Archive

Archive for the ‘Undocumented Commands’ Category

time

February 7th, 2009

This can be used to time a command. “time snapmirror quiesce vol_name” would quiesce the named snapmirror, and then print the time it took after running the command.

Undocumented Commands , , ,

source

February 7th, 2009

source <filename>

This will read and process the contents of whatever file you point to it. So if the filename is full of commands (for arguments sake /etc/rc), running source on it (“source /etc/rc”) would run all the commands.

If you have a series of commands that need to be run regularly, this would be a very efficient way of scripting this, just one ssh command could execute a whole series of commands.

Undocumented Commands , , ,

sd

February 7th, 2009

Interrogates 2 snapshots and can calculate the differences between them. Not necessarily block or size changes, but prints files, directories, ACL’s and so on. Potentially very useful, but also potentially very intensive!

Undocumented Commands , , ,

rtfo

February 7th, 2009

This can be very useful if there is a comms issue with the filer, or perhaps with replication that you are having problems monitoring or troubleshooting.

Make sure the gateway or router in question is in “/etc/mdgateways”. Then run “rtfo start”.

“rtfo status” will show you the counters. Rtfo will periodically run a ping, so this counter will constantly go up. Probably best not to run this command constantly. If there are any errors detected, then further investigation would be needed maybe in conjunction with [cref ifinfo]

Undocumented Commands , , , ,

result

February 7th, 2009

Prints the exit code of the previous run command. Very useful if you want to script the filers and have some error logging.

Undocumented Commands , , , , ,

raid_config

February 7th, 2009

filer1*> raid_config reconfig
maximum disks possible 56
maximum raid groups possible 400
maximum traditional volumes and aggregates possible 200
maximum stripes possible 32
average raid group size 14
No. of disks in system  3
No. of groups in system  1
No. of traditional volumes and aggregates in system  1
Maximum write disks active  56
Maximum write groups active 4
Maximum reconstructs  2
Maximum reconstruction size 8
Maximum disk copy operations  4
RAID buffer reserve  4132
RAID adjust buffer reserve  0
average raid disk write queue length  50
average raid disk read queue length   50
average raid disk write transfer rate 50
average raid disk read transfer rate  100

My favourite output from this command is “maximum disks possible”. I can never remember the max spindle count without looking it up, so this makes it easier!

If only this could be altered!

Undocumented Commands , ,

label / labelmaint

February 7th, 2009

I can only be hypothetical about this command, I can’t try it on any of our real filers (disks are expensive!), and on the simulator all disk labels are apparently invalid!

To change or edit a disks label, you need to put the disk into maintenance mode first.

labelmaint isolate disk_id

Once a disk is in maintenance, you can start relabelling it.

usage: label list | summary | examine | examinev1 | compare | edit | editv1 |
keep |restore | save | makeraid | makespare | buildtree | wipe | wipev1 |
vuinfo

Undocumented Commands , , ,

kt

February 6th, 2009

kt [-hfrt] [-s sk_cycles_per_msec]

        -f      forget sk_cycles_per_msec in registry
        -h      start or restart uninterruptable heartbeat message on console
                use ‘setflag heartbeat_debug_interval’ to set interval
               in minutes (default is 0 [off])
        -r      print current time in RTC
        -s <value>      set sk_cycles_per_msec
        -t      print current time parameters

I haven’t figured this one out yet…

Undocumented Commands , ,

inodepath

February 6th, 2009

filer1*> inodepath
usage: inodepath { -v <name> | -f <fsid> | -u <uuid> }
                 [-ainp] [ -d <snapid> | -s <snapname> ] <inum>
For more information, try ‘inodepath help’ or ‘man inodepath’
filer1*> inodepath help
usage: inodepath { -v <name> | -f <fsid> | -u <uuid> }
                 [-ainp] [ -d <snapid> | -s <snapname> ] <inum>
Where <inum> and one of -f, -u, or -v are required on the command line and:
  -f <fsid>        A volume -fsid-
  -u <uuid>        A volume -uuid- (currently not implemented)
  -v <name>        A -volume- name
  -a               List -all- possible names
                   (default is just the primary name)
  -i               List parent -info- values
                   (default is to not list them)
  -n               List file -names-
                   (default is to not list them)
  -p               Do not list -path- names
                   (default is to list them)
  -d <snapid>      A -snapshot- name
  -s <snapname>    A snapshot -id-
  <inum>           An inode number from the specified volume
                   (or snapshot within the volume)
For more information, try ‘inodepath help’ or ‘man inodepath’

I honestly can’t see a constructive use for this out in the field or even in the lab!

Undocumented Commands , ,

ifinfo

February 6th, 2009

ifinfo -a

Output a lot of config / debug information on the interfaces. Could be useful if troubleshooting network problems in combination with [cref rtfo].

Undocumented Commands , ,



This site is not affiliated or sponsored in anyway by NetApp or any other company mentioned within.