Last year the Microsoft PowerShell Team announced that Microsoft was going to provide SSH server software, a staple for remote management of Unix and Linux systems and server software available by default with OS X. I would certainly be appreciative of having the same command line tools, such as AWK, grep, etc. available under a Bash shell on Linux systems on Microsoft Windows systems, also. And, since I usually install a third-party SSH server application, such as Bitvise SSH Server, Copssh, freeSSHd, PowerShell SSH Server for Windows, or SilverSHield, to be able to remotely connect to Microsoft Windows for command line troubleshooting, I'm hoping that will be available by some this year or next as well, though I'm not aware of the date it is expected to become available.

References:

1. Ubuntu’s bash and Linux command line coming to Windows 10
   By: Peter Bright
   Date: March 30, 2016
   Ars Technica
2. Windows 10: /bin/bash-supporting “Anniversary Update” coming this summer
   By: Peter Bright
   Date: March 30, 2016
   Ars Technica
3. Microsoft bringing SSH to Windows and PowerShell
   By: Peter Bright
   Date: June 2, 2015
   Ars Technica
4. Windows Services for UNIX Version 3.0
   Microsoft TechNet
5. Looking Forward: Microsoft Support for Secure Shell (SSH)
   By: PowerShell Team
   Date: June 3, 2015
   Windows PowerShell Blog

[/os/windows] permanent link

[ More Info ]

[/os/os-x/802_1x] permanent link

[ More Info ]

[/os/unix/commands] permanent link

[/network/web/html/css] permanent link

One way you can add a new entry in the registry for a program to run when the user logs in is to use the regedit program that comes with the Windows operating system and which is normally found in the C:\Windows directory.

[ More Info ]

[/os/windows/registry] permanent link

[ More Info ]

[/os/os-x] permanent link

-v, --invert-match
       Invert the sense of matching, to select non-matching lines.  (-v
       is specified by POSIX.)

E.g., suppose I have a file names.txt containing the following names:

$ cat names.txt
John Smith
Paul McCartney
Bob Jones
Allen Smith
Greg Smith
Bob Smith
Carl Smith
John Doe

If I want to view all lines except for those containing "Allen", I can use grep -v "Allen" names.txt. But what if I want to exclude more than one pattern, e.g. any line containing "Allen" or "Bob". I could pipe the output of one grep command to another grep command with grep -v "Allen" names.txt | grep -v "Bob". Or you can perform a logical disjunction using the "pipe" character, i.e., "|", aka a "vertical bar".

$ grep -v "Allen\|Bob" names.txt
John Smith
Paul McCartney
Greg Smith
Carl Smith
John Doe
$

In the above example, I am instructing grep to ignore any lines containing either Allen or Bob in the line. Because the pipe character has another meaning to the Bash shell, i.e., it is used by the shell to "pipe" the output of one command to another with the output of the first command becoming the input of the second, its meaning must be "escaped" to be processed by grep as a logical disjunction symbol. That is done by preceding the character with a backslash, which is an escape character.

[/os/unix/commands] permanent link

I was able to access webpages that were not blog postings. When I checked the site's error log file, I saw many entries similar to the following indicating that others were experiencing the same problem when accessing the site:

[Tue Mar 22 11:43:12.276013 2016] [cgi:error] [pid 24979] [client 136.243.36.80:
52035] AH01215: Magic number checking on storable file failed at /usr/lib64/perl
5/vendor_perl/Storable.pm line 381, <DATA> line 32, at /home/jdoe/public_html/su
pport/blog/plugins/calendar line 322.

I use a Blosxom calendar plugin, so I checked line 322 in the calendar plugin file and found the following code:

    if (!Storable->can('lock_retrieve')) {
        debug(1, "cache disabled, Storable::lock_retrieve not available");
        $use_caching = 0;
        return 0;
    }
    $cache = (-r $cachefile ? Storable::lock_retrieve($cachefile) : undef);

The line starting with "$cache" is line 322. I checked the cache file for the plugin which is under the plugins/state directory for the blog software and saw it was zero bytes in length.

# ls -al /home/jdoe/public_html/support/blog/plugins/state/.calendar.cache
-rw-r--r--. 1 apache apache 0 Mar 22 11:42 /home/jdoe/public_html/support/blog/plugins/state/.calendar.cache

So I deleted the file; it will be recreated automatically when the blog is accessed after it is deleted.

# rm /home/jdoe/public_html/support/blog/plugins/state/.calendar.cache
rm: remove regular empty file ‘/home/jdoe/public_html/support/blog/plugins/state/.calendar.cache’? y

Deletion of the .calendar.cache file fixed the problem. When I refreshed the page in my browser for a blog posting from years ago I had been attempting to view I could then see it and access other blog postings as well. I also saw the file had been recreated.

# ls -al /home/jdoe/public_html/support/blog/plugins/state/.calendar.cache
-rw-r--r--. 1 apache apache 94578 Mar 22 21:54 /home/jdoe/public_html/support/blog/plugins/state/.calendar.cache

[/network/web/blogging/blosxom] permanent link

[More Info ]

[/os/os-x/software/vlc] permanent link

[ More Info ]

[/os/windows/network] permanent link

[ More Info ]

[/os/os-x] permanent link

$ networksetup -listallhardwareports | grep -A 1 "Wi-Fi"
Hardware Port: Wi-Fi
Device: en0
$ networksetup -getairportnetwork en0
Current Wi-Fi Network: Hogwarts

[/os/os-x] permanent link

<string>14.6.0</string>

I only want to see the 14.6.0, so I can use the grep command with the -o option to specify I ony want the text that matches a specified pattern displayed. The pattern I can use is '[0-9.]\+'.

[ More Info ]

[/os/os-x] permanent link

$ gksudo modprobe -r psmouse

(gksudo:2607): Gtk-WARNING **: cannot open display:

I then switched back to the graphical user interface (GUI) console by hitting Ctrl-Alt-F7. I was then able to move the mouse pointer using the touchpad.

References:

1. Resetting the mouse in Ubuntu
   Date: August 1, 2015
   MoonPoint Support

[/os/unix/linux/ubuntu] permanent link

# hostname
moonpoint

Yet, if I tried entering any option for the command, I received the error message "Name or service not known".

# hostname -a
hostname: Name or service not known
# hostname -i
hostname: Name or service not known
# hostname -f
hostname: Name or service not known
# hostname --fqdn
hostname: Name or service not known

[ More Info ]

[/os/unix/linux/centos] permanent link

There are a variety of methods you can use to check on which account PHP is running under, e.g., <?php passthru("whoami"); ?>, though some methods may not work on some systems.

[ More Info ]

[/languages/php] permanent link

I was able to stop it from continually crashing by choosing not to reopen all of the spreadsheets that had previously been opened. When it displayed the message "This workbook contains macros. Do you want to disable macros before opening the file?", I chose "Do Not Open" for some of the workbooks that previously had been open.

I was then able to successfully open the application and continue working on other workbooks that had been open in Excel.

I checked for the crash logs related to the problem in ~/Library/Logs/DiagnosticReport. I saw two crash reports for today, March 11, 2016.

$ ls ~/Library/Logs/DiagnosticReports/
Google Chrome_2016-02-24-131912_GSSLA15122293.crash
Microsoft Excel_2016-03-11-114210_GSSLA15122293.crash
Microsoft Excel_2016-03-11-141557_GSSLA15122293.crash
PGP Engine_2016-02-22-212124_GSSLA15122293.crash

When I checked both Excel crash reports, I found that the cause was logged as a stack overflow. I looked for the line that immediately follows the "Application Specific Information:" line in the files, which is why I used the -A 1 argument to the grep command.

GSSLA15122293:/ johndoe1$ grep -A 1 "Application Specific Information:" ~/Library/Logs/DiagnosticReports/Microsoft\ Excel_*.crash | grep -v "Application"
/Users/johndoe1/Library/Logs/DiagnosticReports/Microsoft Excel_2016-03-11-114210_GSSLA15122293.crash-[20077] stack overflow
--
/Users/johndoe1/Library/Logs/DiagnosticReports/Microsoft Excel_2016-03-11-141557_GSSLA15122293.crash-[23349] stack overflow

[/software/office] permanent link

CVE-2016-0960
CVE-2016-0961
CVE-2016-0962
CVE-2016-0963
CVE-2016-0986
CVE-2016-0987
CVE-2016-0988
CVE-2016-0989
CVE-2016-0990
CVE-2016-0991
CVE-2016-0992
CVE-2016-0993
CVE-2016-0994
CVE-2016-0995
CVE-2016-0996
CVE-2016-0997
CVE-2016-0998
CVE-2016-0999
CVE-2016-1000
CVE-2016-1001
CVE-2016-1002
CVE-2016-1005
CVE-2016-1010

[ More Info ]

[/network/web/browser] permanent link

$ sudo systemsetup -getnetworktimeserver
Password:
Network Time Server: time.example.com

If you wish to set the time server you can use the -setnetworktimeserver command, e.g.:

$ sudo systemsetup -setnetworktimeserver us.pool.ntp.org

[/os/os-x] permanent link

[ More Info ]

[/network/web/server/apache] permanent link

$ uname --nodename
localhost.localdomain

You can also use the hostname command on a Linux or OS X system to show or set the host name. To show the host name, issue the command without any parameters

$ hostname
localhost.localdomain

On some versions of Linux, e.g., CentOS and Ubuntu, you can also view the host name using the sysctl command; it won't work on OS X, however.

[ More Info ]

[/os/unix] permanent link

PS C:\> get-process

Handles  NPM(K)    PM(K)      WS(K) VM(M)   CPU(s)     Id ProcessName
-------  ------    -----      ----- -----   ------     -- -----------
    100       9     1472        848    71    14.06  41452 adb
1018385      17     3028       1008    99 5,303.64 101076 AdobeARM
     79       7     1200       1336    44           56112 armsvc
    127      10     7956       9444    37   101.42 120812 audiodg
   9244     279   233224      22368  1019            1916 avp
   1334      72    51656       3436   263 1,856.22  11692 avp
    427      34    18864      11000  1510            1984 certsrv
    321      29    52752      71604   501     3.73   6524 chrome
    249      25    44344       9356   290   801.09   7820 chrome
<text snipped>
    268      20     4876      11112    60          121708 w3wp
    227      21   618252      31084   688            5428 wbengine
     84       8      892        324    41             656 wininit
    181      10     2432       2752    58             684 winlogon
    364      44    13596      17920   153 2,894.47 102896 WinSCP
   1672      20     9228      18008    81           52428 WmiPrvSE
   4884      20    16476       8936    61            3252 WSSBackup


PS C:\>

The column values are as follows:

• Handles: The number of process handles that the process opened. A handle is an integer that Windows assigns to processes. For instance, each process thread is typically assigned a handle.
• NPM(K): Non-paged memory the process is using, in kilobytes.
• PM(K): Pageable memory the process is using, in kilobytes.
• WS(K): Process working set, in kilobytes. The value refers to the number of memory pages that the process recently accessed.
• VM(M): Virtual memory the process is using.
• CPU(s): Processor time used on all processors, in seconds.
• Id: Process Identifier (PID).
• ProcessName: The name of the process.

[ More Info ]

[/os/windows/PowerShell] permanent link

WORKDAY(start_date, num_days, [holidays])

For OpenOffice Calc, you need to separate the parameters with a semicolon rather than a colon. I.e.:

WORKDAY(Start_date; num_days; [holidays])

If you separate the parameters with a comma in Calc, it will display Err:508.

"Holidays" is enclosed in brackets to indicate it is an optional parameter, you would not actually include the brackets if you specified holidays.

The arguments to the function have the following meaning:

• Start_date - Required. A date that represents the starting date for a calculation
• num_days - Required. The number of nonweekend and nonholiday days before or after the start_date. A positive value yields a future date while a negative value yields a past date.
• holidays - Optional. An optional list of one or more dates to exclude from the working calendar, such as state and federal holidays and floating holidays. The list can be either a range of cells that contain the dates or an array constant of the serial numbers that represent the dates.

E.g., I need to determine the date when work requests that should be completed within 5 business days after approval should be completed. I can display the required implementation date for each request in a worksheet by using the formula =WORKDAY(cell,5) where cell holds the approval date. E.g., if the approval date is in cell A2, I can set the formula for the required implemenation date to be WORKDAY(A2,5) for Excel or Google Sheets or WORKDAY(A2;5) for Apache OpenOffice Calc.

If there was no date in cell A2, Excel will display January 6, 1900 in the cell holding the implemenation date, e.g. 1/6/00, if the date format is D/M/YY (day/month/2-digit year), but OpenOffice Calc and Google Sheets will display January 5, 1900. Microsoft Excel stores dates as sequential serial numbers so they can be used in calculations and, by default, uses January 1, 1900 as serial number 1. If you have a date field containing January 1, 2008 that is equivalent to 39,448, which you would see if you put that day in a cell formatted to hold a date then changed the format to be a number. January 1, 2016 is 42,370. January 1, 2008 is 39,448 days after January 1, 1900 and January 1, 2016 is 42,370 days after that date.

Why do Google Sheets and Apache OpenOffice Calc show January 5, 1900 as the calculated date when there is no date in the start_date field while Microsot Excel displays January 6, 1900? Long before Microsoft became a dominant player in the spreadsheet market, Lotus 1-2-3 from Lotus Development Corporation was a dominant spreadsheet in the personal computer market. There was a bug in Lotus 1-2-3 due to the developers assuming that the year 1900 was a leap year; it was not. When Microsoft released Excel they had to compete with Lotus 1-2-3, the then dominant spreadsheet for personal computers, so Microsoft needed to ensure that spreadsheets created in Lotus 1-2-3 returned the same results in Excel, so for compatibility retained the incorrect assumption regarding the year 1900 - see the Microsoft article Excel incorrectly assumes that the year 1900 is a leap year and Excel Date Conversion (Days from 1900) regarding making an allowance for that incorrect assumption. Google Sheets and Apache OpenOffice Calc did not retain that incorrect assumption regarding the year 1900 and so you will see January 5 used in certain calculations in those spreadsheets where January 6, 1900 is used by Exel, which is why sometims you may get differing results for date calculations between those spreadsheet programs.

If an argument to the workday function is not a valid date, e.g., 2/30/16 for February 30, 2016, which is not a valid date, you will see #VALUE! displayed by the function in all three spreadsheets. In all three spreadsheets, if days is not an integer, the number is truncated to an integer value. E.g., if you used the formula =WORKDAY(A2,5.7) you would get the same result as entering WORKDAY(A2,5).

[/os/windows/office/excel] permanent link

#!/usr/bin/python

# download_page
# download a webpage to a specified file. The script takes two parameters:
# the URL of the page to download and a file name to be used to hold
# the downloaded web page.

import urllib2, sys

try:
   sys.argv[1]
except IndexError:
   print "Error - URL missing! Usage: ./download_page.py download_page_url outfile"
   sys.exit(1)
else:
   url = sys.argv[1]

try:
   sys.argv[2]
except IndexError:
   print "Error - missing output file name! Usage: ./download_page.py download_page_url outfile"
   sys.exit(1)
else:
   outfile = sys.argv[2]

page = urllib2.urlopen(url)
source = page.read()

downloadFile = open(outfile, 'w')
downloadFile.write(source)
downloadFile.close()

The sys module is imported to check the command line arguments using sys.argv[x], where x. is the number specifying the argument; sys.argv[0] is always the name of the script itself, in this case download_page.py, so sys.argv[1] should be the URL of the webpage to be saved and sys.argv[2] the file name for the output file. The file name can contain a location for the output file, e.g., mydir/somepage.html. If a directory is specified with the file name, the script doesn't check to ensure the directory exists and will exit with a Python "No such file or directory" error message should that error occur. If no directory path is included with the file name, the directory from which the script is run will be used to store the downloaded webpage.

The script will print error messages if the URL and output file name are omitted from the command line. It can be run using python ./download_page.py or ./download_page.py, if for the latter option you have first changed the file permissions on the program to mark it as executable, e.g., with chmod 755 download_page.py.

download_page.py

[/languages/python] permanent link

[ More Info ]

[/network/web/search] permanent link