
cnxshow
cnxshow -v
cnxshow | tail +10 | cut -f1 -d' '
# * Kernel parameters for the Memory Channel
sysconfig -q rm
(see: rm_errors, rm_shutdowns - and manpage: sys_attrs_rm)
- Filesystem relocation on a cluster (cfsmgr)
cfsmgr -a SERVER=${HOSTNAME} -d ${D}
# where: HOSTNAME=is_the_host_name and D=domain_name
cfsmgr -h host2 -r -a SERVER=host1 /my_fs
# move: "/my_fs" from host2 to host1
# * Show FileSystems status (raw output)
cfsmgr -F raw
25. Network (net)
# * With "trailers" option you will increment performance (see manpage: ifconfig)
# but not all routers support this option. (PAY ATTENTION!)
ifconfig ee0 10.41.8.221 netmask 255.255.255.0 filter trailers
# * Switch of a interaece in NetRain (trunking) config
ifconfig ee1 switch # if Netrain is made by: ee0, ee1 (ee0 UP); switch ee0 to ee1
# * IP down (to create a netrain, interfaces must be without IP)
ifconfig ee0 down delete abort
# * Get/Show IP aliases
# Tru64 v5 "ifconfig" will show
# Tru64 v4 use instead "netstat -in"
ifconfig ee0
netstat -in | grep ee0 | awk '{print $4}' | tail +3
# * Get/Show all network interfaces
ifconfig -a
# * Get/Show routing table
netstat -nr
# * Set media speed of a network interfaces
lan_config -i ee0 -s 100 -x 1
26. Remote Procedure Call (RPC)
# * Get/Show all port associated with rpc rpcinfo -p # See: man portmap (port 111); man inetd (on 5.1B) # * Get/Show connected users "rpc.rusersd" rusers -al # * Get/Show systems state: uptime, load; "rpc.statd" # (will do a broadcast on the net) rup ruptime # * Send a message to all the network (like: wall) "rpc.rwalld" rwall rwall hostname shutdown -b # (broadcast to say: i will reboot soon) # * Test rpc "rpc.sprayd" spray # (see: man spray)
27. LSM change fault disk
# (is the same also for version 5) voldisk list voldg -g dg-1 -k rmdisk rzh69 voldisk rm rzh69 disklabel -z rzh69 disklabel -rw rzh69 auto voldisksetup -i rzh69 privlen=1024 nconfig=1 # privlen may change (5 use 4096) voldg -g dg-1 -k adddisk rzh69=rzh69 volrecover -sb rzh69 iostat rz43 1 volprint -Aht | grep wo - disklabel export (disklabel cloning) disklabel -r dsk35 > /tmp/dsk35 disklabel -Rr dsk39 /tmp/dsk35
28. Raid Controller
swxcrmgr # (before export display: export DISPLAY=1.1.1.1:0.0) /usr/lbin/hsxterm -b BUS -t TARGET -l LUN
29. Hardware Error (binary errorlog)
# * General query
ca u -R |more
wsea x analyze | more
wsea u ana | more
wsea u -o sum | more
uerf -R | more
dia -R
# * wsea
wsea n sum since=dd-mmm-yyyy,hh:mms
wsea n sum since=today
wsea n sum since=yesterday
wsea n sum begin=dd-mmm-yyyy,hh:mms
wsea n sum end=dd-mmm-yyyy,hh:mms
# it is the same with "ana" option (analize)
wsea n ana since=dd-mmm-yyyy,hh:mms
# * dia
# Provides a detailed report all I/O-related error events in reverse order.
dia -o terse -i ios -R
# Provides a short summary of all disk-related events as they occur.
dia -o brief -i disk -c
# start time / end time
dia -t s:dd-mmm-yyy,HH:MM:SS e:dd-mmm-yyy,HH:MM:SS
dia -t s:27-May-2006,18:21:00 s:27-May-2006,18:22:00 -o full
# Get/Show all type of events
dia ls evt | more
dia ana -i ios -i disk
# * uerf
# Use uerf if you don't have: wsea, dia, ca
# (use the last example)
# Provides a full report of all error events.
uerf -c err -o full
# Provides a detailed report of all error events in reverse order.
uerf -c err -o terse -R
# Provides a short summary of all error events as they occur.
uerf -c err -o brief -n
# In revers order
uerf -R -o full -f /cluster/members/{memb}/var/adm/binary.errlog | more
30. kdbx debugging crash
# * Some dbx commands to show system information kdbx -k vmunix /dev/mem (kdbx) p (page_size*physmem) / (1024*1024) -> memory in MBytes (kdbx) swap -> show swap (kdbx) p utsname -> show hotname # * Debugging on Tru64 5.X (crash analysis) cd /var/adm/crash dbx -k vmunix.0 vmzcore.0 p panicstr t # * Debugging with kdbx kdbx -k vmunix.0 vmzcore.0 sum t p utsname kps p $pid p *pmsgbuf # * Debugging with crashdc crashdc vmunix.0 vmzcore.0
31. Make a floppy disk with ufs Filesystem
disklabel -z floppy0c disklabel -rw floppy0c newfs /dev/rdisk/floppy0c mount -t ufs /dev/disk/floppy0c /mnt/floppy
32. Backup Restore
# * Backup on tape (ntape = not rewind tape) mt rew vdump -0 -f /dev/ntape/tape0_d1 /mnt # * Restore from tape mt rew vrestore -f /dev/ntape/tape0_d1 -x -D /mnt # * Remote Backup (ssh or rsh) tar cf - file | (ssh remote_host "cd remote_dir;tar xf -") vdump -0f - | ssh remote_host "/sbin/vrestore -D -f - " vdump -0f - | ssh remote_host "dd if=- of=" # * Remote vdump and local vrestore on a specific directory ssh remote_host "vdump -0f - " | vrestore -D -f -" - device tape creation (you need to create tape devices if you boot from CDROM) dn_setup -install_tape - device standard creation /dev/MAKEDEV std
33. LSM removal from boot disks (standalone system)
1) Modify link of /etc/fdmns/root_domain, usr_domain, var_domain 2) Edit: "/etc/sysconfigtab" and remove LSM (see also swap) 3) Remove all, execpt "types" from: "/etc/vol/" 4) Remove all from: "/dev/vol" and "/dev/rvol" 5) Edit: "/etc/inittab and remove 3 rows of LSM (vol*) - Persistent Reservation (PR) SCSI disk (clear reservation disk) # # * Script to clear Persistant Reservation (PR) # Works only for HP storages (may be useful) /usr/sbin/cleanPR # * How to remove PR from a disk # (shutdown every node of a cluster execpt one) scu scu> sbtl bus target lun scu> show keys scu> preserve clear key 0x8... - The scu command # * Get/Show disk information # sbtl -> set bus target lun # show nexus -> show bus target lun # tur -> test unit ready (disk in fail or good) # show inquiry -> show disk information # show capacity -> show disk size scu scu> sbtl 1 0 0 scu> show nexus ; tur ; show inquiry ; show capacity scu> show edt capabilities # * Make some disk diagnostic # Read from disk bus 3 target 1 lun 0 scu scu> sbtl 3 1 0 scu> set statistics on scu> set watch off scu> read media ending 100000 compare off bs 512 -> slow scu> read media ending 100000 compare off bs 914k -> top of the top usage # * Change disk attributes # Change wce (write cache enable -> danger) scu scu> sbtl bus target lun scu> show layout cache scu> show pages cache scu> set default savable off scu> set pages cache wce on scu> show pages cache # * Get/Show all scsi paths and capabilitis scu scu> show path all capa # # * Clear Persistent Reservation (PR) of a SCSI disk # (same as done before) scu scu> sbtl bus target lun scu> show keys scu> preserve clear key 0x8...
34. Commands: hwmgr, dsfmgr
# Information from : "man hwmgr_ops" # See: scsimgr for systems in verision 4.x # * Get/Show all disks and path_state plus transfer bytes hwmgr -get attr current -cat disk | grep -E "dev_base_name|path_state|path_xfer" # * Get/Show components in state: "no good" (only for v5.1B) # with "hwmgr -v" they are mapped with an 'X' hwmgr status component -ngood # * Remove state "no good" from a component (you need to hnow hwid first) hwmgr unindict -id 2 # * Remove all componets that are no more configured (removed componets) hwmgr refresh component # * Remove all path stale from scsi disks (speedup boot) hwmgr refresh scsi -all # * Get/Show all disks and paths hwmgr show scsi -type disk # * Scan bus scsi # Change "#N" with bus number (ex. 8) if you want to scan a specific BUS # (usefull to perform a scan without impact to other location - tape in primis) hwmgr -scan scsi [-bus #N] # * Get/see fiber adapter topology hwmgr -sho fibre hwmgr -sho fibre -adapter -topology # * Remove a component (ex. disco) # change "#N" with hwid (ex. 121) hwmgr -delete component -id #N # Not documented: remove hwid (valid for devices: tape, disk and scp) dsfmgr -Z rm_hwid #N 0 dsfmgr -R hwid #N # * Make a disk blink for 60 seconds (usefull to locate a disk) # 66 is the'hwid of a disk (hwmgr show scsi -type disk) # (use: "dd if=/dev/rdisk/dskXc of=/dev/null" -> to see a disk blinking) hwmgr locate component -id 66 -time 60 # * Get/Show all category hwmgr get category # * Get/Show network interfaces: name , speed, link status, negotiation type hwmgr get attr -cat network -a name -a media_speed -a full_duplex -a link # attr -> attribute # cat -> category # * Get/Show network interfaces: name and media type (fiber or UTP) hwmgr get attr -cat network -a name -a media_type # * Get/Show platform type/name hwmgr get attr -cat platform -a name # * Get/Show cpu type and speed hwmgr get attr -cat cpu -a sub_category -a speed # * Resets the device instances for all devices to the lowest possible (not all version) # Edit file: "/etc/dfsc.dat" (first rows) dsfmgr -v -I
35. Tunning and performance issues
# * Get/Show CPU lock info and statistics # (see manpages) lockinfo -percpu sleep 60 | more sched_stat sleep 60 | more # * Get/Show network statistics echo p tcbhashstats | dbx -k /vmunix /dev/mem netstat -s netstat -p tcp netstat -p udp netstat -p ip netstat -i -d 1 netstat -m sysconfig -q socket # drop_packets ? # * Cpu information # (if file: wtmp is big it is slow ...) pset_info # * RAD Information statistics sched_stat vmstat -R # * process RAD e NUMA information ps -ef -o pid,c,cputime,pcpu,user,longtname,nswap,RAD,state,command ps -ef -o pid,c,cputime,pcpu,user,longtname,nswap,state,command ps -ef -o NUMA # * Marvel (GS1280) NUMA cpu sysconfig -c marvel_pfm export DISPLAY=host:0. xmesh sysconfig -u marvel_pfm # * I/O on disks hwmgr -v d -type disk dd if=/dev/rdisk/dsk60c of=/dev/null bs=64k & hwmgr -get attr -id 357 | grep xfer # * Performance and statistics of disks # (for AdvFS si relative section) # * Get/Show read/write in bytes to disk drdmgr dsk0 drdmgr -F raw dsk0 # * csfmgr cfsmgr -a statistics # * Particular mount options (too speedup) mount -o delayed -o noatimes /dev/disk/dsk3c /tmp_files # (see: "man mount") # delayed -> Delays synchronously flushing metadata updates to disk # noatimes -> Marks file access time changes made for reads of regular files # in memory, but does not flush them to disk until other file # modification occur. # * Device timedev (Tru64) # Execute the following command in all cluster members # will speedup gettimeofday() call - so Oracle mknod /dev/timedev c 15 0 chmod ugo+r /dev/timedev # * gettimeofday() with microtime resolution # You must rebuild your kernel options MICRO_TIME
36. Quota filesystem
quotacheck -guv cluster_usr#usr
# Unaligned Access Message Control
uac
37. Striping
# Binary Stripping (to reduce text sixe)
# * Object Strip
/usr/ucb/ostrip
# Max strip (-z compres, but binary "slowe")
/usr/ucb/ostrip -gmcsxz /bin/ls # (on 4.0 does not work - use strip)
# (-g not useful)
# * Exe Strip
strip
# * Manipulates object file comment sections
mcs -pv /bin/ls
mcs -d -t TAGDESC /bin/ls
- Show hour:minuteseconds of a file (ls can't do this)
perl -e 'foreach(@ARGV){$t =localtime ( ( ( stat ( $_ ) ) [9] ) ); printf("%-20s%s\n",$_,$t);}' *
38. RMC: console environment; boot variables
# * Get/Show serial number
consvar -g sys_serial_num
SYS=$(echo "pd rpb->rpb_ssn;quit" | dbx -k /vmunix /dev/mem 2>/dev/null | tail -1)
echo "Serial Number = $SYS"
# * Booted kernel
sizer -b
# * Boot disk
consvar -g bootdef_dev
consvar -g bootded_dev
# * Firmware level
consvar -g version | cut -f2 -d'=' | awk '{print $1}'
# With "-v" it is more verbose
consvar -v -g version
# * "All" variables
consvar -v -l
# * Network adapters mode
# Adapters [tu DE500]: ewa0 -> tu0 ewb0 -> tu1
# DE500 must be configured in FastFD (also at switch level - port)
consvar -g ewa0_mode
# Adapters DE600 etc. must be configured at auto (autonegotiatio)
consvar -g eia0_mode
consvar -g ega0_mode
# * Boot parametrs
sysconfig -q bparm
# * Set RMC boot variable (it will be in memory: volatile)
consvar -s variable value
# * Save all values in eprom (non-volatile)
consvar -a
39. Memory test: for Alpha 8400 – 4100
Quote: >>> test ms7cc* # * Memeory test Quote: >>> set mode adv >>> set SIMM_CALLOUT on >>> init # * ESCAPE Quote: >>> [ESC][ESC]rmc -> GS40 >>> [ESC][ESC]scm -> GS320 >>> [ESC][ESC]MBM -> GS1280 (Marvel) # * wwidmgr (Get/Set fiber channel disks) Quote: >>> set mode diag >>> init # * Get/Show connected disks Quote: >>> wwidmgr -sho dev -full # * Get/Show fiber channel adapters (ex. emx) Quote: >>> wwidmgr -sho ada -full # # * Set a disk to be a boot disk # UDID taken form: "-sho dev" Quote: >>> wwidmgr -quickset -udid #N # * Reset Nvram (you need this when you change a fiber adapter) Quote: >>> wwidmgr -set ada -item 9999
40. Trace
# * Trace of a process
PID="1234"
trace_v51.exe -P -f -T -b 16384 -s 16384 -p ${PID} 2>&1 | tee /tmp/trace_${PID}.logReply Edit Delete Attach ToDo Report By rh77408 (Before you ask, the answer is "no") Date 2016-06-28 15:35Tru64 license details: (or licence if you are US based)
/usr/sbin/lmf list full
Product Name: ADVFS-UTILITIES
Producer: DEC
Issuer: DEC