***********************************************************************************

****
pmx[hfdnsckw] [<mofilter>|<mogroup>] [<counter-filter>] [-l <zipfile>|<directory>]
[-w <webdirectory>] [-m <minushours>] [-p <plushours>] [-s
<startdate>[.<starttime>]] [-e <enddate>[.<endtime>]] [-a|-d|-h] [-o
<outputFormat>] [| <unix-command>]
***********************************************************************************
****
Display counter values, extracted from the statistics ROP files.

This command processes the ROP files by using the utilities

Arguments:
- the first argument (mofilter/mogroup) is to specify the MO instances whose
counters shall be printed: can be identified by an MO group or by a regular
expression matching the LDN of the MOs.
- the second argument (counterfilter) is to specify a regular expression matching
the name(s) of the counter(s) that shall be printed.
If the first and second arguments are omitted, the XML ROP files are parsed and the
result is stored on file instead of being displayed to screen. The result files
will be stored in the same location on the workstation as where the XML ROP files
are stored, ie: ~/moshell_logfiles/logs_moshell/pmfiles/nodeipaddress. This
location can be changed with the user variable pm_logdir.

Options:
- The "h" option (eg. "pmxh") is for displaying the counters side-by-side ("h" as
in "horizontal"). Otherwise, there is one line for each MO instance and counter.
- The "f" option (eg. "pmxf", or "pmxhf") is to prevent pmx from rechecking if
there are any newer xml files ("f" as in "fast").
- The "d" option is mandatory for parsing PEG counters which are not reset at the
end of each ROP period. In "pmom" command, these counters are marked with the tag
"peg,noReset". The command "pmomd . . . peg,noreset" will show all such counters.
- The "n" option is for aggregating all counters on MO level.
- The "k" option is for keeping the counters marked with the suspected faulty tag
"<sf>TRUE</sf>", without this option they are discarded.
- The "s" option is for flagging the counters with negative or empty value, or
those marked with the suspected faulty tag "<sf>TRUE</sf>". The counters are then
displayed with a label showing "SuspectedFaulty", "NegativeCounter", or,
"EmptyCounter".
- The "c" option is for displaying the number of counter instances in the ROP
file. Should be identical to the number of active counter instances shown in
"pgetsn".
- The "w" option is for generating a web report in table and graph format. The "-
w" switch can be used to specify the location of the webpage. Currently works best
with Firefox, avoid IE.

Switches:
- The "-m", "-p", "-s", "-e" are used for specifying a time span, see "h pmr" for
more information. By default, the time span will only cover the latest ROP file
(equivalent to "-m 0.25").
- The "-a", "-d", "-h" switch is for specifiying if any time aggregation should be
used. If the switch is not included, then no time aggregation will occur. Othewise,
time aggregation can be done on hour basis ("-h"), day basis ("-d") or on the whole
timespan ("-a").
- The "-l" option is for specifying a directory containing ROP files. In this
case, pmx will parse the ROP files stored in that directory instead of from the
node. Can also be used in offline mode.
 - The "-o" switch is for specifying the output format. By default, the output is
text but it is possible to specify: "csv", "html", or "htmltab".
- The "-w" switch is for specifiying the path where the web report will be stored.
Only applicable with option "w" ("pmxw"). When "-w" is not given, a default
location will be chosen for the web report.

It is possible to pipe the command into a unix command (eg: grep, sort, etc).
Examples:
>> pmx utrancell downtime --> all counters matching "downtime" on all MOs
matching "utrancell" will be displayed for the last ROP period
>> pmxh utrancell downtime ---> same as above but the counters will be
displayed side-by-side
>> pmxh utrancell downtime -m 3 -a --> same as above but the last 3 hours (12 ROP
periods) will be read and aggregated into one value
>> pmxhf utrancell downtime -m 3 -a | sort -k 2 --> print same as above without
rechecking for new ROP files ("f" option), then sort on the second field ("| sort -
k 2")
>> pmx -m 2 --> fetch and parse all ROP files from the last two hours, result to
be stored on disk
>> pmxf -s 19000101 ---> parse all ROP files that currently exist in the pmlog
directory in the workstation
>> pmxs -m 0.25 ---> show the counters marked as suspected faulty for the last 15
minute ROP period.
>> pmxw utrancell=iub-10-1 pmTotNoRrcConnectReq$ -m 24 --> show the evolution in
graphical format (web) over the past 24 hours of the pmTotNoRrcConnectReq counter
on the MO utrancell=iub-10-1

Note: only counters that are included in an active performance scanner will be
displayed.
Performance scanners can be read with the "pgets" command and created with the
"pcr" command or from OSS.
In the case of RAN nodes (RNC, RXI, RBS, eNodeB), the OSSRC does not allow external
tools (such as moshell) to create/delete/modify a PM scanner. Any attempt to do
this will result in OSSRC reverting the change. This is a feature of OSS (write
TR/CRs on OSS, not to MoShell!) and applies only to RAN nodes, not MGW.