Search Results

Search found 6890 results on 276 pages for 'gnome shell'.

Page 212/276 | < Previous Page | 208 209 210 211 212 213 214 215 216 217 218 219  | Next Page >

• Can the expect utility handle a case where the process it spawns also spawns a sub process?

  - by davidparks21

  I'm trying to use expect to handle rsync over an ssh shell, but it gets stuck. If I run my rsync command it works (simplified here): It prompts me for my password and copies files to the server: rsync -e ssh -<other_params> If I then enclose that in expect: expect -d -c "spawn rsync -e ssh -<other_params>" -c "expect password:" -c "send mypass\r" It does not execute properly, the program exists and no files are copied. Even the debug mode isn't giving many clues. My best guess is that rsync is spawning the ssh process, and the ssh process is what needs to be interacted with, but send is picking up the rsync process id and sending the input there. Any thoughts?

  Read the article

• Ubuntu hangs on boot when NFS-mounting entries in /etc/fstab, but they mount cleanly otherwise

  - by lorin

  I'm managing several Ubuntu 9.10 servers that NFS mount several folders (including /home). I'd like these folders to be mounted at boot time. I would like to have several entries in my /etc/fstab to accomplish this, e.g. 192.168.1.100:/home /home nfs rw 0 0 192.168.1.100:/usr/ansys_inc /ansys_inc nfs ro 0 0 Unfortunately, with this configuration, the servers usually (although not always) hang during the bootup sequence when trying to do the NFS mount. if I comment out these fstab entries, reboot the machine, uncomment them and mount them manually using the shell, the folders mount cleanly. I'm not sure how to go about debugging this problem. It seems like it has something to do with the boot sequence, that some relevant process hasn't been started by the time the OS tries to mount the folders.

  Read the article

• How can I update Firefox add-ons automatically?

  - by Maelstrom

  Similar to this question, is it possible to update installed plugins via the command line? I'm running YSlow with beacon reporting as a nightly cron job under OSX: /Applications/Firefox.app/Contents/MacOS/firefox-bin -no-remote -P YSlow http://www.example.com/ & PID=$! sleep 300 kill $PID This dumps FF into the background and grabs the PID, waits 300 seconds (for the page to load) then kills it. If there is an update pending, the browser "hangs" waiting for a confirmation. If I do click on the "install updates" link, everything works and then Firefox launches a new process - the $! returned by the shell is no longer valid. Can I update a plugin from the command line without confirmation? Can I curl the XPI into a file and install it without confirmation?

  Read the article

• Starting multiple applications in Ubuntu Unity

  - by Black

  I would like to start multiple GUI applications with a single script or command in Ubuntu 12. By now, I have a shell script that starts an application in the foreground and waits for the termination of the application afterwards starts several applications (like browser, mailer, IRC client) in the background The script is working, however all the applications are getting the same icon and are treated like different windows of one application, i.e. the script. Is there a way to start applications from a script, that makes Unity display the icons of the applications, e.g. the Thunderbird icon, instead of a single default icon for the script? The script looks like this: ! /bin/bash wait for termination... /usr/bin/libreoffice path/to/document in background /usr/bin/thunderbird & /usr/bin/pidgin &

  Read the article

• unable to properly execute binaries from PHP

  - by Lowgain

  I was building an app on a SUSE box, and had a binary called create_group for instance, which had a suid bit and allowed my PHP app to call exec('create group grpname'); and create a new group (there are others for users, etc). The binary was a small c script that calls setuid(0) and then runs the user creation stuff. This worked perfectly on the SUSE box I recently moved my project to Ubuntu and everything works fine except these binaries. I can run them from the shell and they work okay, but when I get the PHP app to run them it just does nothing. Is there anything Ubuntu would be doing differently that I'm missing?

  Read the article

• Configure Git to use Beyond Compare for image diff

  - by Barney

  Because we work with a number of sprites, the kind of specialised diff views provided by Beyond Compare would be ideal to see which one of 2 versions I'm after when conflicts arise. I've already configured Git to use Beyond Compare as my primary diff and merge tool as described in their integration guide — it specifically goes into how to configure TortoiseSVN to use it for images, and I've found these articles talking about .gitattributes in general and how to script interactions from a *nix shell — but it's not obvious to me how I can use the advice provided by these guides to make a simple change that would say "use the default diff & merge bindings for files determined to be images, too". For the record, I'm doing all this on Windows :P

  Read the article

• 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

• How to connect with MySQL server if it won't connect via the socket?

  - by cwd

  I have an account on a shared server. I have jailshell access and also PhpMyAdmin. I want to run mysql commands via SSH but I'm getting an error: $ mysql -u mySqlUser -p mySqlPw Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' I can connect with PHP and phpMyAdmin, so would it be possible to call mysql from the shell and have it connect via an ip and port instead of the socket? The file /var/lib/mysql/mysql.sock does not exist - maybe that is intentional, and the only thing in /etc/my.cnf is [mysqld] skip-innodb More Info I don't have access to change system settings. I did a search in /var for mysql.sock but found nothing. However, phpMyAdmin might be connecting via a socket somehow: Really it would just be great if I could connect via IP. Also tried these two syntaxes: $ mysql -u mySqlUser -p mySqlPw -h localhost $ mysql -u mySqlUser -p mySqlPw -h localhost -P 3306 Both with the same result: ERROR 2002 (HY000): Can't connect to local MySQL server through socket '/var/lib/mysql/mysql.sock' (2)

  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

• 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

• 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

• 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

• 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 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

• 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

• 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

• 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

• how to manage credentials/access to multiple ssh servers

  - by geoaxis

  I would like to make a script which can maintain multiple servers via SSH. I want to control the authentication/authorization in such a manner that authentication is done by gateway and any other access is routed through this ssh server to internal services without any further authentication/authorization requirements. So if a user A can log into server_1 for example. He can then ssh to server_2 without any other authentication and do what ever he is allowed to do on server_2 (like shut down mysql, upgrade it and restart it. This could be done via some remote shell script). The problem that I am trying to solve is to come up with a deployment script for a JavaEE system which involves databases and tomcat instances. They need to be shutdown and re-spawned. The requirement is to have a deployment script which has minimal human interaction as possible for both developers and operation.

  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

• 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

• 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

• 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

• windows 2003 - why can't serial port be accessed remotely?

  - by Danny Staple

  we have recently installed the updates on some of our servers, which have a bit of hardware attached via USB that presents itself as a serial (COM) port. The strange behaviour is that if I start a cmd shell on the server via VNC, I can open the serial port. If I run a service and start it from there (telnet, jenkins) then I receive a "not found" error for it. IE: C:\Documents and Settings\some_user>echo 1 >COM4: C:\Documents and Settings\some_user> on the local cmd will work, and on the remote telnet will give: C:\Documents and Settings\some_user>echo 1 >COM4: The system cannot find the file specified. C:\Documents and Settings\some_user> I cannot see any security settings on the Device manager settings panel for this device.

  Read the article

• "Cannot allocate memory" while no process seems to be using up memory

  - by omat

  I am not competent on server issues, any help is much appreciated. When try to start a python/django shell on a linux box, I am getting OSError: [Errno 12] Cannot allocate memory. free -m seems to confirm I am out of memory: total used free shared buffers cached Mem: 590 560 29 0 3 37 -/+ buffers/cache: 518 71 Swap: 0 0 0 But I cannot see what is eating up the memory with top or ps aux: PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 1 root 20 0 24336 908 0 S 0.0 0.2 0:00.68 init 2 root 20 0 0 0 0 S 0.0 0.0 0:00.00 kthreadd 3 root 20 0 0 0 0 S 0.0 0.0 0:04.85 ksoftirqd/0 How can I identify the leak? Thanks. BTW, I am not sure if it is relevant, but the machine I am talking about is an AWS EC2 instance with Ubuntu 12 running.

  Read the article

< Previous Page | 208 209 210 211 212 213 214 215 216 217 218 219  | Next Page >