Search Results

Search found 8028 results on 322 pages for 'unix shell'.

Page 238/322 | < Previous Page | 234 235 236 237 238 239 240 241 242 243 244 245  | Next Page >

• sudo prompts for password over ssh

  - by Joe Watkins

  I have sudo set up for a shell script as follows on "hostname" (sudo -l output): (suser) NOPASSWD: /path/script* sudoers content is: myuser ALL=(suser) NOPASSWD: /path/script* this works fine, so I can run the following, logged in locally on hostname, without need for password: sudo -u suser /path/script however, when I use ssh (with keys set up, so no password require) to login and run, as follows: ssh hostname sudo -u suser /path/script I get prompted for a password, and when the password is entered I get: Sorry, user myuser is not allowed to execute '/path/script' as suser on hostname. Why? NB the following does not prompt for password at any point: $ ssh hostname $ sudo -u suser /path/script

  Read the article

• Windows equivalent to this Makefile

  - by Sridhar Ratnakumar

  The advantage of writing a Makefile is that "make" is generally assumed to be present on the various Unices (Linux and Mac primarily). Now I have the following Makefile: PYTHON := python all: e installdeps e: virtualenv --distribute --python=${PYTHON} e installdeps: e/bin/python setup.py develop clean: rm -rf e As you can see this Makefile uses simple targets and variable substitution. Can this be achieved on Windows? By that mean - without having to install external tools (like cygwin make); perhaps make.cmd? Typing "make installdeps" for instance, should work both on Unix and Windows.

  Read the article

• Securing files on IPhone

  - by clearbrian

  Hi Is there a way to decompile the binary from an IPhone app. I jailbroke my IPhone and was surprised to find other app's dbs wide open to be copied. So I exported my most important table and hardcoded it into code. Instead of loading table into array from a db I just generated code to fill the array and kept only the most basic DB info so relationships still work. Took a while but now works fine. I was just wondering am I safe, could someone decompile the binary for the app easily and extract the data. In Java its easy to decompile *.class files though thats bytecode where I presume iphone apps are more low level. I know IPhone sdk 4 can mark files as secure. Anyone know can this be overridden by jailbreaks or is this an unix lock?

  Read the article

• Any good way to set the exit status of a Cocoa application?

  - by buglesareking

  I have a Cocoa app which interacts with a server and displays a GUI. If there is a fatal error, I display an alert and exit. I'd like to set the exit status to a non-zero value to reflect that an error occurred, for ease of interaction with some other UNIX based tools. Unfortunately I've been unable to find a good way to do so - NSApplication doesn't seem to have any way to set an exit status. At the moment, I've subclassed NSApplication and added an exitStatus ivar (which I set in my app delegate when necessary), then overridden -terminate: so that it calls exit(exitStatus). This works fine, but it seems a bit grungy to me, not to mention that I may be missing something important that the stadnard `terminate: is doing behind the scenes. I can't call [super terminate:sender] in my subclassed method, because that exit()s without giving me a chance to set the status. Am I missing something obvious?

  Read the article

• Which is better for multi-use auth, MySQL, PostgreSQL, or LDAP?

  - by Fearless

  I want to set up an Oracle Linux 6 server that gives users secure IMAP email (with dovecot), Jabber IM, FTP (with vsftpd), and calDav. However, I want each user logon to be able to authenticate all services (e.g. Joe Smith signs up once for a username and password that he can use for email, ftp, and his calendar). My question is, which database service will be best suited for that application? Also, is there a way to link the database with the preexisting server shell logins (e.g. so I can read the root account's LogCheck emails on a different device)?

  Read the article

• Chrome SSL Security Issue under Windows systems?

  - by fraido

  The Fortify.net website allows you to check what SSL Encryption key is used by your browser. I gave it a try with the browsers I've on my machine and these are the results Fedora 9 Firefox 3.0.8 = AES cipher, 256-bit key Chrome 4.0.249.30 = AES cipher, 256-bit key Windows XP SP3 IE 6.0.2x = RC4 cipher, 128-bit key Firefox = AES cipher, 256-bit key Chrome 4.1.249.1042 (42199) = RC4 cipher, 128-bit key .... WHAT!!?!! Chrome is using RC4 128-bit (as IE6 does) that is well known as been very weak! Chrome under Unix works fine... I'm wondering how is this possible? Do you have this issue or is there a way to change the default key to be AES 256bit? I'm using Chrome as the main browser under Windows and I'm really considering to switch back to Firefox

  Read the article

• Open a custom remote powershell remotely

  - by Yann

  I have 2 computers. On the computer A, I have a custom module written in C# for powershell 3.0 and installed via a MSI. I also have a shortcut that open powershell with the module already loaded. I can just double click on my shortcut and run my command Do-Something on this computer without any problem, like the Exchange Server powershell. But now I would like to do it from a remote session on computer B in C#. So my question is, how can I open a remote powershell session to computer A with my module already loaded and the shell configured so I can just run my command and obtain the same result than if I run it on computer A?

  Read the article

• How do I easily elevate when running a .jar file?

  - by Merlyn Morgan-Graham

  When trying to run an installer Jar file, I am getting an error saying that write access is denied to create a directory under the Program Files folder. Right click - Run as Administrator is not available on Jar files (I assume because it is Java.exe that consumes them - they are not themselves treated as directly executable by the shell). What is the quickest and simplest way to run a .Jar file with elevation? I am evaluating this tool to recommend for our dev team, and they will manually install it on their boxes. I'd prefer an option that doesn't require them to type anything.

  Read the article

• How to synchronize users, passwords, hosts, etc without NIS

  - by joshxdr

  I am administering a very small solaris 2.6 network with 4 boxes total. Is it possible to use scp or similar to replace NIS for synchronizing users, groups, hosts, etc? This network is only a small part of my job and I don't want to spend too much time on it, and I am worried the setup and maintenence of NIS will not pay off. I need it to behave like a proper multi-user system, when a user logs into any machine, the users, passwords, hosts, etc. are always the same. Is there an easy way to do this with scp? Right now I copy /etc/passwd from one box to another with scp, but sometimes I make mistakes or forget a step, and scp inside of shell scripts don't seem to works so well since they require password authentication. Any recommendations would be welcome.

  Read the article

• Security: Unable to display current owner

  - by Jim McKeeth

  I have a user with local administrative rights on their Windows 7 64-bit box. They extracted a file from a zip file they downloaded with the Zip shell extension to their desktop on a non-encrypted or compressed NTFS volume. Unfortunately they cannot open it. I tried to take ownership of it and it just reports "Access is Denied" and for the current owner it reports "Unable to display current owner." At this point the file cannot be deleted, executed or modified. I have actually seen this exact same happen in Vista before, but I can't for the life of me remember how to fix it. Any ideas how it happened or how to fix it? I'd be happy just to delete it without a format at this point.

  Read the article

• Can I autologin a specific user on a specific VT, but still have other login shells running?

  - by Michael

  I'm running a linux system with Ubuntu Karmic on it, and I use it to run a webcam and a program that continuously outputs on the console. Is there any way I can force a user to login on VT2, and then switch VTs to VT2, but still have other login consoles running so I can hook a keyboard up to it and CTRL-ALT-F# to get to another terminal? I have a feeling this needs to go in my rc.local file, but nothing in the login manpage seems to suggest it can do what I want it to do. If there is no way to login a user on the vt, launching the program directly on a specific VT will work as well, since the user's shell in /etc/passwd is set to the program anyway.

  Read the article

• Sending mail via sendmail from python

  - by Nate

  If I want to send mail not via SMTP, but rather via sendmail, is there a library for python that encapsulates this process? Better yet, is there a good library that abstracts the whole 'sendmail -versus- smtp' choice? I'll be running this script on a bunch of unix hosts, only some of which are listening on localhost:25; a few of these are part of embedded systems and can't be set up to accept SMTP. As part of Good Practice, I'd really like to have the library take care of header injection vulnerabilities itself -- so just dumping a string to popen('/usr/bin/sendmail', 'w') is a little closer to the metal than I'd like. If the answer is 'go write a library,' so be it ;-)

  Read the article

• Cannot import SQLite with Python 2.6

  - by David McLaughlin

  I'm running Python 2.6 on Unix and when I run the interactive prompt (SQLite is supposed to be preinstalled) I get: [root@idev htdocs]# python Python 2.6 (r26:66714, Oct 23 2008, 16:25:34) [GCC 3.2.2 20030222 (Red Hat Linux 3.2.2-5)] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import sqlite Traceback (most recent call last): File "<stdin>", line 1, in <module> ImportError: No module named sqlite >>> import sqlite Traceback (most recent call last): File "<stdin>", line 1, in <module> ImportError: No module named sqlite >>> How do I resolve this?

  Read the article

• How do I launch a process as a specific user at startup on OS X?

  - by Scott Bonds

  I would like to run a script as a particular user on startup (not on login). I thought a launchd LaunchDaemon would do it, but 'man launchd' says: "If you wish your service to run as a certain user, in that user's environment, making it a launchd agent is the ONLY supported means of accomplishing this on Mac OS X. In other words, it is not sufficient to perform a setuid(2) to become a user in the truest sense on Mac OS X." They aren't kidding--when I try to run my script as a LaunchDaemon it doesn't work. In particular I'm trying to automate some keychain operations using the 'security' command, and it won't let me change the default keychain when I run the script through LaunchDaemon, though the script works fine when run using sudo from a shell. A LaunchAgent won't work, because the goal is for the proces to run without a user logging in and LaunchAgents only run when someone logs in. I looked at cron and the @reboot directive and that looks promising, but I read that cron is deprecated on OSX.

  Read the article

• socket programming: How do I handle out of band data

  - by soulmerge

  I just looked into wikipedia's entry on out-of-band data and as far as I understand, OOB data is somehow flagged more important and treated as ordinary data, but transmitted in a seperate stream, which profoundly confuses me. The actual question would be (besides "Could someone explain what OOB data is?"): I'm writing a unix application that uses sockets and need to make use of select() and was wondering what to do with the exceptfds parameter? Do I need to put all my sockets into this parameter and react to such events? Or do I just ignore them?

  Read the article

• How do I shorten the repository URL using svn+ssh similar to svnserve -r?

  - by Marcus

  In the svnbook, it shows you how to shorten the URL to your repositories when using svnserve as a daemon, using -r like: svnserve -d -r /usr/local/repositories That way, you can refer to the repository you need right after the hostname in the URL without revealing any of the local path (which is /usr/local/repositories/project1): svn checkout svn://host.example.com/project1 However, now that I am switching to svn+ssh, I have the local path back in my repository URL: svn checkout svn+ssh://host.example.com/usr/local/repositories/project1 Does anyone know how to hide that local path and use a shorter URL as up above, using svn+ssh and WITHOUT using a UNIX soft link on the svn server? (you still end up with an extra string in the URL if you use a soft link...) UPDATE: The solution to this can be found in the accepted answer over on ServerFault (the green-checked answer). Yay!

  Read the article

• crontab doesn't work on centos 6

  - by sbxmal

  /etc/crontab: SHELL=/bin/bash PATH=/sbin:/bin:/usr/sbin:/usr/bin MAILTO=root HOME=/ 01 * * * * root run-parts /etc/cron.hourly 02 4 * * * root run-parts /etc/cron.daily 22 4 * * 0 root run-parts /etc/cron.weekly 42 4 1 * * root run-parts /etc/cron.monthly */2 * * * * /usr/bin/php -q /home/dorsen/public_html/cron-test.php */2 * * * * http://www.mysiteurl.com/cron-test.php - also doesn't work If i run this file directly in browser - all fine What i did : root@vded10625 [~]# yum remove cronie-anacron .... Complete! root@vded10625 [~]# yum install cronie-noanacron sysstat .... Complete! root@vded10625 [~]# service crond start Starting crond: [ OK ] root@vded10625 [~]# chkconfig crond on root@vded10625 [~]# service crond restart Stopping crond: [ OK ] Starting crond: [ OK ] Will be pleased for any help

  Read the article

• removing a line from a text file?

  - by Blackbinary

  Hi all. I am working with a text file, which contains a list of processes under my programs control, along with relevant data. At some point, one of the processes will finish, and thus will need to be removed from the file (as its no longer under control). Here is a sample of the file contents (which has enteries added "randomly"): PID=25729 IDLE=0.200000 BUSY=0.300000 USER=-10.000000 PID=26416 IDLE=0.100000 BUSY=0.800000 USER=-20.000000 PID=26522 IDLE=0.400000 BUSY=0.700000 USER=-30.000000 So for example, if I wanted to remove the line that says PID=26416.... how could I do that, without writing the file over again? I can use external unix commands, however I am not very familiar with them so please if that is your suggestion, give an example. Thanks!

  Read the article

• Manipulating the case of the build macros in Visual Studio: $(TargetDir)

  - by miked

  I've come across a weird problem today in Visual Studio 2005. If I create a new configuration "NewConfiguration" for one of my projects. The output directory referred to by the project in $(TargetDir) is "/path/to/build/area/newconfiguration". Note the loss of capital letters, I'd expect it to be in "/path/to/build/area/NewConfiguration". In another project that I created yesterday, the capital letters are there. Normally this would be a problem, but it's part of a fairly complicated build system where some of it's on unix and we need to worry about case sensitive filename. Does anyone know where the source string for the Visual Studio macros like $(TargetDir) are stored so that I can get the case to be consistent and match what we need for our build system?

  Read the article

• Is there some way Linux editors can tell the programming language without the file extension?

  - by vfclists

  I am editing some scripts on Linux without the languages file extensions, and it seems that the editors, namely vi, nano and gedit are not applying syntax highlighting because the filenames don't use the language extensions. Is there some parameters to be passed or some setting that can enable them to recognize the language? Update: After some googling I realize that bash has that ability, at least to do some parsing or check the shebang at the top determine the language. By default Ubuntu does not install the complete vim package, so after installing it, the shell files are recognized. I don't know about nano or gedit, but vi and its graphical counterpart will do.

  Read the article

• Secure against c99 and similar shells

  - by Amit Sonnenschein

  I'm trying to secure my server as much as i can without limiting my options, so as a first step i've prevented dangerous functions with php disable_functions = "apache_child_terminate, apache_setenv, define_syslog_variables, escapeshellarg, escapeshellcmd, eval, exec, fp, fput, ftp_connect, ftp_exec, ftp_get, ftp_login, ftp_nb_fput, ftp_put, ftp_raw, ftp_rawlist, highlight_file, ini_alter, ini_get_all, ini_restore, inject_code, mysql_pconnect, openlog, passthru, php_uname, phpAds_remoteInfo, phpAds_XmlRpc, phpAds_xmlrpcDecode, phpAds_xmlrpcEncode, popen, posix_getpwuid, posix_kill, posix_mkfifo, posix_setpgid, posix_setsid, posix_setuid, posix_setuid, posix_uname, proc_close, proc_get_status, proc_nice, proc_open, proc_terminate, shell_exec, syslog, system, xmlrpc_entity_decode" but i'm still fighting directory travel, i can't seems to be able to limit it, by using a shell script like c99 i can travel from my /home/dir to anywhere on the disc. how can i limit it once and for all ?

  Read the article

• How to disable Utility Manager (Windows Key + U)

  - by Skizz

  How do I disable the Windows Key+U hotkey in Windows XP? Alternatively, how do I stop the utility manager from being active? The two are related. The utilty manager is currently providing a potential security hole and I need to remove it*. The system I'm developing uses a custom Gina to log in and start a custom shell. This removes most Windows Key hotkeys but the Win+U still pops up the manager app. Update: Things I've tried and don't work: NoWinKeys registry setting - this only affects explorer hotkeys; Renaming utilman.exe - program reappears next login; Third party software - not really an option, these machines are audited by the clients and additional, third party software would be unlikely to be accepted. Also, the proedure needs to be reasonably straightforward - this has to be done by field service engineers to existing machines (machines currently in Russia, Holland, France, Spain, Ireland and USA). * The hole is via the internet options in the help viewer the utility app links to.

  Read the article

• Environment variable for volume names in Windows?

  - by Shinrai

  I'm trying to write a batch file that will look at the volume label of the current drive and report if it's not equivalent to a certain string. Is there a default variable in the shell for this? Can I define one? Am I SOL and I'll actually have to do some (shudder) programming? EDIT: If this is possible in PowerShell that would work fine. (For the curious, we ship our machines with software cloning as a rapid bootable backup solution since most of our customers are daytraders and aren't interested in RAID due to urgency of getting-the-hell-back-to-work-right-away if there's a software corruption problem, and we want to make it immediately obvious if they're booting to the backup drive unintentionally, like say the primary failed entirely. The hope was just to write a simple batch file that would autostart on boot and throw a warning in the event of a problem.)

  Read the article

• ssh into my mac from my ipod

  - by mechko

  Ok so this is what I got, but it don't look right... The set up is as follows: My Mac is on a wireless network, presumably behind a firewall over which I have no admin privileges. Further, said wireless network changes my internal IP randomly at random times, so I do not know what my ip is. Note that localhost is a variable ip I forward some port or the other to my dns: ssh -R 19999:localhost:22 sourceuser@remotehost connect to my remotehost from my machine: ssh sourceuser@remotehost and then connect to the port I forwarded: ssh localhost -p 19999 [solved] Second question, there are a bunch of ssh clients for the iphone. Which one should I use? EDIT: Since jailbreak is not illegal, simply frowned upon by the Corporation, I've decided to go with that. I have a nice shell with openssh and the works. Best of all, it's free and it does exactly what you expect.

  Read the article

• Add directory to $PATH if it's not already there

  - by Doug Harris

  Has anybody written a bash function to add a directory to $PATH only if it's not already there? I typically add to PATH using something like: export PATH=/usr/local/mysql/bin:$PATH If I construct my PATH in .bash_profile, then it's not read unless the session I'm in is a login session -- which isn't always true. If I construct my PATH in .bashrc, then it runs with each subshell. So if I launch a Terminal window and then run screen and then run a shell script, I get: $ echo $PATH /usr/local/mysql/bin:/usr/local/mysql/bin:/usr/local/mysql/bin:.... I'm going to try building a bash function called add_to_path() which only adds the directory if it's not there. But, if anybody has already written (or found) such a thing, I won't spend the time on it.

  Read the article

< Previous Page | 234 235 236 237 238 239 240 241 242 243 244 245  | Next Page >